JP7757142B2 - Clinical support system and clinical support device - Google Patents

Description

The embodiments disclosed in this specification and drawings relate to a clinical support system and a clinical support device.

Traditionally, doctors and other medical professionals have been required to provide medical care based not only on biological aspects such as test results, but also on the social and psychological aspects of patients. As a result, medical professionals collect information about psychological and social aspects through communication with patients.

However, it can be difficult to collect information on psychological and social aspects, for example, and it is even more difficult to utilize this information while taking an overall view. Therefore, there is a demand for technology that supports communication during medical treatment.

Special Publication No. 2014-503894

One of the problems that the embodiments disclosed in this specification and drawings attempt to solve is to support communication during medical treatment. However, the problems that the embodiments disclosed in this specification and drawings attempt to solve are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems.

A clinical support system according to an embodiment includes a first acquisition unit, a first analysis unit, a second acquisition unit, a second analysis unit, and a display control unit. The first acquisition unit acquires patient information related to a first patient, medical interview information indicating the symptoms of the first patient, and conversation information recording the conversation of the first patient. The first analysis unit generates patient characteristic information indicating psychological and social aspects of the patient through an analytical process using the patient information, medical interview information, and conversation information. The second acquisition unit acquires the patient characteristic information of the first patient generated by the first analysis unit and the conversation information of a second patient whose patient characteristic information is similar. The second analysis unit analyzes characteristics of the conversation between the second patient and a medical professional based on the conversation information acquired by the second acquisition unit. The display control unit displays the analysis results by the second analysis unit.

FIG. 1 is a block diagram showing an example of the configuration of a clinical support system according to the first embodiment. FIG. 2 is a block diagram showing an example of the configuration of the clinical support apparatus according to the first embodiment. FIG. 3 is a diagram illustrating an example of the data configuration of the analysis target information. FIG. 4 is an explanatory diagram illustrating an example of a conversation analysis screen according to the first embodiment. FIG. 5 is a flowchart illustrating an example of a display process executed by the clinical support device according to the first embodiment. FIG. 6 is a block diagram showing an example of the configuration of a clinical support apparatus according to the second embodiment. FIG. 7 is a flowchart illustrating an example of a display process executed by the clinical support apparatus according to the second embodiment. FIG. 8 is a flowchart illustrating an example of a display process executed by the clinical support device according to the first modification of the second embodiment.

The following describes the clinical support system and clinical support device according to the embodiments, with reference to the drawings. In the following embodiments, parts with the same reference numerals perform similar operations, and redundant explanations will be omitted where appropriate.

(First embodiment)
Doctors and other medical professionals are expected to provide holistic medical care that takes into account not only biological aspects but also the psychological and social aspects of patients.More specifically, medical professionals are expected to determine treatment plans that take into account various aspects, rather than determining treatment plans based solely on biological aspects such as the results of diagnostic imaging and clinical tests.

For example, patients with chronic illnesses such as lifestyle-related diseases may not be cured by treating them from a biological perspective solely through the administration of medication. For example, in the case of lower back pain, it is preferable for medical professionals to not only prescribe a medicated plaster, but also to provide advice tailored to the patient's lifestyle, such as their work.

Healthcare professionals can expect greater improvement if they implement treatment that encourages lifestyle improvements rather than simply administering medication. Furthermore, by determining treatment plans that take into account the patient's parental status and work status, patients can improve adherence (patients actively participating in treatment and medication and receiving treatment in accordance with their decisions) through treatment that suits their lifestyle both during and after treatment, leading to better quality of life and better outcomes. Furthermore, patients can concentrate on their treatment with peace of mind knowing that explanations and treatment are tailored to their individual personalities.

In this way, medical professionals can provide better medical care by determining treatment plans based on biological, psychological, and social aspects. In other words, medical professionals are expected to provide so-called holistic medical care that includes not only biological aspects but also psychological and social aspects. Medical professionals also collect information about psychological and social aspects through communication with patients.

However, medical professionals communicate with patients based on their own training and experience. However, they may find it difficult to gather information related to psychological and social aspects of patients. Furthermore, medical professionals must decide which psychological and social aspects of patients to record in electronic medical records. Furthermore, medical professionals are required to utilize this information while maintaining a comprehensive overview.

On the other hand, experienced medical professionals are able to communicate smoothly with patients based on their own experience and knowledge. Therefore, there is a need for technology that supports medical practice by helping medical professionals learn from others by referring to key points in communication between other medical professionals and patients.

FIG. 1 is a block diagram showing an example of the configuration of a clinical support system 1 according to the first embodiment. The clinical support system 1 includes a Hospital Information System (HIS) 10, a Radiology Information System (RIS) 20, a Picture Archiving and Communication System (PACS) 30, a Laboratory Information System (LIS) 40, a patient terminal 50, a patient measurement terminal 60, a clinical support device 70, and a display device 80. Each system and device is connected to each other via a network 90 for mutual communication. The configuration shown in FIG. 1 is an example, and the number of each system and device may be changed as desired. Devices not shown in FIG. 1 may also be connected to the network 90.

The hospital information system 10, radiology information system 20, medical image management system 30, and clinical test information system 40 are realized by computer equipment such as a server or workstation.

The hospital information system 10 stores electronic medical record information 11 and medical interview information 12. The electronic medical record information 11 is information that records the progress of a patient's medical treatment. For example, the electronic medical record information 11 includes information for identifying the patient, personal information such as the patient's age, gender, and family composition, the name of the disease the patient has, the name of the prescribed medication, and the treatment period. The medical interview information 12 is information that indicates the patient's responses to a medical interview. For example, the medical interview information 12 includes responses to the patient's current symptoms, responses to the patient's living environment, and responses to a personality assessment.

The radiology information system 20 stores image interpretation report information 21. The image interpretation report information 21 contains findings from a doctor or other person who interpreted image information 31 captured by a medical image diagnostic device.

The medical image management system 30 stores image information 31. The image information 31 includes images captured by a medical image diagnostic device. Examples of medical image diagnostic devices include X-ray CT (Computed Tomography) devices, MRI (Magnetic Resonance Imaging) devices, X-ray diagnostic devices, and ultrasound diagnostic devices. The image information 31 conforms to the DICOM (Digital Imaging and Communications in Medicine) standard.

The clinical test information system 40 stores clinical test result information 41 and clinical test report information 42. Clinical test result information 41 is information showing the results of clinical tests. Clinical test report information 42 is information containing findings on the clinical test results.

The patient terminal 50 is a terminal used by the patient. The patient terminal 50 may be realized, for example, by a smartphone, tablet terminal, wearable terminal, or personal computer. The patient terminal 50 also stores conversation information 51, which is a record of the patient's conversation. For example, the conversation information 51 is information about conversations with medical professionals such as doctors, nurses, and counselors. Specifically, the conversation information 51 includes audio information of recorded conversations. The conversation information 51 may also include video information such as videos or still images that include the conversation. If video information is included, the conversation information 51 includes video that shows the patient's facial expressions.

The patient measurement terminal 60 is a terminal that measures the patient's vital signs. The patient measurement terminal 60 is realized, for example, by a wearable terminal. The patient measurement terminal 60 measures vital signs such as blood pressure, pulse, body temperature, and respiration. The patient measurement terminal 60 may also record conversations. The patient measurement terminal 60 then stores personal health information 61 containing the vital sign measurement results.

The patient measurement terminal 60 is not limited to a wearable device, but may also be a body fat scale, an activity monitor, or other devices. The patient measurement terminal 60 may measure body fat, activity, or other items. Furthermore, the clinical support system 1 may be equipped with multiple patient measurement terminals 60. Furthermore, in the clinical support system 1, each of the multiple patient measurement terminals 60 may measure different measurement targets.

The clinical support device 70 is a device that supports medical professionals such as doctors in clinical practice. The clinical support device 70 is realized, for example, by computer equipment such as a server or workstation. The clinical support device 70 analyzes the biological, social, and psychological aspects of a patient based on information acquired from each system and device of the clinical support system 1. The clinical support device 70 also analyzes information that does not belong to the biological, social, or psychological aspects of the patient, i.e., other aspects. The clinical support device 70 also generates information indicating the biological, social, psychological, and other aspects of the patient. The clinical support device 70 may be installed in a facility such as a hospital, or may be a server on the Internet.

The display device 80 is a device capable of displaying various types of information. The display device 80 is realized by a computer device such as a personal computer or a tablet terminal. The display device 80 displays various types of information generated by the clinical support device 70. For example, the display device 80 displays a conversation analysis screen G1 (see Figure 4). The display device 80 may also be a transmissive display that displays different information on the medical professional side and the patient side. The display device 80 is an example of a display control unit.

By viewing the information displayed on the display device 80, medical professionals such as doctors can refer to the analysis results of conversations between other medical professionals and patients during medical treatment. This allows medical professionals to refer to conversations between other medical professionals and patients during medical treatment.

Next, we will explain the clinical support device 70.

Figure 2 is a block diagram showing an example of the configuration of a clinical support device 70 according to the first embodiment. The clinical support device 70 includes a network interface 710, an input interface 720, a display 730, a memory circuitry 740, and a processing circuitry 750.

The NW interface 710 is connected to the processing circuit 750 and controls the transmission and communication of various data between each device connected via the network 90. For example, the NW interface 710 is realized by a network card, network adapter, NIC (Network Interface Controller), etc.

The input interface 720 is connected to the processing circuit 750 and converts input operations received from an operator (medical professional) into electrical signals and outputs them to the processing circuit 750. Specifically, the input interface 720 converts input operations received from the operator into electrical signals and outputs them to the processing circuit 750. For example, the input interface 720 may be implemented by a trackball, switch buttons, a mouse, a keyboard, a touchpad that performs input operations by touching the operation surface, a touchscreen that integrates a display screen and a touchpad, a non-contact input circuit using an optical sensor, an audio input circuit, or the like. Note that in this specification, the input interface 720 is not limited to those equipped with physical operating components such as a mouse and keyboard. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the device and outputs these electrical signals to a control circuit is also included as an example of the input interface 720.

The display 730 is connected to the processing circuit 750 and displays various information and image data output from the processing circuit 750. For example, the display 730 may be implemented as a liquid crystal display, a CRT (Cathode Ray Tube) display, an organic EL display, a plasma display, a touch panel, etc.

The memory circuitry 740 is connected to the processing circuitry 750 and stores various data. The memory circuitry 740 also stores various programs that the processing circuitry 750 reads and executes to realize various functions. For example, the memory circuitry 740 may be implemented using semiconductor memory elements such as RAM (Random Access Memory) or flash memory, a hard disk, an optical disk, etc.

The memory circuit 740 stores analysis target information 741, reference conversation information 742, and standard treatment information 743.

Figure 3 shows an example of the data configuration of the analysis target information 741. The analysis target information 741 is information acquired by each system and device in the clinical support system 1, and specifies the information to be analyzed as biological, social, psychological, or other aspects of the patient.

The analysis target information 741 associates categories, analysis items, and analysis target items. Categories are divided into biological aspects, psychological aspects, social aspects, and others. Biological aspects are classifications that indicate the biological aspects of the patient. For example, biological analysis items such as the patient's various test results, diseases, and physical condition are associated with the biological aspects. Furthermore, biological aspects are not limited to biological aspects, and analysis items related to physical aspects, medical aspects, and pathological aspects may also be associated with the biological aspects. Psychological aspects are classifications that indicate the psychological aspects of the patient. For example, psychological analysis items such as the patient's intentions and personality are associated with the psychological aspects. Furthermore, psychological aspects may also include the patient's emotional state. Emotional states include feelings such as depression and anger. Social aspects are classifications that indicate the social aspects of the patient. For example, social analysis items such as work, interpersonal relationships, hobbies, and educational background are associated with the social aspects. Other analysis items that do not belong to the biological, psychological, or social aspects are associated with the social aspects. Analysis items are items for analyzing a patient. Analysis target items are information that specifies whether or not an analysis item is to be analyzed. In other words, the clinical support device 70 analyzes the analysis items that are set as analysis targets in the analysis target items. Also, the association between the classification of analysis target information 741 and analysis items shown in Figure 3 is an example and may be changed as desired.

Reference conversation information 742 is information that has multiple pieces of conversation information 51 to be referenced. Reference conversation information 742 has multiple pieces of information that associate patient characteristic information, conversation information 51, and medical professional identification information. Patient characteristic information is the analysis results for each analysis item of the patient's biological, psychological, social, or other aspects. Conversation information 51 is information that records conversations between patients in the patient characteristic information and medical professionals. Medical professional identification information is identification information for identifying medical professionals who conversed with patients in the conversation information 51. Furthermore, medical professional identification information may be the medical professional's years of experience, the medical department to which the medical professional belongs, the medical professional's classification such as specialist, a combination of these pieces of information, or other information.

Standard treatment information 743 is information in which standard treatment plans for injuries and illnesses are registered. Standard treatment information 743 includes injury/illness identification information for identifying the injury or illness, and treatment plan information that indicates the treatment plan. Injury/illness identification information is, for example, the name of the injury or illness, or code information for identifying the injury or illness. Treatment plan information is information that indicates the standard treatment for the injury or illness. For example, treatment plan information is information that indicates the treatment plan recommended by each facility such as a hospital, or the treatment plan recommended by an academic society.

The processing circuitry 750 controls the overall operation of the clinical support device 70. The processing circuitry 750 has, for example, a patient designation function 751, an information acquisition function 752, an analysis target designation function 753, a feature analysis function 754, a similarity calculation function 755, a similar patient selection function 756, a conversation information acquisition function 757, a conversation analysis function 758, a standard treatment acquisition function 759, and an information generation function 760. In the embodiment, each processing function performed by the components of the patient designation function 751, the information acquisition function 752, the analysis target designation function 753, the feature analysis function 754, the similarity calculation function 755, the similar patient selection function 756, the conversation information acquisition function 757, the conversation analysis function 758, the standard treatment acquisition function 759, and the information generation function 760 is stored in the memory circuitry 740 in the form of a computer-executable program. The processing circuitry 750 is a processor that reads and executes programs from the memory circuitry 740 to realize the functions corresponding to each program. In other words, once each program has been loaded, the processing circuit 750 will have the functions shown in the processing circuit 750 in Figure 2.

Note that while Figure 2 illustrates a single processor realizing the patient designation function 751, information acquisition function 752, analysis target designation function 753, feature analysis function 754, similarity calculation function 755, similar patient selection function 756, conversation information acquisition function 757, conversation analysis function 758, standard of care acquisition function 759, and information generation function 760, it is also possible to combine multiple independent processors to form the processing circuit 750, and have each processor execute a program to realize the functions. Also, while Figure 2 illustrates a single memory circuit such as memory circuit 740 storing programs corresponding to each processing function, it is also possible to have multiple memory circuits distributed and have the processing circuit 750 read corresponding programs from individual memory circuits.

The term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), GPU (Graphical Processing Unit), or an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor achieves its functions by reading and executing a program stored in the memory circuitry 740. Note that instead of storing a program in the memory circuitry 740, the processor may be configured so that the program is directly embedded in its circuitry. In this case, the processor achieves its functions by reading and executing the program embedded in its circuitry.

The patient designation function 751 accepts the designation of a patient for whom a treatment plan is to be created. In other words, the patient designation function 751 accepts the designation of a patient for whom the physical, social, and psychological aspects of the patient will be analyzed. The patient designation function 751 accepts the designation of a patient from, for example, the display device 80. For example, the patient designation function 751 accepts information that can identify the patient, such as the patient's name or a patient code for identifying the patient. Alternatively, the patient designation function 751 accepts the designation of a patient by selecting from a list of patients.

The information acquisition function 752 acquires patient information related to the patient designated by the patient designation function 751, medical interview information 12 indicating the patient's responses to medical interviews, and conversation information 51 recording the patient's conversation. The information acquisition function 752 is an example of a first acquisition unit. Here, the patient information is information monitoring the patient, such as electronic medical record information 11, radiology report information 21, image information 31, clinical test result information 41, clinical test report information 42, and personal health information 61. For example, when a patient is designated by the patient designation function 751, the information acquisition function 752 acquires the patient information, medical interview information 12, and conversation information 51 of the designated patient.

The analysis target designation function 753 designates the analysis target items, which are the analysis items to be analyzed by the characteristic analysis function 754, from among the analysis items for the biological, psychological, and social aspects of the patient. In other words, the analysis target designation function 753 designates the analysis target items of the analysis target information 741.

More specifically, the analysis target designation function 753 accepts the designation of an analysis target to be analyzed using the patient information, medical interview information 12, and conversation information 51 acquired by the information acquisition function 752. For example, the analysis target designation function 753 accepts the designation of an analysis target from information received by the NW interface 710 or an operation accepted by the input interface 720. For example, the analysis target designation function 753 accepts the designation of an analysis target from the display device 80 via the NW interface 710. The analysis target designation function 753 then designates the analysis target items of the designated analysis target information 741 as the analysis target.

The feature analysis function 754 generates patient feature information indicating biological, psychological, social, and other aspects of the patient through an analytical process using the patient information, medical interview information 12, and conversation information 51. The feature analysis function 754 is an example of a first analysis unit. The analytical process involves extracting, determining, analyzing, and generating new information from the patient information, medical interview information 12, and conversation information 51. The patient feature information is information indicating the patient's features in biological, psychological, social, or other aspects. The patient feature information is also the analysis result for each analysis item of the patient's biological, psychological, social, or other aspects.

More specifically, the feature analysis function 754 generates patient feature information for each of the analysis target items of the patient's biological, psychological, and social aspects. The feature analysis function 754 generates patient feature information that corresponds to the analysis target items of the biological aspects specified by the analysis target information 741 from the information contained in the patient information. Specifically, the feature analysis function 754 generates patient feature information that indicates the coexisting diseases, physical condition, etc. contained in the patient information.

Furthermore, the feature analysis function 754 generates patient feature information for the analysis target items of psychological and social aspects specified by the analysis target information 741 from the patient information, medical interview information 12, and information included in the conversation information 51. For example, the feature analysis function 754 generates patient feature information from the medical interview information 12 that corresponds to the analysis target items of psychological and social aspects specified by the analysis target information 741. Specifically, the feature analysis function 754 extracts answers to questions about personality judgment, work, interpersonal relationships, etc. included in the medical interview information 12.

In addition, the feature analysis function 754 generates patient feature information including conversation content information and emotional information by analyzing the conversation information 51. The conversation content information is information that indicates the content of the conversation. The emotional information is information that indicates the patient's emotions when each phrase in the conversation was spoken.

More specifically, the feature analysis function 754 generates patient feature information including conversation content information and emotional information corresponding to the analysis target items of psychological and social aspects specified by the analysis target information 741. For example, the feature analysis function 754 generates conversation content information indicating the content of the conversation by performing natural language processing on the conversation information 51. For example, if the conversation information 51 includes the utterance "My lower back hurts," the feature analysis function 754 generates the sentence "My lower back hurts."

The feature analysis function 754 also generates emotional information that indicates the emotions of the patient who spoke. Here, if the patient's symptoms are worsening, the patient will often speak with negative emotions. On the other hand, if the patient's symptoms are improving, the patient will often speak with positive emotions. For example, the feature analysis function 754 generates emotional information that indicates the patient's emotions expressed by the tone of voice, hesitation, nuance, etc. when saying, "My lower back hurts." In this way, the feature analysis function 754 generates patient feature information in which emotional information is associated with conversation content information.

For example, the feature analysis function 754 may utilize artificial intelligence such as AI (Artificial Intelligence). For example, the feature analysis function 754 generates emotional information indicating the emotions of the speaker of the conversation included in the conversation information 51 using a trained model generated by machine learning. The trained model is generated by supervised learning, in which the conversation information 51 is input as training data on the input side, and emotional information indicating the emotions of the speaker is input as training data on the output side. Note that the trained model is not limited to this type of machine learning, and may be generated by other methods as well.

Furthermore, the feature analysis function 754 may generate information indicating the patient's emotional information and values by comparing the conversation information 51 with other information such as patient information and medical interview information 12. For example, if the patient answers that their standing work is moderate in intensity in the medical interview information 12 and that their standing work is six hours in the conversation information 51, the feature analysis function 754 can generate information indicating the patient's values indicating that they perceive six hours as moderate intensity.

Furthermore, as time passes, the patient's social and psychological situation changes. Therefore, the characteristic analysis function 754 may generate patient characteristic information at regular intervals, or may generate patient characteristic information for each medical event such as an examination or notification.

Note that the characteristic analysis function 754 is not limited to generating all of the patient's biological, psychological, social, and other aspects, but may also generate patient characteristic information indicating psychological and social aspects.

The similarity calculation function 755 calculates the degree of similarity between the patient characteristic information generated by the characteristic analysis function 754 and each of the multiple pieces of patient characteristic information contained in the reference conversation information 742. More specifically, the similarity calculation function 755 calculates the similarity by comparing each analysis item contained in the patient characteristic information. For example, the similarity calculation function 755 calculates the similarity by counting the number of matching analysis items. Note that the similarity calculation function 755 is not limited to a perfect match, and may determine that the analysis items match when they can be determined to be approximately identical. The above-mentioned method of calculating similarity is one example, and other methods may also be used for calculation.

The similar patient selection function 756 selects a similar patient, who is a patient similar to the patient specified by the patient specification function 751. A similar patient is an example of a second patient. More specifically, the similar patient selection function 756 selects a similar patient based on the similarity between the patient characteristic information of the patient generated by the characteristic analysis function 754 and each patient characteristic information in the reference conversation information 742. For example, the similar patient selection function 756 selects as similar patients patients whose patient characteristic information has a similarity equal to or greater than a threshold. Note that the similar patient selection function 756 may select as similar patients patients patients whose patient characteristic information has the highest similarity, up to a set number of similarities, without being limited to the threshold, or may select as similar patients using other methods.

The conversation information acquisition function 757 acquires the patient characteristic information of the patient generated by the characteristic analysis function 754 and the conversation information 51 of a similar patient whose patient characteristic information is similar. The conversation information acquisition function 757 is an example of a second acquisition unit. More specifically, the conversation information acquisition function 757 acquires the conversation information 51 of a similar patient selected from one or more similar patients selected by the similar patient selection function 756 from the reference conversation information 742. Alternatively, when the similar patient selection function 756 selects one similar patient, the conversation information acquisition function 757 may acquire the conversation information 51 of the similar patient selected by the similar patient selection function 756 from the reference conversation information 742.

The conversation analysis function 758 analyzes the characteristics of the conversation between the patient and the medical professional based on the conversation information 51 acquired by the conversation information acquisition function 757. The conversation analysis function 758 is an example of a second analysis unit. More specifically, the conversation analysis function 758 analyzes the names of each stage of the conversation based on the conversation information 51 acquired by the conversation information acquisition function 757. For example, the stages of a conversation during a medical examination may include an introduction section in which the patient's recent condition is asked, an announcement section in which the patient is informed of the illness they are suffering from, an explanation section in which the illness they are suffering from is explained, and a conclusion section in which future plans are explained. The conversation analysis function 758 performs natural language processing on the conversation information 51 acquired by the conversation information acquisition function 757 to analyze which stage the conversation falls into.

The conversation analysis function 758 also analyzes the duration of each stage of the conversation between the similar patient and the medical professional based on the conversation information 51 acquired by the conversation information acquisition function 757. That is, the conversation analysis function 758 analyzes how many minutes of conversation took place in the introductory section, the announcement section, the explanation section, and the conclusion section. The conversation analysis function 758 also analyzes the total duration of the conversation between the similar patient and the medical professional, the duration of the speech by the medical professional, and the duration of the speech by the patient.

The conversation analysis function 758 also analyzes the emotions of the similar patient at each stage of the conversation between the similar patient and the medical professional, based on the conversation information 51 acquired by the conversation information acquisition function 757. For example, the conversation analysis function 758 analyzes emotional values that quantify the positive, negative, and other emotions of the similar patient at each stage of the conversation. Note that the conversation analysis function 758 is not limited to positive and negative, and may also analyze whether the similar patient is feeling joy, anger, sadness, or happiness. Here, the feature analysis function 754 generates emotional information that indicates the emotions of the patient uttered at each stage. For example, the conversation analysis function 758 analyzes the emotions of the similar patient at each stage by acquiring emotional information for each stage.

The conversation analysis function 758 also analyzes keywords for each process in the conversation between similar patients and medical professionals based on the conversation information 51. For example, the conversation analysis function 758 analyzes keywords that indicate the theme of the conversation in each process. The conversation analysis function 758 analyzes the theme by performing natural language processing on the linguistic information in the conversation information 51 for each process. The conversation analysis function 758 then acquires the analyzed theme as a keyword for each process. Note that the conversation analysis function 758 is not limited to the theme of the conversation in each process, and may also acquire characteristic remarks as keywords.

The conversation analysis function 758 also extracts treatments that were considered for implementation in the conversation between the second patient and the medical professional based on the conversation information 51. For example, the conversation analysis function 758 uses natural language processing to extract proposed treatment information indicating the treatment proposed by the medical professional to the patient. The conversation analysis function 758 also uses natural language processing to extract selected treatment information indicating the treatment selected for implementation.

The standard treatment acquisition function 759 acquires the standard treatment that is typically performed for the illness or injury of the patient specified by the patient designation function 751 from the standard treatment information 743. More specifically, the standard treatment acquisition function 759 acquires illness or injury identification information, such as the name of the illness or injury of the patient specified by the patient designation function 751, from the electronic medical record information 11. The standard treatment acquisition function 759 also acquires treatment policy information indicating the standard treatment that corresponds to the acquired illness or injury identification information from the standard treatment information 743.

The information generation function 760 generates information to be displayed on the display device 80. The display device 80 then displays the various information generated by the information generation function 760. More specifically, the information generation function 760 generates information indicating the analysis results obtained by the conversation analysis function 758. The information generation function 760 is an example of an information generation unit. For example, the information generation function 760 generates a conversation analysis screen G1 that indicates the analysis results obtained by the conversation analysis function 758 of a conversation between a similar patient and a medical professional.

Here, Figure 4 is an explanatory diagram showing an example of the conversation analysis screen G1 according to the first embodiment. The conversation analysis screen G1 includes patient summary information G11, treatment selection flow information G12, first conversation flow information G13a, and second conversation flow information G13b.

Patient summary information G11 is information that provides an overview of similar patients. More specifically, patient summary information G11 is information that indicates the similarities between similar patients, the medical professionals who have conversed with the similar patients, and the current status of the similar patients. The information generation function 760 acquires patient characteristic information for analysis items determined to match by the similarity calculation function 755, medical professional identification information associated with conversation information 51 in the reference conversation information 742, and electronic medical record information 11 of similar patients selected by the similar patient selection function 756.

The information generation function 760 also generates patient summary information G11 based on the patient characteristic information, electronic medical record information 11, and medical professional identification information. That is, the information generation function 760 generates information indicating the similarities between similar patients based on the patient characteristic information for the analysis items determined to match by the similarity calculation function 755. By presenting the similarities in this way, medical professionals can inform patients of the similarities between similar patients and explain treatment plans, etc. This often makes patients feel that they are being treated with compassion, allowing medical professionals to communicate smoothly. The information generation function 760 also generates information indicating the medical professionals who spoke with the similar patient based on the medical professional identification information associated with the conversation information 51. The information generation function 760 also generates information indicating the current status of the similar patient based on the electronic medical record information 11 of the similar patient selected by the similar patient selection function 756.

The information generation function 760 may also present information indicating the patient's characteristics in the patient summary information G11. For example, the information generation function 760 may display whether the patient is one who will avoid risky treatments, or one who will choose risky treatments if they are expected to have favorable effects. This allows medical professionals to determine what type of explanation is appropriate for the patient. Furthermore, the information generation function 760 may present a list of paraphrases so that explanations can be rephrased according to the patient's type. The information generation function 760 may also present examples of failed communication with similar patients.

The treatment selection flow information G12 indicates the flow leading up to the selection of a treatment to be administered to a similar patient. For example, the treatment selection flow information G12 indicates the standard treatment typically administered for the illness or injury of a similar patient, the proposed treatment proposed by a medical professional to the similar patient, and the selected treatment. The information generation function 760 generates the treatment selection flow information G12 based on the treatment policy information acquired by the standard treatment acquisition function 759, the proposed treatment information extracted by the conversation analysis function 758, and the selected treatment information. In other words, the information generation function 760 generates information indicating the standard treatment typically administered for the illness or injury of a similar patient, the treatment proposed by a medical professional to the similar patient, and the selected treatment. By referring to the flow for narrowing down the selected treatments shown in the treatment selection flow information G12, medical professionals can refer to the conversation flow used to explain the treatment. Note that the treatment selection flow information G12 shown in Figure 4 narrows down the selected treatments through three stages: standard treatment, proposed treatment, and selected treatment. However, the treatment selection flow information G12 may have four or more stages.

The first conversation flow information G13a and the second conversation flow information G13b are information indicating the flow of conversation between a similar patient and a medical professional during the medical professional's examination of the similar patient. When the first conversation flow information G13a and the second conversation flow information G13b are not distinguished, they are referred to as conversation flow information G13. In the conversation analysis screen G1 shown in FIG. 4, the information generation function 760 generates two pieces of conversation flow information G13, the first conversation flow information G13a and the second conversation flow information G13b, because the similar patient and the medical professional have examined each other twice. In other words, the information generation function 760 generates information indicating the number of conversation flow information G13 corresponding to the number of examinations. The conversation flow information G13 includes conversation time information G131, emotion value information G132, process information G133, and emotion distribution information G134.

Conversation time information G131 is information indicating the duration of a conversation between a similar patient and a medical professional. For example, conversation time information G131 includes the total duration of conversation between the similar patient and the medical professional, the duration of speech by the medical professional, and the duration of speech by the patient. In this way, the information generation function 760 generates information indicating the duration of each step in a conversation between a similar patient and a medical professional. More specifically, the information generation function 760 generates the conversation time information G131 based on the total duration of conversation between the similar patient and the medical professional, the duration of speech by the medical professional, and the duration of speech by the patient, which are analyzed by the conversation analysis function 758.

Emotional value information G132 is information that includes a graph showing changes in emotional value and keywords for each emotional value. For example, emotional values are a numerical representation of the similar patient's positive or negative emotions. The graph of emotional value information G132 shows emotional values over time in a conversation between a similar patient and a medical professional during a medical examination. Keywords indicate themes of the conversation, characteristic remarks, and remarks that significantly change the patient's emotions. In this way, the information generation function 760 generates information that shows the emotions of the similar patient at each stage in the conversation between the similar patient and the medical professional. The information generation function 760 generates information that shows keywords for each stage in the conversation between the similar patient and the medical professional. More specifically, the information generation function 760 generates emotional value information G132 based on the emotional values and keywords analyzed by the conversation analysis function 758.

Process information G133 is information indicating the name of each process in the conversation between a similar patient and a medical professional during a medical examination. The information generation function 760 generates process information G133 based on the name of each process in the conversation analyzed by the conversation analysis function 758. In the process information G133 shown in Figure 4, each process is of equal length, but the length may also correspond to the conversation time of each process.

Emotion distribution information G134 is information that indicates the distribution of emotions of similar patients during a medical examination. For example, the emotion distribution indicates the distribution of emotions of joy, anger, sadness, and happiness of similar patients. The information generation function 760 generates emotion distribution information G134 based on the emotions of joy, anger, sadness, and happiness of similar patients analyzed by the conversation analysis function 758.

Note that the emotion distribution information G134 shown in FIG. 4 is just an example and may be changed as desired. For example, the emotion distribution information G134 is not limited to all of the patient summary information G11, treatment selection flow information G12, and conversation flow information G13, but may include some of this information, or selected information. For example, when treatment selection flow information G12 is selected, the information generation function 760 generates emotion distribution information G134 that includes the treatment selection flow information G12. In other words, when treatment selection flow information G12 is selected, the information generation function 760 generates emotion distribution information G134 that does not include the patient summary information G11 or the conversation flow information G13. In this case, the conversation analysis function 758 does not perform analysis related to the conversation flow information G13. This allows the information generation function 760 to generate emotion distribution information G134 more quickly.

Next, we will explain the various processes performed by the clinical support system 1.

Figure 5 is a flowchart showing an example of display processing executed by the clinical support device 70 according to the first embodiment. The display processing is processing for displaying characteristics of conversations during medical examinations of similar patients.

The patient designation function 751 accepts the designation of the patient to be the subject of the conversation analysis screen G1 (step S1).

The information acquisition function 752 acquires patient information such as electronic medical record information 11, radiology report information 21, image information 31, clinical test result information 41, clinical test report information 42, and personal health information 61 (step S2).

The information acquisition function 752 acquires the medical interview information 12 (step S3).

The information acquisition function 752 acquires the conversation information 51 (step S4).

The characteristic analysis function 754 generates patient characteristic information indicating biological, psychological, social, and other aspects of the patient through analytical processing using the patient information, medical interview information 12, and conversation information 51 (step S5).

The similarity calculation function 755 calculates the similarity between the patient characteristic information generated by the characteristic analysis function 754 and each of the multiple patient characteristic information contained in the reference conversation information 742 (step S6).

The similar patient selection function 756 selects similar patients based on the similarity calculated by the similarity calculation function 755 (step S7).

The conversation information acquisition function 757 acquires conversation information 51 of similar patients from the reference conversation information 742 (step S8).

The conversation analysis function 758 analyzes the conversation information 51 of similar patients acquired by the conversation information acquisition function 757 (step S9).

The standard treatment acquisition function 759 acquires treatment policy information indicating the standard treatment according to the injury or illness of the similar patient selected by the similar patient selection function 756 from the standard treatment information 743 (step S10).

The information generation function 760 generates a conversation analysis screen G1 based on the analysis results obtained by the conversation analysis function 758 and the standard of care acquired by the standard of care acquisition function 759 (step S11). The information generation function 760 transmits the generated conversation analysis screen G1 to the display device 80 (step S12). As a result, the display device 80 displays the conversation analysis screen G1.

The clinical support device 70 then ends the display process.

As described above, the clinical support system 1 according to the first embodiment acquires patient information, medical interview information 12, and conversation information 51. Furthermore, the clinical support system 1 generates patient characteristic information indicating psychological and social aspects of the patient through an analysis process using the patient information, medical interview information 12, and conversation information 51. The clinical support system 1 also acquires the extracted patient characteristic information and conversation information 51 of similar patients whose patient characteristic information is similar. Furthermore, the clinical support system 1 analyzes the characteristics of conversations between similar patients and medical professionals based on the acquired conversation information 51. The clinical support system 1 then generates information showing a conversation analysis screen G1 based on the analysis results. This allows medical professionals to refer to conversations between similar patients and other medical professionals. Therefore, the clinical support system 1 can support communication during medical treatment.

Furthermore, medical professionals can consider treatment plans that take into account the biological, psychological, and social aspects of the patient. As such, the clinical support system 1 can promote holistic, patient-centered medical care.

(Variation 1)
In the clinical support device 70 according to the first embodiment, the reference conversation information 742 includes the patient characteristic information of similar patients. That is, the patient characteristic information of similar patients is stored in the storage circuitry 740 of the clinical support device 70. However, the patient characteristic information of similar patients may be stored in a storage circuitry of another device other than the storage circuitry 740 of the clinical support device 70. Alternatively, the clinical support device 70 may generate the patient characteristic information of similar patients each time it compares the patient characteristic information of a patient specified by the patient specification function 751 with that of a patient specified by the patient specification function 751.

Second Embodiment
The clinical support device 70a according to the second embodiment further includes a condition input function 761 that accepts acquisition conditions for acquiring the conversation information 51. Fig. 6 is a block diagram showing an example of the configuration of the clinical support device 70a according to the second embodiment. The processing circuitry 750a of the clinical support device 70a further includes the condition input function 761.

The condition input function 761 accepts acquisition conditions for acquiring conversation information 51. The condition input function 761 is an example of an input unit. More specifically, the condition input function 761 accepts acquisition conditions for acquiring conversation information 51 from similar patients selected by the similar patient selection function 756.

For example, the condition input function 761 accepts input specifying a medical professional as an acquisition condition. That is, the condition input function 761 accepts input of medical professional identification information for identifying medical professionals who have spoken with similar patients as an acquisition condition. The medical professional identification information may be the name of the medical professional or code information for identifying the medical professional. The medical professional identification information may also be the number of years of experience of the medical professional, the medical department to which the medical professional belongs, the classification of the medical professional such as specialist, or other conditions.

The conversation information acquisition function 757 acquires conversation information 51 of similar patients based on the similarity between the patient characteristic information generated by the characteristic analysis function 754 and the patient characteristic information of similar patients, and the acquisition conditions. More specifically, the similar patient selection function 756 selects similar patients based on the similarity between the patient characteristic information generated by the characteristic analysis function 754 and the patient characteristic information of similar patients. The conversation information acquisition function 757 acquires conversation information 51 of similar patients selected by the similar patient selection function 756, which satisfies the acquisition conditions.

For example, if the acquisition condition is medical professional identification information, the conversation information acquisition function 757 acquires conversation information 51 between the specified medical professional and the similar patient from the conversation information 51 of the similar patient. In other words, the conversation information acquisition function 757 acquires conversation information 51 that is associated with the patient characteristic information of the similar patient and that is associated with the medical professional identification information indicated by the acquisition condition from the reference conversation information 742.

Next, we will explain the various processes executed by the clinical support system 1 according to the second embodiment.

Figure 7 is a flowchart showing an example of display processing executed by the clinical support device 70a according to the second embodiment.

Steps S21 to S27 perform the same processing as steps S1 to S7 of the display processing performed by the clinical support device 70 according to the first embodiment.

The condition input function 761 accepts medical professional identification information for identifying medical professionals who have conversed with similar patients as an acquisition condition for acquiring conversation information 51 (step S28).

The conversation information acquisition function 757 acquires conversation information 51 of similar patients that matches the medical professional identification information from the reference conversation information 742 (step S29).

The conversation analysis function 758 analyzes the conversation information 51 acquired by the conversation information acquisition function 757 (step S30).

The standard treatment acquisition function 759 acquires treatment policy information indicating the standard treatment according to the injury or illness of the similar patient selected by the similar patient selection function 756 from the standard treatment information 743 (step S31).

The information generation function 760 generates a conversation analysis screen G1 based on the analysis results by the conversation analysis function 758 (step S32). The information generation function 760 transmits the generated conversation analysis screen G1 to the display device 80 (step S33). As a result, the display device 80 displays the conversation analysis screen G1.

The clinical support device 70a then ends the display process.

As described above, the clinical support system 1 according to the second embodiment accepts acquisition conditions for acquiring conversation information 51. For example, the clinical support system 1 accepts medical professional identification information as an acquisition condition. This allows medical professionals to view the analysis results of the communication of medical professionals that they wish to use as reference.

(Variation 1)
In the second embodiment, the condition input function 761 is described as accepting input of medical professional identification information as an acquisition condition. The condition input function 761 according to the first modification of the second embodiment accepts input of an importance level for patient characteristic information as an acquisition condition. Furthermore, the conversation information acquisition function 757 acquires conversation information 51 of similar patients based on the importance level, the patient characteristic information of the patient extracted by the characteristic analysis function 754, and the patient characteristic information of similar patients.

More specifically, the condition input function 761 accepts input to set the importance of classifications of patient characteristic information and analysis items. For example, if the importance of "psychological aspects" of patient characteristic information is set higher as an acquisition condition, the similarity calculation function 755 calculates the similarity by weighting the "psychological aspects." This increases the similarity of patient characteristic information with similar "psychological aspects." Therefore, the similar patient selection function 756 selects patients with similar "psychological aspects" as similar patients. Therefore, the conversation information acquisition function 757 acquires conversation information 51 of similar patients whose psychological aspects are similar to those of the patient specified by the patient specification function 751.

Furthermore, if the importance of "educational background," an analysis item of the social aspects of patient characteristic information, is increased as an acquisition condition, the similarity calculation function 755 calculates the similarity by weighting "educational background." This increases the similarity of patient characteristic information with similar "educational background." Therefore, the similar patient selection function 756 selects patients with similar "educational background" as similar patients. Therefore, the conversation information acquisition function 757 acquires conversation information 51 of similar patients whose "educational background" is similar to that of the patient specified by the patient specification function 751.

The condition input function 761 may also accept input to set the importance for multiple categories or analysis items.

Next, we will explain the various processes executed by the clinical support system 1 according to variant 1 of the second embodiment.

Figure 8 is a flowchart showing an example of display processing executed by the clinical support device 70a according to variant 1 of the second embodiment.

Steps S41 to S45 perform the same processing as steps S1 to S5 of the display processing performed by the clinical support device 70 according to the first embodiment.

The condition input function 761 accepts input to set the importance level as an acquisition condition for acquiring conversation information 51 (step S46).

The similarity calculation function 755 calculates the similarity between the patient characteristic information generated by the characteristic analysis function 754 and the multiple patient characteristic information contained in the reference conversation information 742, based on the importance set by the condition input function 761 (step S47).

The similar patient selection function 756 selects similar patients based on the similarity calculated by the similarity calculation function 755 (step S48).

The conversation information acquisition function 757 acquires the conversation information 51 of similar patients from the reference conversation information 742 (step S49).

The conversation analysis function 758 analyzes the conversation information 51 acquired by the conversation information acquisition function 757 (step S50).

The standard treatment acquisition function 759 acquires treatment policy information indicating the standard treatment according to the injury or illness of the similar patient selected by the similar patient selection function 756 from the standard treatment information 743 (step S51).

The information generation function 760 generates a conversation analysis screen G1 based on the analysis results by the conversation analysis function 758 (step S52). The information generation function 760 transmits the generated conversation analysis screen G1 to the display device 80 (step S53). As a result, the display device 80 displays the conversation analysis screen G1.

The clinical support device 70a then ends the display process.

As described above, the clinical support system 1 according to the first variant of the second embodiment accepts inputs that set the classification of patient characteristic information and the importance of analysis items as acquisition conditions. This allows medical professionals to specify similar patients in more detail.

(Variation 2)
In this embodiment, the information generation function 760 generates the conversation analysis screen G1. Then, it has been explained that the display device 80 displays the conversation analysis screen G1. The information generation function 760 may display the conversation analysis screen G1 on the display 730 of the clinical support device 70, or may display it on another device.

At least one of the embodiments described above can support communication during medical treatment.

While several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments may be embodied in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments may be made without departing from the spirit of the invention. These embodiments and their variations are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as set forth in the claims.

1 Clinical Support System 10 Hospital Information System (HIS)
11 Electronic medical record information 12 Interview information 20 Radiology Information Systems (RIS)
21 Image interpretation report information 30 Medical image management system (PACS: Picture Archiving and Communication Systems)
31 Image information 40 Laboratory Information System (LIS)
41 Clinical test result information 42 Clinical test report information 50 Patient terminal 51 Conversation information 60 Patient measurement terminal 61 Personal health information 70, 70a Clinical support device 80 Display device 741 Analysis target information 742 Reference conversation information 743 Standard treatment information 751 Patient designation function 752 Information acquisition function 753 Analysis target designation function 754 Feature analysis function 755 Similarity calculation function 756 Similar patient selection function 757 Conversation information acquisition function 758 Conversation analysis function 759 Standard treatment acquisition function 760 Information generation function 761 Condition input function G1 Conversation analysis screen G11 Patient summary information G12 Treatment selection flow information G13 Conversation flow information G13a First conversation flow information G13b Second conversation flow information G131 Conversation time information G132 Emotion value information G133 Process information G134 Emotion distribution information

Claims (9)

a first acquisition unit that acquires patient information related to a first patient, medical interview information indicating symptoms of the first patient, and conversation information in which a conversation of the first patient is recorded;
a first analysis unit that generates patient characteristic information indicating psychological and social aspects of the patient by performing an analysis process using the patient information, the medical interview information, and the conversation information;
a second acquisition unit that acquires the patient characteristic information of a first patient generated by the first analysis unit and the conversation information of a second patient whose patient characteristic information is similar to the patient characteristic information of the first patient;
a second analysis unit that analyzes characteristics of a conversation between the second patient and a medical professional based on the conversation information acquired by the second acquisition unit;
a display control unit that displays the analysis result by the second analysis unit;
A clinical support system equipped with: the second analysis unit analyzes the time of each process in the conversation between the second patient and the medical professional based on the conversation information;
The display control unit displays the time of each process in the conversation between the second patient and the medical professional.
The clinical support system according to claim 1 . the second analysis unit analyzes the second patient's emotions at each stage of the conversation between the second patient and the medical professional based on the conversation information;
The display control unit displays the emotion of the second patient at each stage of the conversation between the second patient and the medical professional.
The clinical support system according to claim 1 or 2. the second analysis unit analyzes keywords in each step of the conversation between the second patient and the medical professional based on the conversation information;
the display control unit displays the keywords for each step in the conversation between the second patient and the medical professional.
The clinical support system according to any one of claims 1 to 3. further comprising an input unit that accepts an acquisition condition for acquiring the conversation information;
the second acquisition unit acquires the conversation information of the second patient based on the acquisition condition.
The clinical support system according to any one of claims 1 to 4. the input unit accepts an input specifying the medical worker;
the second acquisition unit acquires the conversation information between the designated medical worker and the second patient from the conversation information of the second patient.
The clinical support system according to claim 5 . the input unit accepts an input for setting an importance level for the patient characteristic information;
the second acquisition unit acquires the conversation information of the second patient based on the importance, the patient characteristic information of the first patient generated by the first analysis unit, and the patient characteristic information of the second patient.
The clinical support system according to claim 5 or 6. The display control unit displays a standard treatment for the illness or injury of the second patient, a treatment proposed to the second patient by a medical professional, and a selected treatment.
The clinical support system according to any one of claims 1 to 7. a first acquisition unit that acquires patient information related to a first patient, medical interview information indicating symptoms of the first patient, and conversation information in which a conversation of the first patient is recorded;
a first analysis unit that generates patient characteristic information indicating psychological and social aspects of the patient by performing an analysis process using the patient information, the medical interview information, and the conversation information;
a second acquisition unit that acquires the conversation information of the second patient based on a similarity between the patient characteristic information of the first patient generated by the first analysis unit and the patient characteristic information of the second patient;
a second analysis unit that analyzes characteristics of a conversation between the second patient and a medical professional based on the conversation information acquired by the second acquisition unit;
an information generating unit that generates information indicating the analysis result by the second analyzing unit;
A clinical support device comprising: