JP5089413B2 - Multiple disease simulation system - Google Patents

Description

  The present invention relates to a simulation system for simulating a disease state transition by treatment when a plurality of diseases are simultaneously developing.

  Conventionally, models based on epidemiological studies centered on specific diseases such as heart disease and diabetes, patient basic information (age, gender, height, weight, etc.), laboratory data, diagnostic results, medication data, etc. Proposals have been made to simulate how symptoms change with treatment. For example, Patent Document 1 uses the equation for predicting the probability of occurrence of cardiovascular disease, which is the result of an epidemiological study, and describes the risk of developing a cardiovascular disease such as myocardial infarction in the future and the therapeutic goal, A technique for simulating by inputting test values and past history data relating to an inspected person is disclosed.

  By the way, there are cases in which several diseases occur at the same time in some patients. In such cases, treatment of a certain disease improves symptoms for that disease, but another disease that is simultaneously suffering On the other hand, there are many cases where the symptoms worsen.

However, until now, no proposal has been made for simulating how each disease changes by treatment, targeting a case in which a plurality of diseases occur simultaneously. In such a case, there has been no proposal of a system capable of simulating the transition of each disease according to a subsequent treatment plan based on the situation after a certain period from the start of treatment.
Japanese Patent Laid-Open No. 2002-24401

  The present invention is a multi-disease simulation system capable of visualizing and simulating how each disease changes due to treatment when a plurality of diseases coexist due to the circumstances as described above. The issue is to provide.

  Further, the present invention is a case of multiple diseases that can simulate the transition of each disease according to a subsequent treatment plan based on the situation after the lapse of a certain period from the start of treatment in a case where a plurality of diseases coexist. Another object is to provide a simulation system.

In order to solve the above-described problems, the present invention firstly includes a server computer and a plurality of client computers connected via a network, and a plurality of database systems operating on the server computer or on the network. A disease simulation system,
The database system has an experimental medical database that stores various laboratory data in association with the type of disease and the degree of disease symptoms;
The client computer
A data input unit that accepts input of clinical disease data, case names of multiple diseases for a specific patient
Equipped with a control program for simulating multiple diseases, and numerically defined improvement factors and deterioration factors for diagnosis results for a certain disease, and the health status corresponding to the diagnosis results for each case step by step A main memory unit that holds the classified grades,
A storage unit that captures and stores necessary clinical test data from a database system acquired via a server computer,
A processing unit for performing calculations necessary for simulation of multiple diseases, and
It has a display unit that visualizes and displays the simulation results of multiple diseases,
The processing unit
A weighting factor deriving unit for deriving the weighting factor of the clinical laboratory data for each case based on the acquired clinical laboratory data,
An improvement factor calculation unit that calculates the total strength of the improvement factors by multiplying the weighting factor for each of the specific patient clinical test data corresponding to the improvement factors,
A deterioration factor calculation unit corresponding to the deterioration factor, which calculates the total strength of the deterioration factors by multiplying the weighting factor for each of the specific patient clinical test data, and
The balance between the total strength of the improvement factors and the total strength of the deterioration factors is used as the diagnosis result, the time series change of each improvement factor and each deterioration factor is calculated, and the value of the diagnosis result is obtained in time series, and the treatment There is provided a simulation system for a plurality of diseases, characterized in that it has a diagnosis result calculation unit for visualizing and displaying a disease state transition by the display unit.

  Second, in the first invention, the client computer accepts input of two case names, and the display unit displays one case state corresponding to the X axis and the other case state corresponding to the Y axis. Provided is a simulation system for multiple diseases, characterized by being visualized and displayed.

  Third, in the first invention, the client computer accepts input of three case names, the first case state is on the X axis, the second case state is on the Y axis, A simulation system for multiple diseases is provided in which the state of each case is visualized and displayed on a display unit in correspondence with the Z axis.

  Fourth, in the first invention, the client computer accepts input of four or more case names, and visualizes and displays the state of each case as a chart graph on the display unit. Provide a system.

  Further, fifthly, in the first to fourth inventions, the diagnosis result calculation unit of the processing unit of the client computer calculates a time series change of each deterioration factor and each improvement factor using a sigmoid function. A simulation system for multiple diseases is provided.

  Further, sixthly, in any one of the first to fifth inventions, the diagnosis result calculation unit of the processing unit of the client computer starts each treatment performed after the start of treatment and starts each deterioration factor and each Provided is a multi-disease simulation system characterized by calculating time-series changes of improvement factors.

  According to the present invention, since the above-described configuration is adopted, it is possible to visualize and simulate how each disease changes due to treatment when a plurality of diseases coexist.

  The present invention also makes it possible to simulate the transition of each disease according to the subsequent treatment plan based on the situation after the lapse of a certain period from the start of treatment in the case where a plurality of diseases coexist.

  Hereinafter, preferred embodiments of the present invention will be described.

  FIG. 1 shows a schematic configuration of a simulation system for multiple diseases according to the present embodiment, FIG. 2 shows a schematic hardware configuration of a client computer, and FIG. 3 shows a functional block diagram of a processing unit of the client computer. Show.

  The server computer 10 is connected to a plurality of client computers 20 via a network such as a LAN, WAN, or the Internet. Although the server computer 10 is connected to the database system 30, the database system 20 may be operated on the server computer 20 or may be operated on a network such as a LAN, WAN, or the Internet.

  In the multiple disease simulation system according to the embodiment of the present invention, when a plurality of diseases occur simultaneously, the genetic constitution is used to grasp the state of the disease as a whole as a result of treatment. It is impossible for the body to simulate the patient's condition taking into account and lifestyle, visualize the patient's clinical laboratory data and diagnosis results in relation to changes in the internal environment, etc., and simulate the change in the state due to treatment It is a system that can navigate treatment methods that may be completely cured.

  The server computer 10 collects experimental medical information, stores it in the database system 20, retrieves necessary data in response to a request from the client computer 20, and provides it to the client computer 20.

  As shown in FIG. 2, the client computer 20 includes a processing unit 11, an external interface unit (I / O) 12, a main storage unit 13, a data input unit 14, a storage unit 15, a display unit 16, and the like.

  The processing unit 11 controls each unit, performs various calculations necessary for the simulation based on experimental medical information sent via the host computer 10, and derives a simulation result. The external interface unit 12 exchanges data with the server computer 10. The main memory unit 13 has a control program and the like necessary for the simulation, and numerically defined and improved improvement factors and deterioration factors for a diagnosis result of a certain disease, and a health state corresponding to the diagnosis result for each case Retain grades classified in stages. The data input unit 14 includes an input unit such as a keyboard, and a person who intends to obtain a desired simulation result using the system inputs information such as data from the input unit in order to perform settings for the simulation. Accept. The storage unit 15 takes in and stores necessary clinical test data from the database system 30 acquired via the server computer 10. The display unit 16 includes a display device such as a display, and presents a necessary menu screen to the user or displays a simulation result.

  The processing unit 15 is a weighting factor deriving unit 21 for deriving a weighting factor of clinical test data for each case based on the acquired clinical test data, and the weight for each clinical test data of a specific patient corresponding to the improvement factor. The improvement factor calculation unit 22 that calculates the total strength of the improvement factors by multiplying by the coefficient, the deterioration that calculates the total strength of the deterioration factors by multiplying the weighting factor for each specific patient clinical test data corresponding to the deterioration factor The factor calculation unit 23 and the balance between the total strength of the improvement factors and the total strength of the deterioration factors are used as the diagnosis result, and the time series change between each improvement factor and each deterioration factor is calculated and the value of the diagnosis result is It has a diagnostic result calculation unit 24 that obtains serially and visualizes and displays a disease state transition due to treatment on a display unit.

  In the present embodiment, for example, a comprehensive gene medical support system (HGM-aid) developed and constructed by the present inventors can be used as the database system 30 (Japanese Patent Laid-Open No. 2005-108248).

  This comprehensive gene medical support system has the following structural features.

  (I) It has a data input unit, a data processing unit, a data storage unit, a data search unit, and a data display unit, and the data storage unit has knowledge for storing clinical data and clinical research data in clinical medicine To store the classification database, the term definition database for storing definitions of the meaning and scope of terms included in clinical data and research data, and the strength of ties between records that make up the data stored in the knowledge classification database A domain management database and a user management database, and the data processing unit generates a discretized data by collating clinical data with a reference value against a knowledge classification database and compressing the amount of information It has digitized data processing function means.

  (Ii) The data processing unit has a data-related strength processing function means for classifying the strength of the connection between records related to the region mediation database by an OR condition and an AND condition.

  (Iii) The data search unit has search function means using an OR condition and an AND condition of related information defined between related knowledge elements.

  (Iv) The data search unit has search function means for limiting the “depth” related to the search order from the knowledge element as the base point and the “weight” of the relationship between the knowledge elements as a threshold value.

  (V) In the data search unit and the data display unit, the knowledge element as the base point is arranged in the center, the primary elements are arranged concentrically, and the secondary elements are arranged linearly or radially from the primary element, The elements related to each other have search graph display means arranged radially from the branching elements.

  Next, the simulation of the system of the present embodiment will be described in detail by giving a specific example.

  In this simulation, a model is used in which a diagnosis result (D) is balanced by a factor that adversely affects the disease (deterioration factor: P) and a factor that improves the disease (improvement factor: Q). Examples of these factors and examples of diagnostic results are shown in Table 1.

Here, in the control program installed in the main storage unit 13 of the client computer 20, P [k] _m and Q [k that can be compared with the diagnosis result D [k] _m for the disease k at a certain time point m. ] Define _m . Further, the values that D [k] _m can take are classified as grades such as three grades such as a healthy state, a mild symptom state, and a severe symptom state.

In order to compare D [k] _m , P [k] _m , and Q [k] _m , the client computer 20 that performs the simulation uses the experimental medical database included in the database system 30 to use these clinical test data. And the diagnosis results classified by grade for each case, and the clinical laboratory values (E ^u _m , F ^v _m : u, v are the cause of the deterioration and the type of the improvement factor) of the deterioration factor and the improvement factor, respectively. In order to convert into discrete values corresponding to the grades of, the degree of influence of the clinical test for each disease is introduced as a weighting factor (W [k] ^u , R [k] ^v ) (Equation 1). This introduction is performed by the weight coefficient introduction unit 21 of the processing unit 15.

The meaning of the weighting factor W [k] ^u of the deterioration factor is a coefficient for normalizing a plurality of deterioration factors with respect to the disease k and replacing them with numerical values that can be compared. Similarly, the weighting factor R [k] ^v of the improvement factor is also normalized by comparing a plurality of improvement factors to a numerical value that can be compared, so that the diagnosis result D [k] _m at a certain time point _m is deteriorated by P [k] _m and the improvement factor Q [k] _m can be expressed in a balance (formula 2).

Here, a breast cancer patient is first exemplified as a reference example of the one-dimensional simulator model (k = 0). FIG. 4 is an explanatory diagram of the one-dimensional simulation model. Tumor exacerbation cause P [0] _m is improved as BCA225 (breast cancer specific antigen), CA15-3 (breast cancer related antigen), CEA (carcinoembryonic antigen) and TK (thymidine kinase activity) which are tumor markers indicating breast cancer Factor Q [0] _m is the total immunity. A diagnostic result D [0] _m is calculated by multiplying these clinical laboratory values by the respective weighting coefficients to convert them into the strength of influence that worsens and improves breast cancer (Equation 3).

It is expressed.

  Since each factor changes in time series, a changed value after a certain time T has elapsed from the start time is predicted. The time-series change of each factor is predicted, and the diagnosis result D [0] after the elapse of time T can be simulated by the balance equation in the same manner as the above equation 3.

  Next, an n-dimensional simulator model (n is an integer of 2 or more) will be described.

Among the deterioration factors P [k] ^u _m and the improvement factors Q [k] ^v _m as factors that change the medical condition, there are elements that change with each other. The correlation between these elements is also estimated in the same manner as the above-described relationship such as the diagnosis result, and is expressed as a balance equation.

There are factors that cause changes accompanying each other in the deterioration factors P [k] ^u _m and the improvement factors Q [k] ^v _m (k = 0, 1) as factors that change the symptoms. The correlation between these elements is also estimated and expressed as a balance equation. Accordingly, the balance surrounding D [k] _m can be expressed as an n-ary linear equation composed of a total of n (= a + b) elements of P [k] _m and Q [k] _m (Equation 4).

When considered as an n-dimensional balance equation, w _n indicates the state of disease k (corresponding to D [k] _m in one dimension (equation 2)), and t _kj is a deterioration factor of disease k. Similarly, s _kj indicates the weight that a certain improvement factor contributes to the improvement of the disease k. When each dimension is considered as a disease k by the n-dimensional balance equation, the health state simulation shown in Equation 2 can be modeled as a health state simulation corresponding to a state in which a plurality of diseases coexist. . When simulating using any two types of diseases, it can be expressed as a two-dimensional graph. When simulating using any three types of diseases, it can be expressed as a three-dimensional graph, and further expressed as a chart graph. Simulation using any 4 or more diseases is possible.

In the transient analysis of the simulator by the diagnosis result calculation unit 24 of the processing unit 11, the analysis is started with the state at a certain time point m as an initial state, and the influence received by treatment such as medication is given as a disturbance event (for example, the influence of medication) ) As a time-series change), and the transient changes in D [k] _m , P [k] _m , and Q [k] _m are analyzed. The overall result of the transient analysis is displayed on the display unit 16 by using an animation of a chart graph, and it is visualized which of the n types of concurrent diseases is worsening or improving. Use an understandable expression.

  As parameters of disturbance events, the slope of the rise, the magnitude of the influence, the time when the disturbance occurs and the delay time are given, and the decay rate after the disturbance reaches the peak is set as a parameter, so that the disturbance event is Express. By setting the degree of influence of the set disturbance event on the disease k and the threshold value to be affected, it is possible to simulate a state in which a plurality of diseases coexist and affect each other. In order to simulate changes due to these disturbances, disturbance events are not given as linear changes, but as continuous changes using a sigmoid function, it is possible to perform a simulation closer to the actual conversion. It becomes.

  There are various types of sigmoid functions, and favorable results could be obtained by using the following functions shown in the lower diagram of FIG. 5. However, the present invention is not limited to this, and other functions are used. Also good.

Here, A represents the upper limit value of the disturbance, B represents the gradient, C represents the delay time, and T represents the time.

  Next, a more detailed description will be given with a specific example. The example shown here is for a two-dimensional simulator model.

  A plurality of deterioration factors P and improvement factors Q are defined as factors that change the medical condition. For each factor, the influence on the X axis and the Y axis is defined as a ratio, and the time series change of each factor is obtained. The sigmoid function described above is used as a basic pattern of time series change. Here, as initial settings, parameters (start time, slope, upper limit, delay time, rate of change, etc.) of change patterns of the respective factors are set. Then, the influence and timing of a certain factor on other factors are set. For example, the diagnosis result calculation unit 24 of the processing unit 11 improves the condition (upper limit value of the deterioration factor) of the disease (deterioration factor) when a certain drug (improvement factor) reaches a certain concentration (threshold). Simulate attenuating).

  As a specific example, assume that a patient has heart failure (k = 0) and breast cancer (k = 1) at the same time.

Heart failure exacerbation factor P [0] _m is BNP (brain natriuretic peptide) and CK-MB (creatine kinase MB), and improvement factor Q [0] _m is total immunity. Further, the tumor deterioration factor P [1] _m is defined as BCA225 (breast cancer specific antigen), CA15-3 (breast cancer related antigen), CEA (carcinoembryonic antigen) and TK (thymidine kinase activity) which are tumor markers indicating breast cancer. The improvement factor Q [0] _m is _defined as total immunity. Also, the diagnostic results D [0] _m and D [1] _m are calculated by multiplying these clinical laboratory values by the respective weighting coefficients and converting them to the strength of influence that worsens and improves heart failure and breast cancer.

  First, a diagnosis result of breast cancer and heart failure at the time of m = 1 is obtained as data of a treatment start point. Therefore, the client computer 20 captures necessary clinical test data and the like from the database system 30 via the server computer 10. Then, the client computer 20 takes in clinical test data and the like corresponding to the state of the patient at the start of treatment.

As diagnosis results at the start of treatment, diagnosis results D [0] ₁ and D [1] ₁ of heart failure and breast cancer are obtained from data at the first visit. The time point at which m = 1 is set as the treatment start point.

  In this example, it is simulated how each disease (here heart failure and breast cancer) changes from treatment start to treatment, and the result is displayed on the display of the display unit 16 of the client computer 20 in the XY coordinates. Visualize and display. The X axis represents heart failure factor and the Y axis represents breast cancer factor. The origin is a healthy state free of both diseases. The time point m = 1 is the treatment start time point, m = 2, 3,... Are the time points after the treatment start specified by the input after the treatment start.

  In this example, in order to predict the state of both diseases when a certain time T has elapsed after the start of treatment, an initial setting is made at the treatment start time point m = 1. FIG. 5 shows an example of input of the default parameters and a sigmoid function representing a time series change of each element after the start of treatment.

  A detailed initial setting screen is shown in FIG. The X-axis label is heart failure and the Y-axis label is breast cancer. “Heart failure factor” and “breast cancer factor” are shown as exacerbation factors, and the influence of these factors on the X axis and Y axis is defined as “5: 1” and “1: 5”, respectively. In addition, in order to obtain the time series change of these factors, it is possible to input “upper limit”, “slope”, “delay”, “rate of change”, and “start (time from the start of treatment)” in the sigmoid function. ing. In addition, as an improvement factor, the schedule of treatment to be performed after the start can be input in time series, and like `` heart failure factor '' and `` breast cancer factor '', in order to obtain the time series change of each element, in the sigmoid function "Upper limit", "Slope", "Delay", "Change rate", and "Start (start time of each treatment)" can be input. In addition, the number of “effect target” (1 is heart failure factor, 2 is breast cancer factor, 3 is total immunity), the effect of the effect on the X-axis and Y-axis, “threshold”, and “efficacy rate” can be entered. It has become.

  FIG. 7 is a diagram showing a display screen in an initial state, where “initial state” indicates a state without both diseases. 8 and 9 show movements up to the state at the first visit, and FIG. 10 shows the state at the first visit.

  Based on the input data, the client computer 20 performs an operation in consideration of the effects of treatments related to each other in the balance equation of Equation 4, and the transition of the state of both diseases at each time point is calculated. Simulate.

  FIG. 11 shows how anti-cancer drug treatment significantly improves breast cancer factor, but has an adverse effect on heart failure factor.

  FIG. 12 shows a situation where the anticancer drug treatment is moderated and the immunotherapy is also used to suppress the deterioration of the heart failure factor, and at the same time, the immunotherapy is pulling in the direction of improvement. .

  FIG. 13 shows a situation in which significant improvement has been observed by continuing mild anticancer drug treatment and immunotherapy.

  Next, what kind of information can be obtained by the multiple disease simulation system according to the present invention will be described with reference to FIG.

  [1] The patient has familial dilated cardiomyopathy, lifestyle changes due to the care of his grandchild, a high fat diet, and breast cancer after about 12 months. At the first visit, breast cancer had spread to the thorax and was diagnosed as inoperable.

  [2] Combined administration of 3 types of anticancer drugs for 1 week according to cancer treatment protocol and treatment to increase total immunity as experimental medical treatment, breast cancer dramatically regressed and metastasis to the rib cage reduced did. At this point, there was a lack of consideration for heart disease with anticancer drugs.

  [3] Furthermore, the combined use of three types of anticancer drugs for one week and treatment to increase total immunity stopped the activity of cancerous virus, which is an indicator of the active phase of cancer, and stopped the growth of cancer. It was considered. Breast cancer improved more dramatically, but heart failure rapidly worsened due to side effects.

  [4] Heart failure was improved only by temporarily postponing the regimen of additional anticancer drugs to enhance total immunity.

  The simulation system for multiple diseases according to the present invention incorporates a model for analyzing diseases and their factors specialized in medical treatment, and this simulation technology is expected to develop in the following direction in the medical field.

  (1) This simulation system was basically created with the use of doctors in mind, but when the simulation system has been recognized, the simulation results are shown to the patient in an easy-to-understand manner. It is possible to understand what kind of illness is predisposed to illness and help to acquire the lifestyle habits necessary to avoid the illness. This is because, from the passive medical treatment that the current doctor plays a central role, the patient actively participates in the medical treatment, and the patient and the doctor are in an equal position, and what kind of treatment is advanced for the patient. It can provide a place to consider whether it is appropriate. This simulation technology can be said to be a powerful tool for reforming and establishing from treatment-centered medicine in Japan to prevention-centered medicine, and from doctor-centered medicine to medicine where patients and doctors cooperate.

  (2) By importing information that interprets descriptive data such as clinic findings by doctors and image diagnostic results into the system as one of the diagnostic parameters, it is possible to more accurately represent the characteristics of disease diagnosis, and diagnosis and treatment The simulation accuracy can be improved. In addition, by establishing a safe and secure medical network infrastructure, input / output functions such as descriptive data such as findings and image data will be brought online, and information described in electronic medical records, etc. will be captured and updated to the latest information at a remote location. Diagnosis and treatment navigation based on this becomes possible.

1 is a diagram showing a schematic configuration of a multiple disease simulation system according to an embodiment of the present invention. It is a block diagram which shows the hardware schematic structure of a client computer. It is a functional block diagram which shows the structure of the process part of a client computer. It is a figure which shows the example of a one-dimensional simulator model. It is explanatory drawing of an initial setting screen and a sigmoid function. It is a figure which shows an initial setting screen in detail. Show. It is a figure which shows the display screen of an initial state. It is a figure which shows the state in the middle to the state at the time of the first medical examination. It is a figure which shows the state in the middle to the state at the time of the first medical examination. It is a figure which shows the state at the time of the first medical examination. It is a figure which shows the state of both diseases by anticancer agent treatment. It is a figure which shows the state of both diseases by making anticancer agent treatment gentle, and using immunotherapy together. It is a figure which shows the state from which the remarkable improvement was seen in both diseases by gentle anticancer agent treatment and continuation of immunotherapy. It is a figure which shows the specific example about what kind of information is obtained by the simulation system of multiple diseases by this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Server computer 20 Client computer 30 Database system 11 Processing part 12 External interface part 13 Main memory part 14 Input part 15 Storage part 16 Display part 21 Weight coefficient introduction part 22 Improvement factor calculation part 23 Deterioration factor calculation part 24 Diagnosis result calculation part

Claims (6)

A multiple disease simulation system in which a server computer and a plurality of client computers are connected via a network, and a database system is operating on the server computer or the network,
The database system has an experimental medical database that stores various laboratory data in association with the type of disease and the degree of disease symptoms;
The client computer
A data input unit that accepts input of clinical disease data, case names of multiple diseases of a specific patient,
Equipped with a control program for simulating multiple diseases, and numerically defined improvement factors and deterioration factors for diagnosis results for a certain disease, and the health status corresponding to the diagnosis results for each case step by step A main memory unit that holds the classified grades,
A storage unit that captures and stores necessary clinical test data from a database system acquired via a server computer,
A processing unit for performing calculations necessary for simulation of multiple diseases, and
It has a display unit that visualizes and displays the simulation results of multiple diseases,
The processing unit
A weighting factor deriving unit for deriving the weighting factor of the clinical laboratory data for each case based on the acquired clinical laboratory data,
An improvement factor calculation unit that calculates the total strength of the improvement factors by multiplying the weighting factor for each of the specific patient clinical test data corresponding to the improvement factors,
A deterioration factor calculation unit corresponding to the deterioration factor, which calculates the total strength of the deterioration factors by multiplying the weighting factor for each of the specific patient clinical test data, and
The balance between the total strength of the improvement factors and the total strength of the deterioration factors is used as the diagnosis result, the time series change of each improvement factor and each deterioration factor is calculated, and the value of the diagnosis result is obtained in time series, and the treatment A simulation system for a plurality of diseases, characterized by having a diagnosis result calculation unit for visualizing and displaying a disease state transition by the display unit.   The client computer accepts input of two case names, and visualizes and displays the state of one case on the display unit in association with the X-axis and the state of the other case on the Y-axis. The multiple disease simulation system described.   The client computer accepts input of three case names, and displays the first case state on the X axis, the second case state on the Y axis, and the third case state on the Z axis. The multiple disease simulation system according to claim 1, wherein the system is visualized and displayed.   The multi-disease simulation system according to claim 1, wherein the client computer receives input of four or more case names, and displays the state of each case as a chart graph on the display unit.   5. The plural according to claim 1, wherein the diagnosis result calculation unit of the processing unit of the client computer calculates a time-series change of each deterioration factor and each improvement factor by using a sigmoid function. Disease simulation system.   6. The diagnosis result calculation unit of the processing unit of the client computer calculates a time-series change of each subsequent deterioration factor and each improvement factor with each treatment performed after the start of treatment as a start time. The simulation system for multiple diseases according to any one of the above.