Depression diagnosis system based on clinical phenotype data

Abstract

The invention discloses a depression diagnosis system based on clinical phenotype data, and relates to the technical field of computer-aided diagnosis. The system comprises a data collection unit, a data processing unit, a database and a clinical data accumulation unit. The health index of a user is obtained through the data collection unit according to an emotion curve and questionnaire answer monitoring; the data processing unit obtains a real-time early warning proportion according to the emotion curve and performs early warning analysis; when the real-time early warning proportion exceeds the reasonable fluctuation range in continuous T1 days or the number of days exceeding the reasonable fluctuation range in continuous T0 days exceeds T11, an early warning prompt signal is generated; when the real-time early warning proportion is greater than a preset value R1 and the health index is greater than a preset value R2, generating an early warning prompt signal; and the data processing unit transmits the early warning prompt signal to an early warning prompt end, so that the treatment condition of the patient can be known in time according to the prompt, and the diagnosis and analysis of the illness state are assisted.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer-aided diagnosis, and in particular relates to a depression diagnosis system based on clinical phenotype data. Background Art

[0002] Depression is an affective mental disorder, mainly manifested by significant and persistent low mood. Its core characteristics are reduced behavior and loss of pleasure. The main clinical manifestations include: significant and persistent low mood, slow thinking, reduced spontaneous speech, inattention, decreased learning and work ability, difficulty in interpersonal communication, etc.; often accompanied by obvious tension, anxiety and irritability, as well as self-blame, self-guilt, and negative suicidal ideas or behaviors.

[0003] For example, Chinese patent CN113693584A discloses a method for selecting depression symptom predictor variables, a computer device and a storage medium. The method of the present invention comprises the following steps: investigating depression patients and normal controls, collecting general information, Hamilton Depression Rating Scale scores and resting functional magnetic resonance imaging scan data; collecting the above information again after the depression patients receive treatment and the individual clinical symptoms are relieved; processing the resting functional magnetic resonance imaging scan data to obtain a DC brain map; based on the obtained DC brain map, analyzing the changes in brain functional activities of depression patients before and after treatment, and finding variables that can characterize the relief of depression symptoms. A computer device for implementing the above method is further provided. It can provide objective support for the prediction of early clinical efficacy of depression, reduce the social burden of the disease, and has a good application prospect.

[0004] For example, Chinese patents CN111513735A and CN115714002A disclose a method for diagnosing / early warning of depression. However, for patients with depression, there is no good auxiliary diagnosis method for problems that arise during their subsequent treatment and rehabilitation processes. Therefore, the present invention provides a depression diagnosis system based on clinical phenotype data. Summary of the invention

[0005] The purpose of the present invention is to provide a depression diagnosis system based on clinical phenotype data, which solves the existing problems by collecting information such as users' emotional curves and questionnaire answer monitoring, and combining it with analysis and processing of historical data.

[0006] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0007] The present invention is a depression diagnosis system based on clinical phenotype data, comprising:

[0008] A data collection unit, the data collection unit is used to obtain the user's health index based on the emotion curve and questionnaire answer monitoring;

[0009] A data processing unit, which is used to obtain the real-time warning ratio according to the sentiment curve and perform warning analysis;

[0010] When the real-time warning ratio exceeds the reasonable fluctuation range for T1 consecutive days, or the number of days that exceed the reasonable fluctuation range within T0 consecutive days exceeds T11, an early warning signal is generated, and T1, T0, and T11 are preset values;

[0011] When the real-time warning ratio is greater than the preset value R1, and the health index is greater than the preset value R2, a warning signal is generated;

[0012] The data processing unit transmits the early warning prompt signal to the early warning prompt terminal.

[0013] Furthermore, the data collection unit performs the following steps when acquiring the user health index:

[0014] Step S1: Obtain the user's emotional curve during hospitalization, extract the number of times the user is depressed every day and the duration of depression every day according to the emotional curve, mark the number of times the user is depressed every day as Di, and mark the duration of depression every day as Si, i=1, 2, 3, ..., n, where n is a positive integer;

[0015] Step S2: Calculate the slope of the sentiment curve for every two consecutive days, and intercept the slope at The dates corresponding to the sentiment curve within the interval are marked as normal fluctuation intervals, and the remaining dates are marked as abnormal fluctuation intervals. k1 and k2 are preset values;

[0016] Step S3: sorting the normal fluctuation intervals and abnormal fluctuation intervals in chronological order, marking the health index 1 of the user on the corresponding date as JK1 or JK2 or JK3 or JK4 or JK5 according to the distribution and slope of the normal fluctuation intervals and abnormal fluctuation intervals after sorting, and the health index 1 is obtained according to the slope of the emotion curve;

[0017] Step S4: When the health index 1 is less than X1, the number of times the user is depressed every day Di and the duration of depression every day Si are obtained. If Di or Si shows a downward trend within the continuous time X2, the health index 2 is marked as ZK1, and X1 and X2 are preset values; otherwise, the health index 2 is marked as ZK2, and the health index 2 is obtained according to the slope of the emotion curve;

[0018] Step S5: obtaining a daily health index 3 according to the user's questionnaire answer monitoring, wherein the health index 3 is obtained by the existing technology, and is obtained according to the existing depression answering system;

[0019] Step S6: Calculate the user health index Z;

[0020] ;

[0021] and , These are preset values.

[0022] Furthermore, the emotional curve adopts a line graph, the horizontal axis of the emotional curve is the number of low moods per day, and the vertical axis is the duration of low moods per day (if the hospitalization time is short, it can be collected multiple times a day, the horizontal axis is the number of low moods in a fixed time period, and the vertical axis is the duration of low moods in a fixed time period).

[0023] Furthermore, the health index 1 is obtained in the following manner:

[0024] If there is no abnormal fluctuation interval before or after the normal fluctuation interval, the user's health index 1 is marked as JK1, and the health index 1 JK1 is the average of the maximum slope and the minimum slope of the normal fluctuation interval of the emotion curve;

[0025] If there is an abnormal fluctuation interval before the normal fluctuation interval, and the slope of the abnormal fluctuation interval is greater than k2, the user's health index 1 is marked as JK2, and the health index 1 JK2 is the average of the maximum slope of the abnormal fluctuation interval and the maximum slope of the normal fluctuation interval;

[0026] If there is an abnormal fluctuation interval after the normal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK3, and the health index 1 JK3 is the average of the maximum slope and the minimum slope of the emotion curve of the abnormal fluctuation interval;

[0027] If there is only an abnormal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK4, and the health index 1 JK4 is the median of the slope of the emotion curve;

[0028] Otherwise, the user's health index 1 is marked as JK5, and the health index 1 JK5 is X3 times the maximum value of the slope of the emotion curve, where X3 is a preset value.

[0029] Furthermore, the health index 2 is obtained in the following manner:

[0030] Health Index 2 ZK1: Get the number of times of depression and the duration of depression in the last two days, and mark them as the number of times of depression DJ1, DJ2, and the duration of depression SJ1, SJ2 in chronological order;

[0031] Health index 2 ZK1=(SJ1-SJ2) / (DJ1-DJ2);

[0032] Health Index 2 ZK2: Take a maximum value and a minimum value from the number of times the user is depressed, and mark them as Dmax and Dmin respectively; take a maximum value and a minimum value from the duration of the user's depression, and mark them as Smax and Smin respectively;

[0033] Health index 2 ZK2= (Smax-Smin) / (Dmax-Dmin).

[0034] Further, the data processing unit performs the following steps when obtaining the warning ratio according to the sentiment curve:

[0035] Intercept the curve between the line segment y=Y1 and the line segment y=Y2 in the emotion curve and mark it as the warning line. Y1 and Y2 are preset values, and Y1<Y2;

[0036] Get the total length of the warning line and mark it as Ly;

[0037] Get the total length Lz of the emotion curve;

[0038] Warning ratio = Ly / Lz.

[0039] Furthermore, the reasonable fluctuation range is set by an administrator, and the real-time warning ratio is obtained once a day.

[0040] Furthermore, it also includes a clinical data accumulation unit; the clinical data accumulation unit is used to accumulate clinical data and perform clinical prospect analysis, and the method of clinical prospect analysis is:

[0041] The data collection unit transmits the emotional curve and the corresponding hospitalization duration to the database;

[0042] The clinical data accumulation unit obtains the emotion curves corresponding to the patients admitted to the hospital in the last TJ days from the database, each emotion curve includes several warning proportions, and TJ is a preset value;

[0043] Calculate the average warning ratio corresponding to each sentiment curve and mark it as the average warning ratio;

[0044] The average warning proportion is selected through the data selection algorithm, and after removing the abnormal data, the remaining average warning proportions are marked as TPj, j = 1, 2, 3, ..., x, and x is a positive integer;

[0045] Calculate the mean of TPj and mark it as the overall proportion of coordination TZq;

[0046] Obtain the number of sentiment curves whose average warning ratio exceeds TZq from all sentiment curves, and mark them as the number of exceeded sentiment curves. The remaining number of sentiment curves is marked as the number of ordinary exceeded sentiment curves.

[0047] When the number of surpassing emotional curves is greater than the number of common surpassing emotional curves, the maximum hospitalization duration and the average of the hospitalization duration in the emotional curve are obtained, and the data range formed by the maximum hospitalization duration and the average of the hospitalization duration is marked as the clinical expected range;

[0048] When the number of transcending emotional curves is ≤ the number of ordinary transcending emotional curves, the average of the hospitalization durations corresponding to the emotional curves is obtained, and the median of all hospitalization durations greater than the average hospitalization duration is obtained, and the data range formed by the average and median hospitalization durations is marked as the clinical expected range;

[0049] The clinical data accumulation unit transmits the clinical expectation range to the data processing unit, and the data processing unit receives the clinical expectation range transmitted by the clinical data accumulation unit and performs clinical expectation calculation.

[0050] Furthermore, the method of performing data selection by using the data selection algorithm includes:

[0051] Obtain the average warning proportions corresponding to all emotion curves, sort the average warning proportions in descending order according to the length of hospitalization, and mark the corresponding average warning proportions as PPj, where j=1, 2, 3, ..., m, and m is a positive integer;

[0052] Calculate the mean of Pj and mark it as the overall proportion Zq;

[0053] Calculate the average hospitalization time corresponding to all emotional curves, and mark it as average hospitalization time RT;

[0054] If Zq≥Q1, and RT≥Q2, then the emotion curves corresponding to the largest average warning proportion and the first average warning proportion are eliminated, and then the overall proportion and average hospitalization time corresponding to the remaining emotion curves are calculated; Q1 and Q2 are preset values;

[0055] Until the overall proportion is less than Q1 and the average length of hospital stay is less than Q2.

[0056] Furthermore, the data processing unit performs the following steps when performing clinical expectation calculation:

[0057] Obtain the real-time warning ratio corresponding to the user's emotional curve;

[0058] If the warning ratio is ≤ the overall coordination ratio TZq, the warning prompt terminal will prompt "maintain the treatment plan unchanged";

[0059] If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time is within the clinical expected range, the warning prompt terminal will prompt "prepare to change the treatment plan";

[0060] If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time exceeds the maximum value of the clinical expected range, the warning prompt terminal will prompt "change treatment plan";

[0061] If an early warning signal is generated, and the early warning ratio is greater than the overall coordination ratio TZq, the early warning prompt terminal will prompt "change treatment plan";

[0062] Otherwise, the early warning prompt terminal prompts "clinical observation".

[0063] The present invention has the following beneficial effects:

[0064] The present invention obtains the user's health index through the data collection unit according to the emotional curve and questionnaire answer monitoring; the data processing unit obtains the real-time warning ratio according to the emotional curve and performs warning analysis; when the real-time warning ratio exceeds the reasonable fluctuation range for T1 consecutive days, or the number of days that exceeds the reasonable fluctuation range within T0 consecutive days exceeds T11, a warning prompt signal is generated; when the real-time warning ratio is greater than the preset value R1, and the health index is greater than the preset value R2, a warning prompt signal is generated; the data processing unit transmits the warning prompt signal to the warning prompt terminal, which can timely understand the patient's treatment situation according to the prompt, and assist in the diagnosis and analysis of the disease.

[0065] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0067] Figure 1 It is a schematic diagram of the structure of the depression diagnosis system based on clinical phenotype data of the present invention;

[0068] Figure 2 Schematic diagram of the emotion curve of the present invention. DETAILED DESCRIPTION

[0069] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0070] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0071] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0072] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0073] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0074] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0075] See also Figure 1 As shown, the present invention is a depression diagnosis system based on clinical phenotype data, including: a data collection unit is connected to a data processing unit in communication, the data processing unit is connected to a database in communication, the data collection unit is connected to a database in communication, a clinical data accumulation unit is connected to a database in communication, and the data processing unit is connected to an early warning prompt terminal in communication; specifically:

[0076] A data collection unit, the data collection unit is used to obtain the user's health index based on the emotional curve and questionnaire answer monitoring; the questionnaire answer monitoring adopts the national standard depression test questions - SDS depression self-test or other versions of depression self-assessment quantitative scale;

[0077] A data processing unit, which is used to obtain the real-time warning ratio according to the sentiment curve and perform warning analysis;

[0078] When the real-time warning ratio exceeds the reasonable fluctuation range for T1 consecutive days, or the number of days that exceed the reasonable fluctuation range within T0 consecutive days exceeds T11, an early warning signal is generated, and T1, T0, and T11 are preset values;

[0079] When the real-time warning ratio is greater than the preset value R1, and the health index is greater than the preset value R2, a warning signal is generated;

[0080] The data processing unit transmits the early warning prompt signal to the early warning prompt terminal.

[0081] As an embodiment provided by the present invention, preferably, depression is automatically identified through a recognition model, and the recognition model is specifically constructed as follows:

[0082] First, sample data is constructed. The sample data includes several pictures of people in low mood. The pictures of people in low mood are obtained by manual judgment and input, and are used as sample data.

[0083] The sample data is used as a convolutional neural network (CNN) model to train the model. Here, the convolutional neural network (CNN) model is used to extract features from pictures of people with low moods.

[0084] After the training is completed, some test data are collected. The test data are several pictures that have been labeled as depressed and not depressed. They are used as test data and input into the convolutional neural network CNN model to verify the success rate of the model. When the success rate exceeds the set ratio B1, the model is marked as a recognition model. Otherwise, sample data and test data are collected again to retrain and test the model until the success rate exceeds the set ratio B1.

[0085] As an embodiment provided by the present invention, preferably, depression refers to abnormal depression of the patient's mood and mood. The patient is sad, has low voice, significantly reduced speech and movements, feels bad about himself, often feels inferior, blames himself, and feels guilty. In severe cases, he has a clear sense of guilt and may even have self-harm or suicidal thoughts or behaviors. The user can be identified as depressed by the recognition model when and where according to the needs, or the start and end time of depression can be input by the user himself / caregiver / doctor / nurse, etc.

[0086] As an embodiment provided by the present invention, preferably, the data collection unit performs the following steps when acquiring the user health index:

[0087] Step S1: Obtain the user's emotional curve during hospitalization, extract the number of times the user is depressed every day and the duration of depression every day according to the emotional curve, mark the number of times the user is depressed every day as Di, and mark the duration of depression every day as Si, i=1, 2, 3, ..., n, where n is a positive integer;

[0088] Step S2: Calculate the slope of the sentiment curve for every two consecutive days, and intercept the slope at The dates corresponding to the sentiment curve within the interval are marked as normal fluctuation intervals, and the remaining dates are marked as abnormal fluctuation intervals. k1 and k2 are preset values;

[0089] Step S3: sorting the normal fluctuation intervals and abnormal fluctuation intervals in chronological order, marking the health index 1 of the user on the corresponding date as JK1 or JK2 or JK3 or JK4 or JK5 according to the distribution and slope of the normal fluctuation intervals and abnormal fluctuation intervals after sorting, and the health index 1 is obtained according to the slope of the emotion curve;

[0090] Step S4: When the health index 1 is less than X1, the number of times the user is depressed every day Di and the duration of depression every day Si are obtained. If Di or Si shows a downward trend within the continuous time X2, the health index 2 is marked as ZK1, and X1 and X2 are preset values; otherwise, the health index 2 is marked as ZK2, and the health index 2 is obtained according to the slope of the emotion curve;

[0091] Step S5: obtaining a daily health index 3 according to the user's questionnaire answer monitoring, wherein the health index 3 is obtained by the existing technology, and is obtained according to the existing depression answering system;

[0092] Step S6: Calculate the user health index Z;

[0093] ;

[0094] and , These are preset values.

[0095] As an embodiment provided by the present invention, preferably, Figure 2 As shown, the emotional curve adopts a line graph, the horizontal axis of the emotional curve is the number of low moods per day, and the vertical axis is the duration of low moods per day (if the hospitalization time is short, it can be collected multiple times a day, the horizontal axis is the number of low moods in a fixed time period, and the vertical axis is the duration of low moods in a fixed time period).

[0096] As an embodiment provided by the present invention, preferably, the health index 1 is obtained in the following manner:

[0097] If there is no abnormal fluctuation interval before or after the normal fluctuation interval, the user's health index 1 is marked as JK1, and the health index 1 JK1 is the average of the maximum slope and the minimum slope of the normal fluctuation interval of the emotion curve;

[0098] If there is an abnormal fluctuation interval before the normal fluctuation interval, and the slope of the abnormal fluctuation interval is greater than k2, the user's health index 1 is marked as JK2, and the health index 1 JK2 is the average of the maximum slope of the abnormal fluctuation interval and the maximum slope of the normal fluctuation interval;

[0099] If there is an abnormal fluctuation interval after the normal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK3, and the health index 1 JK3 is the average of the maximum slope and the minimum slope of the emotion curve of the abnormal fluctuation interval;

[0100] If there is only an abnormal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK4, and the health index 1 JK4 is the median of the slope of the emotion curve;

[0101] Otherwise, the user's health index 1 is marked as JK5, and the health index 1 JK5 is X3 times the maximum value of the slope of the emotion curve, where X3 is a preset value.

[0102] As an embodiment provided by the present invention, preferably, the health index 2 is obtained in the following manner:

[0103] Health Index 2 ZK1: Get the number of times of depression and the duration of depression in the last two days, and mark them as the number of times of depression DJ1, DJ2, and the duration of depression SJ1, SJ2 in chronological order;

[0104] Health index 2 ZK1=(SJ1-SJ2) / (DJ1-DJ2);

[0105] Health Index 2 ZK2: Take a maximum value and a minimum value from the number of times the user is depressed, and mark them as Dmax and Dmin respectively; take a maximum value and a minimum value from the duration of the user's depression, and mark them as Smax and Smin respectively;

[0106] Health index 2 ZK2= (Smax-Smin) / (Dmax-Dmin).

[0107] As an embodiment provided by the present invention, preferably, the data processing unit performs the following steps when obtaining the warning proportion according to the emotion curve:

[0108] Intercept the curve between the line segment y=Y1 and the line segment y=Y2 in the emotion curve and mark it as the warning line. Y1 and Y2 are preset values, and Y1<Y2;

[0109] Get the total length of the warning line and mark it as Ly;

[0110] Get the total length Lz of the emotion curve;

[0111] Warning ratio = Ly / Lz.

[0112] As an embodiment provided by the present invention, preferably, the reasonable fluctuation range is set by an administrator, and the real-time warning ratio is obtained once a day.

[0113] As an embodiment provided by the present invention, preferably, it also includes a clinical data accumulation unit; the clinical data accumulation unit is used to accumulate clinical data and perform clinical expectation analysis, and the method of clinical expectation analysis is:

[0114] The data collection unit transmits the emotional curve and the corresponding hospitalization duration to the database;

[0115] The clinical data accumulation unit obtains the emotion curves corresponding to the patients admitted to the hospital in the last TJ days from the database, each emotion curve includes several warning proportions, and TJ is a preset value;

[0116] Calculate the average warning ratio corresponding to each sentiment curve and mark it as the average warning ratio;

[0117] The average warning proportion is selected through the data selection algorithm, and after removing the abnormal data, the remaining average warning proportions are marked as TPj, j = 1, 2, 3, ..., x, and x is a positive integer;

[0118] Calculate the mean of TPj and mark it as the overall proportion of coordination TZq;

[0119] Obtain the number of sentiment curves whose average warning ratio exceeds TZq from all sentiment curves, and mark them as the number of exceeded sentiment curves. The remaining number of sentiment curves is marked as the number of ordinary exceeded sentiment curves.

[0120] When the number of surpassing emotional curves is greater than the number of common surpassing emotional curves, the maximum hospitalization duration and the average of the hospitalization duration in the emotional curve are obtained, and the data range formed by the maximum hospitalization duration and the average of the hospitalization duration is marked as the clinical expected range;

[0121] When the number of transcending emotional curves is ≤ the number of ordinary transcending emotional curves, the average of the hospitalization durations corresponding to the emotional curves is obtained, and the median of all hospitalization durations greater than the average hospitalization duration is obtained, and the data range formed by the average and median hospitalization durations is marked as the clinical expected range;

[0122] The clinical data accumulation unit transmits the clinical expectation range to the data processing unit, and the data processing unit receives the clinical expectation range transmitted by the clinical data accumulation unit and performs clinical expectation calculation.

[0123] As an embodiment provided by the present invention, preferably, the method for performing data selection by using the data selection algorithm includes:

[0124] Obtain the average warning proportions corresponding to all emotion curves, sort the average warning proportions in descending order according to the length of hospitalization, and mark the corresponding average warning proportions as PPj, where j=1, 2, 3, ..., m, and m is a positive integer;

[0125] Calculate the mean of Pj and mark it as the overall proportion Zq;

[0126] Calculate the average hospitalization time corresponding to all emotional curves, and mark it as average hospitalization time RT;

[0127] If Zq≥Q1, and RT≥Q2, then the emotion curves corresponding to the largest average warning proportion and the first average warning proportion are eliminated, and then the overall proportion and average hospitalization time corresponding to the remaining emotion curves are calculated; Q1 and Q2 are preset values;

[0128] Until the overall proportion is less than Q1 and the average length of hospital stay is less than Q2.

[0129] As an embodiment provided by the present invention, preferably, the data processing unit performs the following steps when performing clinical expectation calculation:

[0130] Obtain the real-time warning ratio corresponding to the user's emotional curve;

[0131] If the warning ratio is ≤ the overall coordination ratio TZq, the warning prompt terminal will prompt "maintain the treatment plan unchanged";

[0132] If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time is within the clinical expected range, the warning prompt terminal will prompt "prepare to change the treatment plan";

[0133] If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time exceeds the maximum value of the clinical expected range, the warning prompt terminal will prompt "change treatment plan";

[0134] If an early warning signal is generated, and the early warning ratio is greater than the overall coordination ratio TZq, the early warning prompt terminal will prompt "change treatment plan";

[0135] Otherwise, the early warning prompt terminal prompts "clinical observation".

[0136] A depression diagnosis system based on clinical phenotype data obtains the user's health index through a data collection unit according to the emotional curve and questionnaire answer monitoring; a data processing unit obtains the real-time warning ratio according to the emotional curve and performs warning analysis; when the real-time warning ratio exceeds the reasonable fluctuation range for T1 consecutive days, or the number of days that exceed the reasonable fluctuation range within T0 consecutive days exceeds T11, a warning prompt signal is generated; when the real-time warning ratio is greater than a preset value R1, and the health index is greater than a preset value R2, a warning prompt signal is generated; the data processing unit transmits the warning prompt signal to the warning prompt terminal.

[0137] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0138] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A depression diagnosis system based on clinical phenotype data, characterized in that: include: A data collection unit, the data collection unit is used to obtain the user's health index based on the emotion curve and questionnaire answer monitoring; A data processing unit, which is used to obtain the real-time warning ratio according to the sentiment curve and perform warning analysis; When the real-time warning ratio exceeds the reasonable fluctuation range for T1 consecutive days, or the number of days that exceed the reasonable fluctuation range within T0 consecutive days exceeds T11, an early warning signal is generated, and T1, T0, and T11 are preset values; When the real-time warning ratio is greater than the preset value R1, and the health index is greater than the preset value R2, a warning signal is generated; The data processing unit transmits the early warning prompt signal to the early warning prompt terminal.

2. The depression diagnosis system based on clinical phenotype data according to claim 1, characterized in that: The data collection unit performs the following steps when acquiring the user health index: Step S1: Obtain the user's emotional curve during hospitalization, extract the number of times the user is depressed every day and the duration of depression every day according to the emotional curve, mark the number of times the user is depressed every day as Di, and mark the duration of depression every day as Si, i=1, 2, 3, ..., n, where n is a positive integer; Step S2: Calculate the slope of the sentiment curve for every two consecutive days, and intercept the slope at The dates corresponding to the sentiment curve within the interval are marked as normal fluctuation intervals, and the remaining dates are marked as abnormal fluctuation intervals. k1 and k2 are preset values; Step S3: sorting the normal fluctuation intervals and abnormal fluctuation intervals in chronological order, marking the health index 1 of the user on the corresponding date as JK1 or JK2 or JK3 or JK4 or JK5 according to the distribution and slope of the normal fluctuation intervals and abnormal fluctuation intervals after sorting, and the health index 1 is obtained according to the slope of the emotion curve; Step S4: When the health index 1 is less than X1, the number of times the user is depressed every day Di and the duration of depression every day Si are obtained. If Di or Si shows a downward trend within the continuous time X2, the health index 2 is marked as ZK1, and X1 and X2 are preset values; otherwise, the health index 2 is marked as ZK2, and the health index 2 is obtained according to the slope of the emotion curve; Step S5: Obtain daily health index three based on the user's questionnaire answers; Step S6: Calculate the user health index Z; ;and , These are preset values.

3. The depression diagnosis system based on clinical phenotype data according to claim 1 or 2, characterized in that: The emotion curve adopts a line graph, the horizontal axis of the emotion curve is the number of times of low mood every day, and the vertical axis is the duration of low mood every day.

4. The depression diagnosis system based on clinical phenotype data according to claim 2, characterized in that: The method for obtaining the health index 1 is as follows: If there is no abnormal fluctuation interval before or after the normal fluctuation interval, the user's health index 1 is marked as JK1, and the health index 1 JK1 is the average of the maximum slope and the minimum slope of the normal fluctuation interval of the emotion curve; If there is an abnormal fluctuation interval before the normal fluctuation interval, and the slope of the abnormal fluctuation interval is greater than k2, the user's health index 1 is marked as JK2, and the health index 1 JK2 is the average of the maximum slope of the abnormal fluctuation interval and the maximum slope of the normal fluctuation interval; If there is an abnormal fluctuation interval after the normal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK3, and the health index 1 JK3 is the average of the maximum slope and the minimum slope of the emotion curve of the abnormal fluctuation interval; If there is only an abnormal fluctuation interval, and the slope of the abnormal fluctuation interval is less than k1, the user's health index 1 is marked as JK4, and the health index 1 JK4 is the median of the slope of the emotion curve; Otherwise, the user's health index 1 is marked as JK5, and the health index 1 JK5 is X3 times the maximum value of the slope of the emotion curve, where X3 is a preset value.

5. The depression diagnosis system based on clinical phenotype data according to claim 2, characterized in that: The method for obtaining the health index 2 is as follows: Health Index 2 ZK1: Get the number of times of depression and the duration of depression in the last two days, and mark them as the number of times of depression DJ1, DJ2, and the duration of depression SJ1, SJ2 in chronological order; Health index 2 ZK1=(SJ1-SJ2) / (DJ1-DJ2); Health Index 2 ZK2: Take a maximum value and a minimum value from the number of times the user is depressed, and mark them as Dmax and Dmin respectively; take a maximum value and a minimum value from the duration of the user's depression, and mark them as Smax and Smin respectively; Health index 2 ZK2= (Smax-Smin) / (Dmax-Dmin).

6. The depression diagnosis system based on clinical phenotype data according to claim 1, characterized in that: The data processing unit performs the following steps when obtaining the warning ratio according to the emotion curve: Intercept the curve between the line segment y=Y1 and the line segment y=Y2 in the emotion curve and mark it as the warning line. Y1 and Y2 are preset values, and Y1<Y2; Get the total length of the warning line and mark it as Ly; Get the total length Lz of the emotion curve; Warning ratio = Ly / Lz.

7. The depression diagnosis system based on clinical phenotype data according to claim 1, characterized in that: The reasonable fluctuation range is set by the administrator, and the real-time warning ratio is obtained once a day.

8. The depression diagnosis system based on clinical phenotype data according to claim 1, characterized in that: It also includes a clinical data accumulation unit; the clinical data accumulation unit is used to accumulate clinical data and perform clinical prospective analysis, and the method of clinical prospective analysis is: The data collection unit transmits the emotional curve and the corresponding hospitalization duration to the database; The clinical data accumulation unit obtains the emotion curves corresponding to the patients admitted to the hospital in the last TJ days from the database, each emotion curve includes several warning proportions, and TJ is a preset value; Calculate the average warning ratio corresponding to each sentiment curve and mark it as the average warning ratio; The average warning proportion is selected through the data selection algorithm, and after removing the abnormal data, the remaining average warning proportions are marked as TPj, j = 1, 2, 3, ..., x, and x is a positive integer; Calculate the mean of TPj and mark it as the overall proportion of coordination TZq; Obtain the number of sentiment curves whose average warning ratio exceeds TZq from all sentiment curves, and mark them as the number of exceeded sentiment curves. The remaining number of sentiment curves is marked as the number of ordinary exceeded sentiment curves. When the number of surpassing emotional curves is greater than the number of common surpassing emotional curves, the maximum hospitalization duration and the average of the hospitalization duration in the emotional curve are obtained, and the data range formed by the maximum hospitalization duration and the average of the hospitalization duration is marked as the clinical expected range; When the number of transcending emotional curves is ≤ the number of ordinary transcending emotional curves, the average of the hospitalization durations corresponding to the emotional curves is obtained, and the median of all hospitalization durations greater than the average hospitalization duration is obtained, and the data range formed by the average and median hospitalization durations is marked as the clinical expected range; The clinical data accumulation unit transmits the clinical expectation range to the data processing unit, and the data processing unit receives the clinical expectation range transmitted by the clinical data accumulation unit and performs clinical expectation calculation.

9. The depression diagnosis system based on clinical phenotype data according to claim 8, characterized in that: The method for performing data selection by using the data selection algorithm includes: Obtain the average warning proportions corresponding to all emotion curves, sort the average warning proportions in descending order according to the length of hospitalization, and mark the corresponding average warning proportions as PPj, where j=1, 2, 3, ..., m, and m is a positive integer; Calculate the mean of Pj and mark it as the overall proportion Zq; Calculate the average hospitalization time corresponding to all emotional curves, and mark it as average hospitalization time RT; If Zq≥Q1, and RT≥Q2, then the emotion curves corresponding to the largest average warning proportion and the first average warning proportion are eliminated, and then the overall proportion and average hospitalization time corresponding to the remaining emotion curves are calculated; Q1 and Q2 are preset values; Until the overall proportion is less than Q1 and the average length of hospital stay is less than Q2.

10. The depression diagnosis system based on clinical phenotype data according to claim 8, characterized in that: The data processing unit performs the following steps when performing clinical expectation calculation: Obtain the real-time warning ratio corresponding to the user's emotional curve; If the warning ratio is less than or equal to the overall coordination ratio TZq, the warning prompt terminal will prompt "keep the treatment plan unchanged"; If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time is within the clinical expected range, the warning prompt terminal will prompt "prepare to change the treatment plan"; If the warning ratio is greater than the overall coordination ratio TZq, and the user's hospitalization time exceeds the maximum value of the clinical expected range, the warning prompt terminal will prompt "change treatment plan"; If an early warning signal is generated, and the early warning ratio is greater than the overall coordination ratio TZq, the early warning prompt terminal will prompt "change treatment plan"; Otherwise, the early warning prompt terminal prompts "clinical observation".

Images