JP7539148B2 - Layered organ region specifying device, layered organ region specifying method, and program - Google Patents

Description

The present invention relates to a technology for identifying regions of one or more layers of an organ having a layered structure from a medical image.

In the prior art, the steps are: setting scanning lines radially on the first image from a center point of the first image determined based on a manually input contour point (input contour point); focusing on the image density change along each scanning line, extracting contour candidate points that have the potential to become contour points for each scanning line other than the scanning line on which the input contour point is located; and, taking one of the multiple input contour points as the start point of the determined contour line of the first image and the other as the end point of the determined contour line, evaluating combinations of multiple contour candidate points obtained by connecting the contour candidate points on each adjacent scanning line between the two input contour points using the characteristic that the organ contour is smooth, and repeating the operation of selecting an optimal combination of contour candidate points between all input contour points, and using the selected optimal combination of contour candidate points as the contour of the first image. A method for automatically extracting organ contours is known that includes the steps of: determining the central point of the first image as a point; setting scanning lines radially from a point corresponding to the central point of the first image for each captured image from the second image onwards; extracting multiple contour candidate points for each scanning line by focusing on the image density change on each scanning line; and evaluating combinations of multiple contour candidate points obtained by connecting the contour candidate points of adjacent scanning lines using the characteristics that the contour points of each captured image from the second image onwards are close to the determined contour points of an image that is substantially close in time or position to the captured image, and that organ contours are smooth, thereby selecting optimal combinations of contour candidate points for each captured image from the second image onwards and determining these as contour points for each captured image (see, for example, Patent Document 1).

Japanese Patent No. 2903995 (page 1, Figure 1, etc.)

However, conventional layered organ region identification devices had the problem that they were unable to properly identify the regions of one or more layers of an organ that has a layered structure of three or more layers in a medical image.

The layered organ region identification device of the present invention includes a medical image display unit that displays a medical image of a layered organ having a first layer, a second layer that is in contact with the outside of the first layer, and a third layer that is in contact with the outside of the second layer, a first reception unit that receives a first instruction that specifies a pixel in the first layer of the medical image, a second reception unit that receives a second instruction that specifies the outer contour of a region including the third layer of the medical image, and a first calculation unit that acquires the pixel value of the pixel specified by the first instruction and determines a set of pixels that satisfy a first approximation condition for the pixel value and are adjacent to the pixel specified by the first instruction as the pixels of the first layer. The device for identifying layered organ regions includes a layer determination unit, a second layer determination unit that determines a set of adjacent, contiguous pixels on the outside of the pixels in the first layer as pixels in the second layer, a third layer determination unit that obtains a representative value of pixel values of two or more pixels located outside the pixels in the second layer and within a contour specified by a second instruction, and determines a set of pixels located inside the contour specified by the second instruction, other than the pixels in the first and second layers, that have a pixel value that satisfies a second approximation condition for the representative value, as pixels in the third layer, and an output unit that outputs information indicating the pixels in the third layer.

This configuration makes it possible to properly identify areas of one or more layers of an organ having a layered structure of three or more layers in a medical image.

Furthermore, in the layered organ region identifying device of the present invention, the first approximation condition may be a condition that the pixel value is within the range from X- _α1 to X+ _α1 (where X is the pixel value of the pixel specified by the first instruction, and _α1 is a value such that a set of pixels satisfying the condition and contiguous to the pixel specified by the first instruction does not overlap with a pixel of the contour specified by the second instruction).

This configuration allows the area of the first layer to be properly identified, and the area of one or more layers to be properly identified.

Furthermore, in the layered organ region identifying device of the present invention, the first layer determination unit may be configured to sequentially detect a set of pixels that satisfy the condition that the pixel value is within the range of X-α to X+α (where X is the pixel value of the pixel specified by the first instruction) and that are consecutive to the pixel specified by the first instruction, by decreasing the value of α from a predetermined initial value until at least a part of the set goes from overlapping with the pixel of the contour specified by the second instruction to no longer overlapping, and to determine the set of pixels detected under a first approximation condition, which is a condition that the value of α is _α1 , which is the α value at which the set no longer overlaps, as the pixels of the first layer.

This configuration allows the area of the first layer to be properly identified, and the area of one or more layers to be properly identified.

In addition, in the layered organ region identifying device of the present invention, the second layer determining unit may determine a set of pixels in the medical image within a range from a pixel adjacent to the outside of the pixels in the first layer to a predetermined number of pixels outward as pixels in the second layer.

This configuration allows the area of the second layer to be properly identified, and the area of one or more layers to be properly identified.

In addition, in the layered organ region identifying device of the present invention, the third layer determination unit may determine a representative point of the pixels of the first layer, determine a contour of a figure having a closed area centered on the representative point and smaller in size than the contour specified by the second instruction, and obtain a representative pixel value for two or more pixels on the contour.

This configuration allows the pixel value range of the pixels in the third layer to be appropriately set, and allows the area of one or more layers to be appropriately identified.

The layered organ region identifying device of the present invention may also be configured such that the second approximation condition is that the pixel value is within the range of Y-γ to Y+γ (where Y is a representative value of the pixel, and γ is a value obtained using the pixel values of two or more pixels on the contour to indicate the range of pixel values that are likely to be pixels of the third layer).

This configuration allows the pixel value range of the pixels in the third layer to be appropriately set, and allows the area of one or more layers to be appropriately identified.

In addition, in the layered organ region identifying device of the present invention, the output unit may further output information indicating the pixels of the first and second layers.

This configuration allows the first, second, and third layer regions to be properly identified.

The layered organ region identifying device of the present invention may be configured such that in the layered organ region identifying device, the layered organ is the uterus, the first layer is the endometrium, the second layer is the junction zone (muscle layer directly below the endometrium), and the third layer is the uterine muscle layer.

This configuration makes it possible to properly identify one or more layered regions in medical images of the uterus, which is a layered organ.

The layered organ region identification device according to the present invention can appropriately identify the region of one or more layers of an organ having a layered structure of three or more layers in a medical image.

FIG. 2 is a block diagram of a layered organ region specifying device according to an embodiment of the present invention; A flowchart for explaining the operation of the layered organ region specifying device. FIG. 13 shows an example of a medical image displayed by the layered organ region specifying device. FIG. 13 shows an example of a medical image displayed by the layered organ region specifying device. FIG. 5(a)-FIG. 5(c) show a specific layer management table of the layered organ region specifying device. FIG. 13 is a schematic diagram showing representative points in a medical image displayed by the layered organ region specifying device. FIG. 13 shows an example of a medical image displayed by the layered organ region specifying device. FIG. 1 is a diagram showing an example of the appearance of a computer system according to an embodiment of the present invention. FIG. 1 shows an example of the configuration of the computer system.

Below, an embodiment of a layered organ region identification device and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, and therefore repeated explanations may be omitted.

(Embodiment)
FIG. 1 is a block diagram of a layered organ region specifying device 1 according to the present embodiment.

The layered organ region identification device 1 includes a medical image display unit 101, a medical image storage unit 102, a first reception unit 103, a second reception unit 104, a first layer determination unit 105, a second layer determination unit 106, a third layer determination unit 107, and an output unit 108.

The medical image display unit 101 displays a medical image of a layered organ having a first layer, a second layer, and a third layer. The second layer is a layer that contacts the outside of the first layer, and the third layer is a layer that contacts the outside of the second layer. A layered organ is an organ that has a layered structure. A layered organ is an organ of a living body. The living body here is usually a human being, but may be an animal other than a human being. A layered organ is, for example, a uterus or a heart. For example, when the layered organ is a uterus, the first layer is the endometrium, the second layer is the junction zone, and the third layer is the myometrium. The junction zone is also called the muscle layer directly below the endometrium. When the layered organ is a heart, for example, the first layer is the ventricular cavity, the second layer is the myocardium, and the third layer is the epicardium. In the following, a medical image is mainly described as an image of a uterus.

A medical image of a layered organ is, for example, a medical image of a layered organ of a living body. A medical image is, for example, an image that can usually be used for clinical or medical purposes. However, it goes without saying that a medical image may be used for other purposes, such as research or academic purposes. A medical image of a layered organ is, for example, a tomographic image of a part including a layered organ of a living body. A medical image is, for example, an image taken by CT (computed tomography), MRI (magnetic resonance imaging), PET (positron emission tomography), or SPECT (single photon emission computed tomography). A medical image is, for example, a grayscale or color raster image. The pixel value of a pixel of a medical image is, for example, the luminance value or saturation of the pixel. A pixel value is usually expressed as a value of 256 gradations, but the number of gradations of a pixel value is not limited to this. The pixel values of a medical image may be, for example, measurement values or signal values acquired when capturing a medical image at a site in a living body corresponding to the pixel, or may be values that have been indexed from these. There is no restriction on the resolution, data structure, file format, etc. of the medical image.

The medical image of a layered organ displayed by the medical image display unit 101 may be a medical image of a layered organ having a layered structure composed of three or more layers, for example, a medical image of a layered organ having a layered structure composed of four or more layers. When the medical image is a medical image of a layered organ having a layered structure composed of four or more layers, any layer of the four or more layers may be considered to be the first layer as long as two layers are arranged on top of each other outside the first layer. In such a case, the layer in contact with the outside of the first layer may be considered to be the second layer, and the layer in contact with the outside of the second layer may be considered to be the third layer.

The medical image of the layered organ displayed by the medical image display unit 101 may be a medical image of a layered image stored in the medical image storage unit 102, or may be a medical image of a layered organ stored in a server device (not shown) or the like provided outside the layered organ region identification device 1. In the following, a case where the medical image display unit 101 displays a medical image of a layered image stored in the medical image storage unit 102 will be mainly described as an example.

The medical image display unit 101 may display the medical image on a monitor (not shown) that the medical image display unit 101 has, or may display the medical image on a monitor (not shown) external to the layered organ region identification device 1. The display of the medical image here also includes the concept of transmitting information for displaying the medical image to another device capable of displaying the medical image, and displaying the medical image on the other device.

The medical image display unit 101 may or may not include a display device such as a monitor. The medical image display unit 101 is realized, for example, by driver software for a display device such as a monitor, or by driver software for a display device such as a monitor and a monitor, etc.

The medical image storage unit 102 stores one or more medical images of a layered organ having a first layer, a second layer, and a third layer as described above.

The medical image storage unit 102 accumulates medical images received by a receiving unit (not shown) of the layered organ region identification device 1 from a device (not shown) that captures medical images or from other devices. The storage of the medical images here may be temporary storage of the medical images received by the receiving unit. Note that the manner in which the medical images are accumulated in the medical image storage unit 102 is not limited to the above.

The medical image storage unit 102 is preferably a non-volatile recording medium, but can also be a volatile recording medium. If the medical image display unit 101 does not display the medical images stored in the medical image storage unit 102, the medical image storage unit 102 may be omitted.

The first receiving unit 103 receives a first instruction that specifies one or more pixels in a first layer of the medical image of the layered organ displayed by the medical image display unit 101. The reception of the first instruction may be, for example, the reception of the coordinates of a pixel in the first layer of the medical image of the layered organ displayed by the medical image display unit 101.

The first reception unit 103 receives a first instruction from a user, for example, via an input device (not shown) such as a keyboard, a mouse, a touch panel, or a tablet. For example, when the first reception unit 103 receives an operation to click or tap a pixel in the first layer of the medical image of a layered organ displayed by the medical image display unit 101 via an input device such as a mouse, the first reception unit 103 receives a first instruction that specifies the clicked or tapped pixel.

Here, the concept of accepting a first instruction includes accepting information input from an input device (not shown) such as a keyboard, mouse, or touch panel, receiving information transmitted via a wired or wireless communication line, and accepting information read from a recording medium. The input means used to accept the first instruction may be anything, such as a keyboard, mouse, or menu screen. The same applies to the acceptance of a second instruction, which will be described later. The first acceptance unit 103 may be realized by a device driver for an input means such as a mouse, keyboard, or touch panel, or control software for a menu screen. The same applies to the second acceptance unit 104.

The second reception unit 104 receives a second instruction specifying the outer contour of the area including the third layer of the medical image of the layered organ displayed by the medical image display unit 101. The area including the third layer is, for example, an area including an area where a set of pixels of the third layer is arranged. The area including the third layer is, for example, an area having a contour that surrounds the third layer from the outside along the outer edge of the third layer. The area including the third layer is an area that may have a portion that protrudes outward from the outer edge of the third layer, a portion that is recessed inward from the outer edge of the third layer, etc. The outside of the third layer is, for example, a portion of the third layer opposite the second layer. The area including the third layer may be, for example, an area including an area where a set of pixels of the third layer is arranged, and may include pixels located outside the outer edge of the third layer. In addition, the area including the third layer may be, for example, an area including an area where a set of pixels of the third layer is arranged, and may not include a portion of the pixels of the third layer that are located inside the outer edge of the third layer. It is preferable that the portion of the region including the third layer that protrudes outward from the outer edge of the third layer or that is recessed inward is close to the outer edge. It is preferable that the region including the third layer is a region that includes at least the entire region of the third layer.

The second instruction specifying the contour that surrounds the area including the third layer from the outside is, for example, an instruction to specify multiple consecutive pixels that surround the area including the third layer from the outside. The consecutive pixels here may be considered to be adjacent pixels above, below, left, right, and so on, and may also be considered to include adjacent pixels in a diagonal direction in addition to adjacent pixels above, below, left, and right.

The second reception unit 104 receives the second instruction, for example, by manual input from the user. The second reception unit 104 receives the first instruction from the user, for example, via an input device (not shown) similar to the first reception unit 103. For example, the second reception unit 104 receives information specifying a set of pixels located at the coordinates of the traced trajectory by the user tracing the outside of the third layer area with a cursor or the like using a mouse or a touch panel on the medical image of the layered organ displayed by the medical image display unit 101 as a second instruction specifying a contour. Note that the second instruction specifying the outside contour of the area including the third layer may be considered, for example, as an instruction to specify the outside contour of the third layer by the user manually inputting it, or an instruction to specify the contour of the third layer by tracing the outside of the third layer, and usually the contour specified in this way is an approximate contour of the outside of the third layer.

The first layer determination unit 105 obtains the pixel value of the pixel specified by the first instruction, and determines a set of pixels that satisfy a first approximation condition for the pixel value and are continuous with the pixel specified by the first instruction as the pixels of the first layer. Determining the pixels of the first layer means associating a label or the like indicating that the pixel is a pixel of the first layer with the pixels of the medical image determined as the pixels of the first layer. Associating a label with a pixel may be associating a label with a pixel identifier or pixel coordinates, etc., and storing the label in a storage unit (not shown), or updating a medical image stored in the medical image storage unit 102, etc., with a medical image to which the label is associated. Also, determining the pixels of the first layer may be obtaining information indicating the contour of an area where the pixels of the first layer are continuous and storing the information in a storage unit (not shown). The information indicating the contour is, for example, information of multiple coordinates indicating the contour, or information of multiple pixels on the contour (for example, pixel identifiers, pixel coordinates, etc.). This is also true for the process of determining pixels of the second layer and the process of determining pixels of the third layer, which will be described later, except that the pixels are different.

The first approximation condition is, for example, a condition for detecting, in a medical image, a pixel that is approximate to the pixel value of a pixel specified by a first instruction as a pixel of the first layer. The first approximation condition is, for example, a condition indicating the range of pixel values of pixels to be detected as pixels of the first layer. The first approximation condition is, for example, a condition for detecting, as the pixel value of the first layer, a pixel having a pixel value within a desired range including the pixel value of the pixel specified by the first instruction. The first approximation condition is, for example, a condition indicating a range of pixel values including the pixel value of the pixel specified by the first instruction, and a pixel having a pixel value within this range is determined to be a pixel that satisfies the condition. The range of pixel values is, for example, expressed by an upper limit and a lower limit of pixel values.

The first approximation condition is set, for example, based on the pixel value of the pixel specified by the first instruction. The first approximation condition is, for example, a condition that, when the pixel value of the pixel specified by the first instruction is X, the pixel value of the pixel is within a range from X- _α1 to X+ _α1 . _α1 here is usually a positive integer, but may be a positive real number. _α1 may be a value specified in advance, or may be a value dynamically determined according to the contour specified by the second instruction, as described later. In addition, _α1 here may be, for example, a fixed value, or may be a value specified according to the value of X. For example, the value of _α1 may be a value specified as a ratio to the value of X.

In the following, in this embodiment, the value of _α1 is expressed as a ratio (specifically, %) to the value of X, and when actually judging whether the pixel value of a pixel satisfies the first approximation condition, the value of _α1 % of X is used instead of _α1 . For example, when the first approximation condition is that the pixel value of a pixel is within the range from X-25% to X+25%, pixels having pixel values within the range from (1-0.25)X to (1+0.25)X are actually detected as pixels that satisfy the first approximation condition.

When the first approximation condition is a condition that pixels whose pixel values are within the range from X- _α1 to X+ _α1 are detected, it is preferable to set _α1 to a value such that a set of pixels that are adjacent to the pixel specified by the first instruction and that satisfy the first approximation condition do not overlap with a pixel of the contour specified by the second instruction.

For example, the first layer determination unit 105 may increase or decrease the value of α from a predetermined value (for example, initial value α ₀ : α ₀ is a positive value) in the pixel value range from X-α to X+α (where α is a variable) to determine the value of α as α1 such that a set of pixels that are adjacent to the pixel specified by the first instruction and have pixel values in the range from X-α to X+α do not overlap with the pixels of the contour specified by the second instruction, and may then determine a _first approximation condition by substituting this _α1 as the above value of α.

For example, the first layer determination unit 105 sequentially detects a set of pixels that are adjacent to the pixel specified by the first instruction and that satisfy the condition that the pixel value is within the range from X-α to X+α by decreasing the value of α from a predetermined initial value. Then, the first layer determination unit 105 determines the set of pixels detected when the detected set of pixels goes from overlapping with the pixel of the contour specified by the second instruction to no longer overlapping as the pixels of the first layer. The condition having the above pixel value range, in which _α1 , the value of α at this time, is substituted for the value of α, becomes the first approximation condition.

Specifically, for example, the first layer determination unit 105 detects a set of pixels that have pixel values within the range of X-α to X+α with a predetermined initial value _α0 substituted as the value of α and that are continuous with the pixel specified by the first instruction. Then, it determines whether the detected set of pixels overlaps with the pixels of the contour specified by the second instruction. If they do not overlap, the condition having the above range of pixel values with the initial value _α0 substituted as _α1 for α is set as the first approximation condition. Then, the set of pixels detected at this time that are continuous with the pixel specified by the first instruction may be determined as the pixels of the first layer.

Furthermore, if there is an overlap, the first layer determination unit 105 may, for example, sequentially decrease a predetermined value (for example, the value "1") from the initial value α ₀ and assign it as the value of α in the above condition, and each time it assigns a decreased value, it detects a group of pixels having pixel values within the range from X-α to X+α that are continuous with the pixel specified by the first instruction. Each time it detects a group of pixels, it determines whether the detected group of pixels overlaps with the pixels of the contour specified by the second instruction. It then determines the detected group of pixels continuous with the pixel specified by the first instruction as the pixels of the first layer. The first layer determination unit 105 may also determine the first approximation condition to be a condition having the above range of pixel values, in which the value of α at the point where there is no overlap is assigned as α ₁ .

If no overlap occurs after substituting the initial value α ₀ , the first layer determination unit 105 may, for example, substitute a value obtained by successively increasing the initial value α ₀ by a predetermined value (for example, the value "1") as the value of α, and each time a value is substituted, detect a set of pixels having pixel values within the range from X-α to X+α that are continuous with the pixel specified by the first instruction. Then, each time a set of pixels is detected, it may be determined whether the detected set of pixels overlaps with the contour pixels specified by the second instruction, and the set of pixels detected immediately before the overlap is determined to be the pixels of the first layer. Alternatively, the first approximation condition may be determined to be a condition having the above range of pixel values, in which α ₁ , which is the value of α immediately before the value of α that caused the overlap, is substituted as the value of α.

In the above, the case where the first instruction is an instruction to specify one pixel has been described, but the first receiving unit 103 may receive an instruction to specify multiple adjacent pixels as the first instruction. In this case, the first layer determination unit 105 may use one of the multiple consecutive pixels specified by the first instruction (e.g., the pixel closest to the center or center of gravity of the multiple pixels) as the one pixel specified by the first instruction by the first layer determination unit 105 described later. In addition, the first layer determination unit 105 may use a representative value (e.g., average value or median value) of the pixel values of the multiple consecutive pixels specified as the pixel value of the pixel specified by the first instruction, and may determine a set of pixels that satisfy a first approximation condition and are consecutive to one of the multiple consecutive pixels specified by the first instruction (e.g., the pixel closest to the center or center of gravity of the multiple pixels) as the pixels of the first layer.

In addition, in the above, as an example of the first approximation condition, a condition that a pixel has a pixel value within the range from X- _α1 to X+ _α1 has been described, but the lower limit value of the pixel value range of this condition may be X- _α2 ( _α2 is a positive value other than _α1 ).

The second layer determination unit 106 determines a set of consecutive pixels adjacent to the outside of the pixels of the first layer as the pixels of the second layer. For example, the second layer determination unit 106 determines a set of pixels within a range from the pixels adjacent to the outside of the pixels of the first layer of the medical image to a predetermined number of pixels toward the outside as the pixels of the second layer. The pixels adjacent to the outside of the pixels of the first layer may be considered as pixels adjacent to the pixels of the outer edge of the first layer. The predetermined number of pixels is, for example, a number of pixels determined according to the thickness of a general second layer. For example, the predetermined number of pixels is a number of pixels between 1 pixel and 5 pixels. However, this number of pixels may be changed depending on the type of layered organ, which layer is the second layer to be detected, etc. The predetermined number of pixels may be determined according to the size of the first layer or the distance from the outside of the first layer to the contour specified by the second instruction. For example, the number of pixels may be obtained by multiplying the number of pixels of the longest part of the first layer by a predetermined coefficient having a predetermined positive value less than 1, or by adding a predetermined number of pixels. Alternatively, the number of pixels may be a certain ratio of the distance from the outside of the first layer to the contour specified by the second instruction, or may be a number of pixels obtained by adding a predetermined number of pixels. The second layer determination unit 106 may also determine the above-mentioned predetermined number of pixels according to the resolution of the medical image, for example, so that the higher the resolution, the greater the number of pixels.

Note that the method of determining the pixels of the second layer here is just an example, and the method of determining the second layer is not limited to the above. For example, the second layer determination unit 106 may determine, as the pixels of the second layer, a set of pixels that are contiguous to a plurality of pixels adjacent to the outside of the pixels of the first layer, and have pixel values within a range from a pixel value obtained by adding a predetermined first value (where the first value is a positive value) to a pixel value obtained by subtracting a predetermined second value (where the second value is a positive value) using the pixel value of the adjacent pixel on the outside as a reference.

The third layer determination unit 107 obtains a representative value of pixel values of two or more pixels located outside the pixels of the second layer and within the contour specified by the second instruction. The outside of the pixels of the second layer here is, for example, the side opposite to the first layer side. The pixels outside the pixels of the second layer here are, for example, pixels located outside the pixels of the second layer and are pixels other than the pixels of the second layer. The pixels outside the pixels of the second layer may be considered to be pixels that do not overlap with either the pixels of the first layer or the pixels of the second layer. The pixels located within the contour specified by the second instruction are pixels located inside the pixels of the contour specified by the second instruction. The inside here is, for example, the second layer side. The pixels within the contour here may be considered to include or not include the pixels of the contour. The representative value is, for example, a representative value obtained by statistically processing the pixel values of two or more pixels. The representative value is, for example, an average value or a median value.

The third layer determination unit 107, for example, determines a representative point of the pixels of the first layer determined by the first layer determination unit 105, determines the contour of a figure having a closed area centered on the representative point and smaller in size than the contour specified by the second instruction, and obtains representative pixel values for two or more pixels on the contour. Determining the contour of a figure may be thought of as determining the pixels on the contour of the figure, or as determining multiple coordinates that indicate the contour of the figure.

For example, the third layer determination unit 107 determines a point that is the center of gravity of the area in which multiple pixels of the first layer are arranged as a representative point of the pixels of the first layer determined by the first layer determination unit 105. Note that, although the point that is the center of gravity is determined as the representative point here, the representative point may be a point other than the center of gravity. For example, the representative point may be the center point, a position shifted by several pixels from the center of gravity, or a point where the major axis and minor axis of the area in which multiple pixels of the first layer are arranged intersect.

The third layer determination unit 107 determines the contour of a figure having a closed area, the size of which is smaller than the contour specified by the second instruction, with the representative point determined above as the center. A figure having a closed area is, for example, a closed figure. The figure having a closed area may be a figure having a shape obtained by reducing the contour specified by the second instruction, or may be a figure having a shape obtained by enlarging the shape of the outer edge of the area in which the pixels of the first layer determined by the first layer determination unit 105 are located, or the shape of the outer edge of the area in which the pixels of the second layer determined by the second layer determination unit 106 are located. In addition, the figure having a closed area may be, for example, a circle.

A figure that is smaller than the contour specified by the second instruction is a figure that does not overlap with the pixels of the contour specified by the second instruction (in other words, has no overlapping parts). For example, the third layer determination unit 107 obtains the distance between the representative point determined above and each pixel of the contour specified by the second instruction, and uses the obtained distances to determine the contour of the figure that does not overlap with the pixels of the contour specified by the second instruction.

In addition, if the figure having a closed area is a figure having a shape obtained by reducing the contour specified by the second instruction, the third layer determination unit 107 may, for example, determine a figure whose overall size is smaller than the contour specified by the second instruction, for example, a contour obtained by reducing the contour specified by the second instruction to a size that is a predetermined factor that is smaller than 100% of the original size.

In addition, if the figure having a closed region is a figure having a shape obtained by enlarging the shape of the outer edge of the region in which the pixels of the first layer determined by the first layer determination unit 105 are located, centered on the representative point determined above, the third layer determination unit 107 may, for example, detect the shortest part of the distance between the pixels of the outer edge of the region in which the pixels of the first layer determined by the first layer determination unit 105 are located and each pixel of the contour specified by the second instruction in the direction connecting the representative point determined above and each pixel of the contour specified by the second instruction, and set the contour of the figure of a size such that the distance between the pixel of the edge corresponding to this short part in the enlarged figure and the center is smaller than the distance between the representative point determined above and the pixel of the contour specified by the second instruction in this short part.

For example, when the figure having a closed region is a perfect circle, the third layer determination unit 107 may determine a perfect circle contour whose radius is shorter than the shortest distance between the representative point determined above and each pixel of the contour specified by the second instruction. For example, the third layer determination unit 107 may determine a perfect circle contour whose center is the representative point determined above and whose radius is β% (β is a positive value less than 100) of the shortest distance between the representative point determined above and each pixel of the contour specified by the second instruction. β is, for example, a predetermined value. Note that the value of β is preferably 80% or less. However, the radius of the perfect circle may be determined in any way as long as it is shorter than the shortest distance between the contour specified by the second instruction and the representative point. For example, a radius shorter than the shortest distance may be determined arbitrarily. For example, β is a predetermined value.

In addition, when the figure having a closed region is a figure having a shape obtained by enlarging the shape of the outer edge of the region in which the pixels of the first layer determined by the first layer determination unit 105 are located, centered on the representative point determined above, the third layer determination unit 107 may, for example, detect the shortest part of the distance between the pixels of the outer edge of the region in which the pixels of the first layer determined by the first layer determination unit 105 are located and each pixel of the edge specified by the second instruction in the direction connecting the representative point determined above and each pixel of the edge specified by the second instruction, and set the edge of the figure of a size such that the distance between the pixel of the edge corresponding to this short part and the center of the enlarged figure is smaller than the distance between the representative point determined above and the pixel of the edge specified by the second instruction in this short part. The same applies when determining the edge of a figure having the shape of the outer edge of the region in which the image of the second layer determined by the second layer determination unit 106 is located.

It is preferable that the third layer determination unit 107 determines the contour of a figure having a closed region, the size of which is smaller than the contour specified by the second instruction and larger than the outer periphery of the second layer. For example, when the figure having a closed region is a figure having a shape obtained by reducing the contour specified by the second instruction with the representative point determined above as the center, the third layer determination unit 107 may, for example, detect the shortest part of the distance between the pixels of the outer edge of the region in which the pixels of the second layer determined by the second layer determination unit 105 are located and each pixel of the contour specified by the second instruction in the direction connecting the representative point determined above and each pixel of the contour specified by the second instruction, and set the contour of the figure of a size such that the distance between the pixel of the edge corresponding to this short part and the center of the reduced figure is smaller than the distance between the representative point determined above in this short part and the pixel of the contour specified by the second instruction, and is larger than the distance between the representative point determined above in this short part and the pixel of the outer edge of the region in which the pixels of the second layer determined by the second layer determination unit are located. For example, if the distance between the representative point in the shortest part detected above and the pixel of the contour specified by the second instruction is _A1 , and the distance between the representative point in the shortest part detected above and the pixel on the outer edge of the area in which the second layer pixels determined by the second layer determination unit 105 are located is _A2 , the third layer determination unit 107 may set a contour that is reduced from the contour specified by the second instruction to _x1 % of the original size (where 100>x> _A2 / _A1 ×100) and centered on the representative point determined above.

Furthermore, in the case where the figure having a closed region is a figure having a shape obtained by enlarging the shape of the outer edge of the region in which the pixels of the second layer determined by the second layer determination unit 106 are located by a magnification greater than 100%, centered on the representative point determined above, the third layer determination unit 107 may, for example, detect the shortest part of the distance between the pixels of the outer edge of the region in which the pixels of the second layer determined by the second layer determination unit 106 are located and each pixel of the contour specified by the second instruction, in the direction connecting the representative point determined above and each pixel of the contour specified by the second instruction, and set the contour of the figure of a size such that the distance between the pixel of the edge corresponding to this short part in the enlarged figure and the center is smaller than the distance between the representative point determined above and the pixel of the contour specified by the second instruction in this short part. For example, if the distance between the representative point in the shortest part detected above and the pixel of the contour specified by the second instruction is _A1 , and the distance between the representative point in the shortest part detected above and the pixel of the outer edge of the area in which the pixels of the second layer determined by the second layer determination unit 105 are located is _A2 , the third layer determination unit 107 may set a contour that is reduced to _x2 % (where 100< _x2 < _A1 / _A2 ×100) of the original size, with the representative point determined above as the center, as the shape of the outer edge of the area in which the pixels of the second layer determined by the second layer determination unit 106 are located.

For example, if the figure having a closed region is a perfect circle, the third layer determination unit 107 may obtain the distance between each pixel on the outer edge of the pixels of the second layer determined by the second layer determination unit 106 and the representative point determined above, and determine the contour of the perfect circle having a radius longer than the shortest of the obtained distances and shorter than the shortest distance between the representative point determined above and each pixel of the contour specified by the second instruction. There is no restriction on how the third layer determination unit 107 determines such a radius. For example, the third layer determination unit 107 may determine the radius of the perfect circle to be any distance within a range longer than the shortest distance between each pixel on the outer edge of the pixels of the second layer and the representative point and shorter than the shortest distance between the representative point and each pixel of the contour specified by the second instruction, or the average distance of these two distances.

How the third layer determination unit 107 determines the contour of a figure having a closed area that is smaller than the contour specified by the second instruction and larger than the outer periphery of the second layer is not limited to the above process. For example, the contour specified by the second instruction may be successively reduced to determine a contour of a size that does not overlap with the area in which the pixels of the second layer determined by the second layer determination unit 106 are located, or the shape of the outer edge of the area in which the pixels of the second layer determined by the second layer determination unit 106 are located may be successively enlarged to determine a contour of a size smaller than the contour specified by the second instruction that has the same shape as the area in which the pixels of the second layer determined by the second layer determination unit 106 are located.

Here, when the layered organ is the uterus, the first layer is the endometrium, and the third layer is the myometrium, the inventors of the present application have found, as a result of their intensive research, that by determining a contour that is reduced from the contour specified by the second instruction to 30% to 80%, preferably 65% to 75%, and more preferably 68% to 72%, centered on the representative point determined above, it is highly likely that a contour that is smaller than the contour specified by the second instruction and larger than the outer periphery of the second layer and has a shape similar to the contour specified by the second instruction will be set. For this reason, when determining a contour that has the same shape as the contour specified by the second instruction (in other words, a contour that has a similar shape) as a figure having a closed area, the contour specified by the second instruction may be reduced from the representative point determined above to 30% to 80%, preferably 65% to 75%, and more preferably 68% to 72%.

The third layer determination unit 107, for example, obtains a representative value of pixel values for one or more pixels on the contour of the figure determined as described above. The representative value obtained by the third layer determination unit 107 is, for example, a value that represents multiple pixels, obtained by statistically processing the pixel values of multiple pixels. The representative value is, for example, the average value or median value of the pixel values of multiple pixels. In the following, an example will be described in which the representative value is obtained and the average value is obtained.

For example, the third layer determination unit 107 preferably obtains a representative value of pixel values for pixels on the contour of the figure determined above, excluding one or more pixels with high pixel values and one or more pixels with low pixel values. The plurality of pixels on the contour here is usually all pixels on the contour, but may be a plurality of pixels excluding a portion. The third layer determination unit 107 preferably obtains a representative value for a plurality of pixels on the contour whose ranks, set in ascending order of pixel values, are between 30% and 70% of the total pixels. As described above, pixels with high and low pixel values may be pixels derived from artifacts (artificial objects) or vascular structures included in the third layer, so that the influence of artifacts (artificial objects) or vascular structures included in the third layer on the detection of the third layer is excluded by obtaining a representative value excluding one or more pixels with high pixel values and one or more pixels with low pixel values as described above. Here, the representative values are obtained for a plurality of pixels whose ranks in ascending order of pixel values are between 30% and 70% of the total pixels, but the ranks of the pixels from which the representative values are obtained are not limited to between 30% and 70% of the total pixels, and for example, the representative values may be obtained for a plurality of pixels whose ranks are between 25% and 80% of the total pixels. According to findings obtained as a result of intensive research by the inventors of the present application, when the layered organ is the uterus and the third layer is the uterine muscle layer, it is preferable to obtain the representative values for the pixels whose ranks are between 30% and 70% as described above in order not to detect pixels other than those in the third layer.

The third layer determination unit 107 may obtain a representative value of pixel values for not only one or more pixels on the contour of the figure determined as described above, but also one or more pixels adjacent to these pixels that are not on the contour. Also, for multiple pixels on the contour of the figure, a representative value of pixel values may be obtained without excluding one or more pixels with high pixel values and one or more pixels with low pixel values as described above.

Note that the process in which the third layer determination unit 107 detects two or more pixels that are located outside the pixels of the second layer and within the contour specified by the second instruction, and the process in which the third layer determination unit 107 obtains a representative value for these pixels, etc., are not limited to the above process.

The third layer determination unit 107 determines, as pixels of the third layer, a set of pixels that have pixel values that satisfy the second approximation condition for the representative values obtained as described above, and that are located inside the contour specified by the second instruction, other than the pixels of the first layer determined by the first layer determination unit 105 and the pixels of the second layer determined by the second layer determination unit 106. Pixels other than the pixels of the first layer determined by the first layer determination unit 105 and the pixels of the second layer determined by the second layer determination unit 106 may be considered to be pixels that are located outside the pixels of the second layer determined by the second layer determination unit 106.

The second approximation condition is, for example, a condition for detecting a pixel having a pixel value that is close to the representative value obtained as described above as a pixel of the third layer. The second approximation condition is, for example, a condition indicating the range of pixel values of pixels to be detected as pixels of the third layer. The second approximation condition is, for example, a condition for detecting a pixel having a pixel value within a desired range including the representative value obtained as described above as a pixel value of the third layer. The second approximation condition is, for example, a condition indicating the range of pixel values including the representative value obtained as described above, and a pixel having a pixel value within this range is determined to be a pixel that satisfies the condition. The range of pixel values is, for example, represented by an upper limit value and a lower limit value of the pixel value. The second approximation condition is, for example, set based on the representative value obtained as described above.

The second approximation condition is, for example, a condition that when the representative value obtained as described above is Y, the pixel value of the pixel is within the range of Y-γ to Y+γ. Here, γ is a positive value obtained using the pixel values of two or more pixels on the contour to indicate the range of pixel values that are likely to be pixels of the third layer. In this second approximation condition, it is preferable that the representative value Y is the average value of the pixel values of one or more pixels of the contour information of the figure determined by the third layer determination unit 107, and γ is twice the standard deviation (σ) (2σ) obtained using the pixel values of one or more pixels of the contour information of the figure determined by the third layer determination unit 107 used to obtain this representative value. For example, when the third layer determination unit 107 obtains an average value to be used as the representative value Y for a plurality of pixels on the contour whose ranks, set in ascending order of pixel values, fall between 30% and 70% of all pixels, the third layer determination unit 107 obtains a standard deviation to be used as γ using the pixel values of the plurality of pixels used to obtain this average value, and obtains a second approximation condition using the obtained average value and twice the value of the standard deviation (2σ). By using such a second approximation condition using the average value and twice the value of the standard deviation (2σ), it becomes possible to detect pixels in the third layer with high accuracy.

The value of γ is not limited to the standard deviation. For example, the value of γ may be a value obtained by subtracting or adding a predetermined value from twice the standard deviation (2σ), or the standard deviation (σ) may be used.

In addition, the representative value used as the Y value is not limited to the average value, but may be a value obtained by adding or subtracting a predetermined value to or from the average value.

The third layer determination unit 107, for example, determines a set of pixels that have pixel values that satisfy the second approximation condition for the representative value acquired as described above, that are located inside the contour specified by the second instruction, other than the pixels of the first layer determined by the first layer determination unit 105 and the pixels of the second layer determined by the second layer determination unit 106, and that are continuous with one or more pixels having pixel values that satisfy the second approximation condition among the pixels on the contour of the figure determined by the third layer determination unit 107 as described above, as the pixels of the third layer. For example, the third layer determination unit 107 detects a set of pixels that are continuous with a plurality of pixels having pixel values that satisfy the second approximation condition among the pixels on the contour of the figure and have pixel values that satisfy the second approximation condition, and determines the set of pixels remaining after excluding the pixels of the second layer determined by the second layer determination unit 106 and the pixels located outside the contour specified by the second instruction from the detected set of pixels as the pixels of the third layer. Furthermore, the third layer determination unit 107 may, for example, in an area outside the pixels of the second layer determination unit 106 and inside the contour specified by the second instruction, detect a set of pixels that are contiguous to a plurality of pixels having pixel values that satisfy the second approximation condition among the pixels on the contour of the figure determined by the third layer determination unit 107 as starting points and have pixel values that satisfy the second approximation condition, detect the pixel set, and determine the detected pixel set as the pixel of the third layer. Note that the pixels on the contour specified by the second instruction may be considered to be pixels inside the contour specified by the second instruction, or may be considered to be pixels outside the contour specified by the second instruction.

The output unit 108 outputs information indicating the pixels of the third layer determined by the third layer determination unit 107. For example, the output unit 108 outputs an identifier or a set of coordinates of the pixels of the third layer. The output unit 108 also outputs, for example, information indicating the contour on which the pixels of the third layer are located (for example, information indicating the pixels of the contour, a set of coordinates of the contour, a closed figure constituted by vector data that draws the contour, etc.). The coordinates here are, for example, the coordinates in the medical image in which the third pixel was determined.

The output unit 108 outputs, for example, information indicating the pixel of the third layer determined by the third layer determination unit 107 in association with the medical image in which the third pixel was detected. For example, the output unit 108 outputs a medical image having information indicating the third pixel acquired for the medical image (for example, medical information having information such as a channel or layer having information indicating the third pixel). The output unit 108 may also output information having information indicating the third pixel in association with the medical image in which the third pixel was detected or an identifier thereof.

The output unit 108 may output, for example, the pixels of the third layer determined by the third layer determination unit 107 in a visually distinguishable manner as the output of information indicating the pixels of the third layer. Outputting in a visually distinguishable manner means, for example, outputting information such that, when an image including pixels of the third layer is displayed, the pixels of the third layer are visually distinguishable from pixels other than the third layer. Displaying in a visually distinguishable manner may, for example, be superimposing a specific color, a specific design, a pattern, or the like on the pixels of the third layer, displaying the pixels of the third layer in a specific color, or displaying the area in which the pixels of the third layer are arranged by surrounding it with a line of a specific color, a dotted line, or the like.

The output unit 108 outputs, for example, a medical image in which the pixels of the third layer determined by the third layer determination unit 107 are visually distinguishable. In addition, information used to visually distinguish such pixels of the third layer may be output in association with the medical image determined by the third layer determination unit 107.

For example, as described above, the output unit 108 may output information such as the identifier of the third pixel and information such as the contour in association with the medical image to a device other than the layered organ region identification device 1, and the device receiving this information may use this information to output (e.g., display) a medical image in which the third pixel is identifiable.

Note that the process of outputting specific pixels or the area in which specific pixels are located in a visually identifiable manner using an image, information indicating the pixels in the image, and information indicating the contours is a well-known technique, so a detailed explanation will be omitted here.

The output unit 108 may output at least one of information indicating the pixels of the first layer determined by the first layer determination unit 105 and information indicating the pixels of the second layer determined by the second layer determination unit 106, similar to the information indicating the pixels of the third layer determined by the third layer determination unit 107. The output of the information indicating the pixels of the first layer is similar to the output of the information indicating the pixels of the third layer, except that the pixels are replaced by the pixels of the first layer, and therefore a detailed description is omitted. For example, the output unit 108 may output the pixels of the first layer determined by the first layer determination unit 105 in a visually distinguishable manner, similar to the pixels of the third layer, as the output of the information indicating the pixels of the first layer. Furthermore, the output of the information indicating the pixels of the second layer is similar to the output of the information indicating the pixels of the third layer, except that the pixels are replaced by the pixels of the second layer, and therefore a detailed description is omitted. For example, the output unit 108 may output the pixels of the second layer determined by the second layer determination unit 106 in a visually distinguishable manner, similar to the pixels of the third layer, as the output of the information indicating the pixels of the second layer.

Note that output here includes, for example, the display of images or information on a monitor, the projection of images or information using a projector, printing with a printer, storing on a recording medium (not shown), sending to an external device capable of displaying or storing images or information, and passing on the processing results to another processing device or another program.

The output unit 108 may or may not include an output device such as a monitor. The output unit 108 may be realized by driver software for an output device, or by a combination of driver software for an output device and an output device, etc.

Next, an example of the operation of the layered organ region identification device 1 will be described using the flowchart in Figure 2.

(Step S101) The medical image display unit 101 determines whether to display a medical image stored in the medical image storage unit 102. For example, when a display instruction receiving unit (not shown) receives an instruction to display one of the medical images stored in the medical image storage unit 102, the medical image display unit 101 decides to display the medical image specified by this instruction. If the medical image is to be displayed, proceed to step S102; if not, return to step S101. Note that, for example, when a medical image is accumulated in the medical image storage unit 102, the medical image display unit 101 may decide to display this medical image.

(Step S102) The medical image display unit 101 displays the medical images stored in the medical image storage unit 102 on a monitor (not shown) or the like. For example, it displays the medical images determined to be displayed in step S101.

(Step S103) The first reception unit 103 determines whether a first instruction has been received for the medical image displayed in step S102. If so, proceed to step S104; if not, return to step S103. Information indicating the specified pixel is stored, for example, in a storage unit (not shown). Note that the medical image display unit 101 may display a dot indicating the pixel indicated by the first instruction superimposed on the medical image.

(Step S104) The second reception unit 104 determines whether a second instruction has been received for the medical image displayed in step S102. If so, proceed to step S105; if not, return to step S104. Information indicating the specified contour (e.g., information about pixels on the contour, etc.) is accumulated in a storage unit (not shown) or the like. Note that the medical image display unit 101 may display a line or the like indicating the contour specified by the second instruction superimposed on the medical image.

(Step S105) The first layer determination unit 105 reads out a condition indicating the range of pixel values from X-α to X+α for acquiring a first approximation condition, and substitutes the initial value α ₀ as the value of α.

(Step S106) The first layer determination unit 105 detects pixels adjacent to the pixel indicated by the first instruction received in step S103 and having pixel values within the range from X-α to X+α.

(Step S107) The first layer determination unit 105 determines whether a portion of the pixels detected in step S106 overlaps with the contour specified by the second instruction received in step S104. If they overlap, the process proceeds to step S108. If they do not overlap, the process proceeds to step S109.

(Step S108) The first layer determination unit 105 changes the value of α, which is a condition indicating the range of pixel values, to a value that is reduced by a predetermined value (e.g., "1"). Then, the process returns to step S10106.

(Step S109) The first layer determination unit 105 determines the set of pixels detected in the previous step S106 as the pixels of the first layer. For example, the first layer determination unit 105 associates information indicating that the determined pixels of the first layer are pixels of the first layer with each other, and stores the information in the medical image storage unit 102 or a storage unit (not shown). Note that the range of pixel values from X-α to X+α, which has the value of α used when detecting this pixel, becomes the first approximation condition, and the value of α at this time corresponds to the above-mentioned _α1 .

(Step S110) The second layer determination unit 106 determines a set of contiguous pixels adjacent to the outside of the first layer pixels determined in step S109 as pixels of the second layer. For example, the second layer determination unit 106 determines a set of pixels within a range from the pixel adjacent to the outside of the first layer pixels determined in step S109 to a predetermined number of pixels outward as pixels of the second layer. For example, the second layer determination unit 106 associates information indicating that the determined second layer pixels are pixels of the second layer with each other, and accumulates the information in the medical image storage unit 102 or a storage unit (not shown), etc.

(Step S111) The third layer determination unit 107 determines a representative point of the pixels of the first layer determined in step S109. For example, the third layer determination unit 107 determines a pixel located at the center of gravity of the area composed of the pixels of the first layer determined in step S109 as the representative point of the pixels of the first layer.

(Step S112) The third layer determination unit 107 determines the contour of a figure having a closed space of a predetermined size with the representative point detected in step S111 as the center. This figure is, for example, the contour of a figure obtained by reducing the contour specified by the second instruction received in step S104. For example, this figure is the contour of a figure obtained by reducing the contour specified by the second instruction received in step S104 to 70% with the representative point determined in step S111 as the center point of the enlargement/reduction. However, 70% here is an example, and is not limited to this range of values. In addition, the third layer determination unit 107 may determine a contour obtained by reducing the contour specified by the second instruction with the representative point as the center, using the distance between the representative point determined in step S111 and the pixel of the contour specified by the second instruction, and the distance between the pixel of the contour outside the pixels of the second layer determined in step S110, and the contour that does not overlap with the outer contour of the pixels of the second layer determined in step S110.

(Step S113) The third layer determination unit 107 acquires the pixel values of the pixels on the contour determined in step S112, and further acquires multiple pixels whose ranking, when sorted from highest pixel value, falls between 30% and 70% of the total. Then, the average value Y of the pixel values of the acquired multiple pixels is acquired as a representative value of two or more pixels on the contour.

(Step S114) The third layer determination unit 107 further obtains the standard deviation σ for the multiple pixels used to obtain the representative value in step 113, and obtains a second approximation condition that the pixel value of the pixel is within the range from Y-γ to Y+γ (where γ is 2σ).

(Step S115) The third layer determination unit 107 sequentially detects consecutive pixels having pixel values that satisfy the second approximation condition obtained in step S114, starting from the pixels on the contour determined in step S112 that are within the range whose pixel values satisfy the second approximation condition obtained in step S113, and merges (combines) the detected pixels.

(Step S116) The third layer determination unit 107 excludes, from the pixels detected in step S115, pixels that overlap with the pixels of the first layer determined in step S109 and pixels that overlap with the pixels of the second layer determined in step S110.

(Step S117) The third layer determination unit 107 excludes pixels that are outside the contour specified by the second instruction received in step S104 from the pixels that remain and were not excluded in step S116.

(Step S118) The third layer determination unit 107 determines the remaining pixels that were not excluded in step S117 as pixels of the third layer. For example, the third layer determination unit 107 associates the determined pixels of the third layer with information indicating that they are pixels of the third layer, and stores them in the medical image storage unit 102 or a storage unit (not shown), etc.

(Step S119) The output unit 108 outputs information indicating the pixels of the third layer determined in step S118. For example, the output unit 108 outputs the pixels of the third layer in a visually distinguishable manner. Note that the output unit 108 may further output information indicating the pixels of the first layer determined in step S109 and the pixels of the second layer determined in step S110. Then, the process returns to step S101.

In the flowchart of FIG. 2, the process of step S103 may be performed after the process of step 104.

In the flowchart in Figure 2, processing ends when the power is turned off or an interrupt occurs to end processing.

The specific operation of the layered organ region identifying device 1 in this embodiment will be described below. Here, it is assumed that the medical image storage unit 102 stores a plurality of medical images of the uterine portion of the human body. These medical images are assumed to be MRI images of cross sections of the uterine portion of the human body perpendicular to the elongation direction of the human body. In such medical images of the uterus, the endometrium, junction zone, and myometrium have a layered structure stacked from the inside to the outside. Here, the description will be made assuming that the first layer is the endometrium, the second layer is the junction zone, and the third layer is the myometrium.

First, if a user operates the layered organ region identification device 1 via an input device such as a keyboard (not shown) to display a medical image of a uterus with the file name "img001" stored in the medical image storage unit 102, the display instruction receiving unit (not shown) receives an instruction to display this medical image "img001", and in response, the medical image display unit 101 displays the medical image with the file name "img001" on the monitor (not shown).

Figure 3 shows an example of a medical image 30 with a file name "img001" displayed by the medical image display unit 101.

Next, the user presses button 31 for inputting a first instruction from the menu button displayed next to medical image 30, visually checks medical image 30 displayed in FIG. 3 with a mouse (not shown) or the like, and performs an operation to specify a pixel in the endometrium, which is the first layer, in this medical image (e.g., an operation of moving the cursor into the endometrium and clicking, etc.). In response to this operation, first reception unit 103 receives a first instruction specifying the specified pixel, and temporarily stores information indicating the specified pixel (e.g., an identifier of the specified pixel, etc.) in a storage unit (not shown) or the like. Here, it is assumed that medical image display unit 101 displays a point representing the pixel specified by the first instruction on medical image 30 shown in FIG. 3.

After receiving the first instruction, the user presses button 32 for inputting a second instruction from the menu button displayed next to medical image 30, and performs an operation with a mouse (not shown) or the like to specify the outer contour of the area including the third layer, the myometrium, in medical image 30 (for example, an operation of moving the cursor along the outer contour of the myometrium while pressing the mouse button). In response to this operation, second reception unit 104 receives a second instruction specifying the outer contour of the area including the myometrium, and temporarily stores information indicating the pixels on the specified contour in a storage unit (not shown) or the like. Here, it is assumed that medical image display unit 101 displays a line representing the contour indicated by the second instruction on medical image 30 shown in FIG. 3.

Figure 4 shows an example of a medical image 30 displayed by the medical image display unit 101, in which a dot 41 indicating the pixel indicated by the first instruction and a line 42 indicating the contour indicated by the second instruction are arranged.

The first layer determination unit 105 reads out the conditions indicating the pixel value range from X-α to X+α for obtaining the first approximation condition, substitutes the pixel value of the pixel indicated by the above-mentioned first instruction as the value of X, and substitutes the initial value _α0 as the value of α. The value of α here is assumed to be a value expressed as a percentage (%) of the pixel value of the pixel indicated by the first instruction. For example, the value of _α0 is assumed to be "30%" here. The first layer determination unit 105 then detects a set of pixels that are continuous with the pixel indicated by the first instruction represented by the point 41 and have pixel values within the range from X-α to X+α (where α is assumed to be _α0 =30%). Specifically, the first layer determination unit 105 detects a set of pixels that have pixel values within the range from (1-0.3)X to (1+0.3)X. Then, it is determined whether a part of the detected pixels overlaps with the contour specified by the second instruction represented by the line 42.

If it does overlap with the contour, the first layer determination unit 105 changes the value of α, a condition indicating the range of pixel values, to a value of "29%" which is a decrease of the above initial value of "30%" by a predetermined value of "1". The first layer determination unit 105 then detects a set of pixels that are adjacent to the pixel indicated by the first instruction represented by point 41 and have pixel values within the range of X-α to X+α, where the above changed α value of "29%" is substituted in place of the initially substituted α value. Specifically, it detects a set of pixels that have pixel values within the range of (1-0.29)X to (1+0.29)X. It then determines whether any part of the detected pixels overlaps with the contour specified by the second instruction represented by line 42.

Here, if even with this value, some of the detected pixels overlap, similarly to the above, the α value of "29%" is further decreased by "1", a set of pixels that are adjacent to the pixel indicated by the first instruction represented by point 41 and have pixel values within the range of X-α to X+α with the decreased α value of "28%" substituted is detected, and a determination is made as to whether some of the detected pixels overlap with the contour specified by the second instruction represented by line 42. The first-layer determination unit 105 decreases the value of α in this way, and repeats the process of determining whether the detected pixels overlap with the contour specified by the second instruction until it is determined that some of the detected pixels do not overlap with the contour specified by the second instruction represented by line 42. Here, for example, if the pixel detected by the first layer determination unit 105 does not overlap with the contour indicated by the second indication shown by line 42 when the value of α is "28%", the first layer determination unit 105 determines the set of pixels adjacent to the pixel indicated by the first indication detected at this time as the pixels of the first layer. Furthermore, the condition in which the value of α at this time is "28%", that is, the condition of detecting pixels having pixel values within the range from X-28% to X+28%, becomes the first approximation condition.

Note that when the initial value α ₀ is substituted for α, if the pixel detected by the first layer determination unit 105 does not overlap with the contour indicated by the second indication shown by the line 42, the first layer determination unit 105 determines the set of pixels adjacent to the pixel indicated by the first indication detected at this time as the pixels of the first layer. Also, the condition under which α ₀ , which is the value of α at this time, is substituted, that is, the condition of detecting pixels having pixel values within the range from X-α ₀ to X+α ₀ , becomes the first approximation condition.

The first layer determination unit 105 determines the set of pixels detected above as the pixels of the first layer, associates the determined first layer pixels with information indicating that they are first layer pixels, and stores them in the medical image storage unit 102. Here, the coordinates of the pixels of the first layer are stored in association with a value indicating that they are first layer pixels.

5(a) is a specific layer management table for managing pixels of layers identified by the layered organ region identifying device 1. The specific layer management table has attributes of "image ID", "coordinates", and "layer". "Image ID" is an identifier of a medical image, and in this case, it is the file name of the medical image. "Coordinates" are the coordinates of a pixel in a medical image. Here, ( _x1 , _y1 ) etc. represent specific coordinates in a medical image. "Layer" indicates the layer to which the pixel indicated by the corresponding coordinates belongs, and a value of "1" indicates that the pixel is a pixel of the first layer, a value of "2" indicates that the pixel is a pixel of the second layer, and a value of "3" indicates that the pixel is a pixel of the third layer.

The record (row) of the pixel determined and stored by the first layer determination unit 105 above is a record with an "image ID" of "img001" and a "layer" of "1".

Next, the second layer determination unit 106 determines the pixels of the medical image 30 as the pixels of the second layer. Here, the second layer determination unit 106 determines the set of pixels within a predetermined range of several pixels outward from the pixels adjacent to the outside of the first layer pixels determined above as the pixels of the second layer. Here, the predetermined range of several pixels is assumed to be within a range of three pixels. The second layer determination unit 106 then associates the determined pixels of the second layer with information indicating that they are pixels of the second layer, and stores them in the medical image storage unit 102.

The information managing the set of pixels in the second layer determined by the second layer determination unit 106 is added to the specific layer management table shown in FIG. 5(a). The specific layer management table to which the information managing the pixels in the second layer determined by the second layer determination unit 106 has been added becomes as shown in FIG. 5(b). In FIG. 5(b), the record (row) of the pixel determined and stored by the second layer determination unit 106 is a record in which the "Image ID" is "img001" and the "Layer" is "2".

Next, the third layer determination unit 107 determines the pixel at the center of gravity of the region composed of the pixels of the first layer determined above as the representative point of the pixels of the first layer. A detailed explanation of the process of determining the center of gravity in the region will be omitted. The center of gravity determined here is assumed to be the pixel with coordinates ( _x500 , _y500 ).

Figure 6 is a diagram showing the representative point 60 and the like determined by the third layer determination unit 107 in the medical image 30. Note that the representative point 60 shown here is a schematic point for the purpose of explanation, and does not necessarily accurately represent the center of gravity of the first layer image determined by the first layer determination unit 105.

Next, the third layer determination unit 107 determines the contour of a figure having a closed area, which has a contour of the same shape as the contour specified by the second instruction (i.e., a shape similar to the contour specified by the second instruction) and is smaller in size than the contour specified by the second instruction, with the representative point 60 determined above as the center. Here, the contour specified by the second instruction is reduced to 70% of its size, with the representative point 60 determined above as the center, to determine the contour of a figure having a closed area 61.

Next, the third layer determination unit 107 acquires the pixel values of each pixel on the contour of the closed region 61 determined above, and sorts the acquired pixel values from the highest pixel value. Then, from the sorted pixel values, pixel values ranked higher than 30% of the total when viewed from the highest pixel value and pixel values ranked lower than 30% of the total when viewed from the lowest pixel value are excluded, and a plurality of pixels ranked from the highest pixel value between 30% and 70% of the total are acquired. For example, if the number of pixel values acquired from the pixels on the contour of the closed region 61 is 500 pixels, the third layer determination unit 107 acquires pixel values ranked from the highest pixel value between 150th and 350th.

The third layer determination unit 107 obtains the average value Y of the multiple pixel values whose highest ranking falls within 30% to 70% of the total as a representative value of two or more pixels on the contour of the closed region 61. The third layer determination unit 107 also obtains a standard deviation σ for the multiple pixel values whose highest ranking falls within 30% to 70% of the total and which were used to obtain the representative value of two or more pixels on the contour of the closed region 61 above. Then, it obtains a second approximation condition that the pixel value of the pixel is within the range of Y-γ to Y+γ (where γ is 2σ).

Next, the third layer determination unit 107 starts from each of the pixels on the contour of the closed region 61 that have pixel values within the range satisfying the second approximation condition obtained above, that is, a plurality of pixels that have pixel values within the range from Y-γ to Y+γ, and sequentially detects consecutive pixels having pixel values that satisfy the second approximation condition obtained above, and merges the detected pixels.

Next, the third layer determination unit 107 excludes, from the multiple pixels obtained by merging, pixels that overlap with the pixels of the first layer determined by the first layer determination unit 105 above and pixels that overlap with the pixels of the second layer determined by the second layer determination unit 106 above. Specifically, from the multiple pixels obtained by merging, it sequentially excludes pixels whose coordinates match the "coordinates" of any one of the records in the specific layer management table shown in FIG. 5(b) where the "image ID" is "img001" and the "layer" is either "1" or "2".

Furthermore, the third layer determination unit 107 sequentially excludes, from the pixels remaining that have not been excluded as described above, pixels whose coordinates are on the outermost side of the contour specified by the second instruction.

Then, the third layer determination unit 107 determines the remaining pixels that were not excluded as pixels of the third layer. For example, the third layer determination unit 107 associates the determined pixels of the third layer with information indicating that they are pixels of the third layer, and stores them in the medical image storage unit 102.

The information managing the set of pixels in the third layer determined by the third layer determination unit 107 is added to the specific layer management table shown in FIG. 5(b). The specific layer management table to which the information managing the pixels in the third layer determined by the third layer determination unit 107 has been added becomes as shown in FIG. 5(c). In FIG. 5(c), the record (row) of the pixel determined and stored by the third layer determination unit 107 is a record in which the "image ID" is "img001" and the "layer" is "3".

The output unit 108 outputs information indicating the pixels of the third layer determined by the third layer determination unit 107. Here, in addition to the image of the third layer, the output unit 108 also outputs information indicating the pixels of the first layer determined by the first layer determination unit 105 and information indicating the pixels of the second layer determined by the second layer determination unit 106.

Here, the output unit 108 first generates an image in which a predetermined first color, red, is superimposed on the pixels of the third layer of the medical image "img001" determined by the third layer determination unit 107. Specifically, it obtains all "coordinate" values of the record (row) in the specific layer management table shown in FIG. 5(b) in which the "image ID" is "img001" and the "layer" is "3", and superimposes the first color onto the pixels at each of the obtained coordinates to composite them. At this time, the first color is composited so that the pixels before the superimposition are visible through the light.

The output unit 108 also generates an image in which a predetermined second color, blue, is superimposed on the pixels of the first layer determined by the first layer determination unit 105 of the medical image "img001" on which the above-mentioned first color has been superimposed. Specifically, it acquires all the "coordinate" values of the record (row) in which the "image ID" in the specific layer management table shown in FIG. 5(b) is "img001" and the "layer" is "1", and superimposes the second color onto the pixels at each of the acquired coordinates, synthesizing them so that the pixels before superimposition are each visible through the color.

The output unit 108 also generates an image in which a predetermined third color, yellow, is superimposed on the pixels of the second layer determined by the second layer determination unit 106 of the medical image "img001" in which the above-mentioned first and second colors are superimposed. Specifically, it obtains all the "coordinate" values of the record (row) in which the "image ID" in the specific layer management table shown in FIG. 5(b) is "img001" and the "layer" is "2", and superimposes the third color onto the pixels of each of the obtained coordinates, synthesizing them so that the pixels before superimposition are visible through each color.

The output unit 108 then displays the image thus generated on a monitor (not shown), thereby outputting the pixels from the first layer to the third layer in a visually distinguishable manner.

Figure 7 is a diagram showing an example of a medical image displayed by the output unit 108, in which the pixels from the first layer to the third layer are displayed in a visually distinguishable manner. In Figure 7, area 71 shows a set of pixels in the first layer, area 72 shows a set of pixels in the second layer, and area 73 shows a set of pixels in the third layer. Note that, for ease of explanation, the first to third colors are represented by different diagonal lines.

The output unit 108 may store the medical image displayed here in the medical image storage unit 102 in association with the medical image "img001."

The output unit 108 may also be configured to output only information indicating the third pixel. For example, the output unit 108 may be configured to display only a medical image in which the first color is superimposed on the pixels of the third layer.

In addition, in this specific example, the layered organ captured in the medical image is the uterus, but the layered organ may be a layered organ other than the uterus, such as the heart.

As described above, according to this embodiment, it is possible to appropriately identify the regions of one or more layers of an organ having a layered structure of three or more layers in a medical image.

In each of the above embodiments, each process (each function) may be realized by centralized processing in a single device (system), or may be realized by distributed processing in multiple devices.

In addition, in each of the above embodiments, if the output performed by the output unit 108 includes display, the configuration for displaying the output unit 108 and the medical image display unit 101 may be realized by a single display unit.

In addition, in each of the above embodiments, the layered organ region identifying device has been described as being standalone, but the layered organ region identifying device may be a standalone device or a server device in a server-client system. In the latter case, the output unit and the reception unit will receive input and output a screen via a communication line.

In addition, in each of the above embodiments, each component may be configured with dedicated hardware, or components that can be realized by software may be realized by executing a program. For example, each component may be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory. During execution, the program execution unit may execute the program while accessing a storage unit (for example, a recording medium such as a hard disk or memory).

The software that realizes the layered organ region identification device in each of the above embodiments is a program such as the following: In other words, this program causes a computer to function as a medical image display unit that displays a medical image of a layered organ having a first layer, a second layer that contacts the outside of the first layer, and a third layer that contacts the outside of the second layer; a first reception unit that receives a first instruction that specifies pixels in the first layer of the medical image; a second reception unit that receives a second instruction that specifies the outer contour of an area including the third layer of the medical image; a first layer determination unit that acquires pixel values of pixels specified by the first instruction and determines a set of pixels that satisfy a first approximation condition for the pixel values and are continuous with the pixels specified by the first instruction as pixels of the first layer; a second layer determination unit that determines a set of continuous pixels adjacent to the outside of the pixels of the first layer as pixels of the second layer; a third layer determination unit that acquires a representative value of pixel values of two or more pixels located outside the pixels of the second layer and within the contour specified by the second instruction, and determines a set of pixels that have a pixel value that satisfies a second approximation condition for the representative value and are located inside the contour specified by the second instruction other than the pixels of the first and second layers as pixels of the third layer; and an output unit that outputs information indicating the pixels of the third layer.

In addition, in the above program, the functions realized by the above program do not include functions that can only be realized by hardware. For example, functions that can only be realized by hardware such as a modem or interface card in an acquisition unit that acquires information or an output unit that outputs information are not included in the functions realized by the above program.

The program may be executed by a single computer or multiple computers. In other words, it may perform centralized processing or distributed processing.

Figure 8 is a schematic diagram showing an example of the appearance of a computer that executes the above program to realize the layered organ region identification device according to the above embodiment. The above embodiment can be realized by computer hardware and a computer program executed thereon.

In FIG. 8, the computer system 900 includes a computer 901 including a CD-ROM (Compact Disk Read Only Memory) drive 905, a keyboard 902, a mouse 903, and a monitor 904.

Figure 9 is a diagram showing the internal configuration of a computer system 900. In Figure 9, the computer 901 includes, in addition to a CD-ROM drive 905, an MPU (Micro Processing Unit) 911, a ROM 912 for storing programs such as a boot-up program, a RAM (Random Access Memory) 913 connected to the MPU 911 to temporarily store application program instructions and provide temporary storage space, a hard disk 914 for storing application programs, system programs, and data, and a bus 915 for interconnecting the MPU 911, the ROM 912, etc. Note that the computer 901 may also include a network card (not shown) for providing a connection to a LAN.

A program that causes the computer system 900 to execute the functions of the layered organ region identifying device according to the above embodiment may be stored in a CD-ROM 921, inserted into the CD-ROM drive 905, and transferred to the hard disk 914. Alternatively, the program may be sent to the computer 901 via a network (not shown) and stored on the hard disk 914. The program is loaded into the RAM 913 when executed. The program may be loaded directly from the CD-ROM 921 or the network.

The program does not necessarily have to include an operating system (OS) or a third-party program that causes the computer 901 to execute the functions of the layered organ region identifying device according to the above embodiment. The program may include only an instruction portion that calls appropriate functions (modules) in a controlled manner to obtain the desired results. How the computer system 900 operates is well known, and a detailed description will be omitted.

The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these are also included within the scope of the present invention.

As described above, the layered organ region identifying device of the present invention is suitable as a device for identifying a desired part in a medical image, and is particularly useful as a device for identifying a desired layer of a layered organ.

REFERENCE SIGNS LIST 1 Layered organ region identifying device 101 Medical image display unit 102 Medical image storage unit 103 First reception unit 104 Second reception unit 105 First layer determination unit 106 Second layer determination unit 107 Third layer determination unit 108 Output unit

Claims (10)

a medical image display unit that displays a medical image of a layered organ having a first layer, a second layer that is in contact with the outside of the first layer, and a third layer that is in contact with the outside of the second layer;
a first receiving unit that receives a first instruction specifying a pixel in the first layer of the medical image;
a second receiving unit that receives a second instruction that specifies an outer contour of a region including the third layer of the medical image;
a first layer determination unit that acquires a pixel value of a pixel designated by the first instruction, and determines a set of pixels that satisfy a first approximation condition for the pixel value and are adjacent to the pixel designated by the first instruction as pixels of a first layer;
a second layer determination unit that determines a set of adjacent pixels adjacent to an outer side of the first layer pixel as a second layer pixel;
a third layer determination unit that obtains a representative value of pixel values of two or more pixels located outside the pixels of the second layer and within the contour specified by the second instruction, and determines a set of pixels that have a pixel value that satisfies a second approximation condition for the representative value and are located inside the contour specified by the second instruction, other than the pixels of the first and second layers, as pixels of a third layer;
An output unit that outputs information indicating the pixels of the third layer. 2. The device for identifying a layered organ region according to claim 1, wherein the first approximation condition is that the pixel value is within the range of X- _α1 to X+ _α1 (where X is the pixel value of the pixel specified by the first instruction, and _α1 is a value such that a set of pixels that are adjacent to the pixel specified by the first instruction and that satisfy the condition do not overlap with a pixel of the contour specified by the second instruction). The first layer determination unit is
3. The device for identifying a layered organ region according to claim 2, further comprising: a detecting unit for detecting a set of pixels that are adjacent to the pixel designated by the first instruction and that satisfy the condition that the pixel value is within the range of X-α to X+α (where X is the pixel value of the pixel designated by the first instruction), the value of α is decreased from a predetermined initial value until at least a part of the set of pixels that satisfies the condition goes from overlapping with the pixel of the contour designated by the second instruction to no longer overlapping, and the device determines the set of pixels detected under the first approximation condition, which is the condition having an α value of _α1 that is the α value at which the set of pixels no longer overlaps, as the pixels of the first layer. The second layer decision unit,
A layered organ region identification device as described in any one of claims 1 to 3, wherein a set of pixels within a range from a pixel adjacent to the outside of the first layer pixel of the medical image to a predetermined number of pixels outward is determined to be the pixels of the second layer. The third layer decision unit,
A layered organ region identification device as described in any one of claims 1 to 4, which determines a representative point of the pixels of the first layer, determines a contour of a figure having a closed area centered on the representative point and smaller in size than the contour specified by the second instruction, and obtains a representative value of the pixel values for two or more pixels on the contour. The second approximation condition is:
The layered organ region identification device of claim 5, wherein the condition is that the pixel value is within the range of Y-γ to Y+γ (where Y is a representative value of the pixel, and γ is a value obtained using the pixel values of two or more pixels on the contour to indicate a range of pixel values that are likely to be pixels of the third layer). The layered organ region identification device according to any one of claims 1 to 6, wherein the output unit further outputs information indicating the pixels of the first and second layers. said layered organ being the uterus and said first layer being the endometrium;
The second layer is a junction zone (muscular layer just below the intima),
The layered organ region identifying device according to claim 1 , wherein the third layer is a uterine muscle layer. A layered organ region determining method performed using a medical image display unit, a first receiving unit, a second receiving unit, a first layer determining unit, a second layer determining unit, a third layer determining unit, and an output unit, comprising:
a medical image display step in which the medical image display unit displays a medical image of a layered organ having a first layer, a second layer in contact with the outside of the first layer, and a third layer in contact with the outside of the second layer;
a first reception step in which the first reception unit receives a first instruction specifying a pixel in the first layer of the medical image;
a second receiving step in which the second receiving unit receives a second instruction specifying an outer contour of a region including the third layer of the medical image;
a first layer determination step in which the first layer determination unit acquires a pixel value of a pixel designated by the first instruction, and determines a set of pixels that satisfy a first approximation condition for the pixel value and are adjacent to the pixel designated by the first instruction as pixels of a first layer;
a second layer determination step in which the second layer determination unit determines a set of consecutive pixels adjacent to the outside of the first layer pixel as a second layer pixel;
a third layer determination step in which the third layer determination unit obtains a representative value of pixel values of two or more pixels located outside the pixels of the second layer and within the contour specified by the second instruction, and determines a set of pixels that have a pixel value that satisfies a second approximation condition for the representative value and are located inside the contour specified by the second instruction, other than the pixels of the first and second layers, as pixels of a third layer;
An output step in which the output unit outputs information indicating the pixels of the third layer. Computer,
a medical image display unit that displays a medical image of a layered organ having a first layer, a second layer that is in contact with the outside of the first layer, and a third layer that is in contact with the outside of the second layer;
a first receiving unit that receives a first instruction specifying a pixel in the first layer of the medical image;
a second receiving unit that receives a second instruction that specifies an outer contour of a region including the third layer of the medical image;
a first layer determination unit that acquires a pixel value of a pixel designated by the first instruction, and determines a set of pixels that satisfy a first approximation condition for the pixel value and are adjacent to the pixel designated by the first instruction as pixels of a first layer;
a second layer determination unit that determines a set of adjacent pixels adjacent to an outer side of the first layer pixel as a second layer pixel;
a third layer determination unit that obtains a representative value of pixel values of two or more pixels located outside the pixels of the second layer and within the contour specified by the second instruction, and determines a set of pixels that have a pixel value that satisfies a second approximation condition for the representative value and are located inside the contour specified by the second instruction, other than the pixels of the first and second layers, as pixels of a third layer;
A program for causing the output unit to function as an output unit that outputs information indicating the pixels of the third layer.

Images