Patient information automatic processing system and method

Abstract

The invention relates to the technical field of digital processing, in particular to a patient information automatic processing system and method. The method comprises the following steps: S1, acquiring first acquisition information, wherein the first acquisition information comprises historical case information; s2, according to the historical case information and the key information of the medicine to be clinically tested, screening a plurality of tested patients matched with the medicine to be clinically tested; s3, acquiring second acquisition information; s4, according to the second collection information, safety level judgment is carried out on each tested patient, safety levels judged by a second processing module include low safety, medium safety and high safety, and the clinical drug test is stopped for the tested patients judged to be low safety and the like; s5, screening the information of each tested patient in the clinical drug test stage into conventional data and key data; according to the invention, the accuracy of clinical drug tests and the safety of tested patients are ensured.

Description

Technical Field

[0001] The present invention relates to the field of digital processing technology, and in particular to a system and method for automatically processing patient information. Background Art

[0002] Clinical trials of drugs require systematic research on drugs on patients or healthy volunteers to confirm or test the clinical and pharmacological effects, adverse reactions, etc. of the drugs in order to ensure the safety and effectiveness of the drugs. Whether it is the selection and confirmation stage of patients or healthy volunteers, or the clinical trial stage of drugs, it is necessary to collect and analyze the information of the subjects to ensure the safety of the subjects and verify the effectiveness of the drugs. Chinese patent CN114187983A discloses a method and device for grouping subjects in a clinical trial project, wherein the method comprises: receiving a target request, wherein the target request is used to request grouping of subjects in a target clinical trial project; responding to the target request, obtaining pre-configured parameters related to the target clinical trial project; generating subject grouping results according to the pre-configured parameters. This patent solves the problem that the enrollment of subjects and the distribution of trial drugs during clinical trials require human participation, resulting in the inability to ensure the credibility and safety of clinical trials. It enables the trial personnel to operate the enrollment and drug distribution at the front desk. The entire process is completed within the system and is invisible to the personnel in a black box, which ensures the reliability of the data, avoids human deviation, and ensures the credibility and safety of the entire clinical trial.

[0003] However, existing clinical trial projects do not monitor patient information throughout the entire process from the preparation stage to the end of the drug clinical trial, resulting in inaccurate patient information monitoring, which in turn affects the accuracy of drug clinical trial results and patient safety.

[0004] Therefore, there is an urgent need to provide a system and method for automatically processing patient information, which can ensure the accuracy of drug clinical trial results and the safety of test patients compared with the existing technology. Summary of the invention

[0005] The present invention solves the above technical problems existing in the prior art and provides a system and method for automatically processing patient information.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows: A method for automatically processing patient information comprises the following steps: S1. Acquire first collected information, where the first collected information includes historical case information; S2. Screen multiple patients who match the drug to be tested based on historical case information and key information of the drug to be tested; S3. Obtain the second collection information; S4. According to the second collection information, judge the safety level of each subject patient. The safety levels judged by the second processing module include low safety level, medium safety level, and high safety level. For the subject patients judged to have a low safety level, stop the clinical drug trial; S5. Screen the information of each subject patient in the clinical drug trial stage into routine data and key data.

[0007] Furthermore, set J interval times in the clinical drug trial stage, set multiple set times within each interval time, and each subject patient has a consultation once at each set time to obtain the corresponding diagnosis report and BMI value. The second collection information includes the diagnosis report and the BMI value.

[0008] Even further, in step S4, according to the second collection information obtained within each interval time, calculate the corresponding safety judgment value, and set the first safety threshold and the second safety threshold , the first safety threshold is less than the second safety threshold, and compare each calculated safety judgment value with , Specifically: When the safety judgment value , judge that the safety level of this subject patient is high safety level; When the safety judgment value , judge that the safety level of this subject patient is medium safety level; When the safety judgment value, judge that the safety level of this subject patient is low safety level.

[0009] Even further, the safety judgment value is calculated by the following formula: ; In the above formula, represents the safety judgment value of each subject patient after the jth interval time, j takes values from 1 to J, represents the first weight value, represents the second weight value, represents the detection error coefficient of the BMI value, represents the total number of set times included in each interval time, n takes values from 1 to N, represents the average value of the maximum and minimum BMI values corresponding to the age of this subject patient, represents the similarity value between the diagnosis report corresponding to the nth set time and the standard diagnosis report, represents the The similarity value between the diagnostic report corresponding to a set time and the standard diagnostic report.

[0010] Furthermore, 、 is calculated according to the following formula: ; ; In the above formula, represents the BMI value of the subject patient before the clinical drug trial, represents the similarity value between the historical case information of the subject patient and the standard diagnostic report.

[0011] Furthermore, in step S5, the specific method for screening is as follows: screen out the similarity values between the diagnostic reports obtained by interviewing each subject patient at each set time and the standard diagnostic report. For each subject patient, establish a change curve with the set time as the abscissa and the similarity value as the ordinate in the same coordinate system. Each change curve encloses a change curve region with the abscissa. Select the common region of the change curve regions corresponding to the subject patients who conduct interviews at the last set time. The points falling within this common region are used as the regular data for clinical drug evaluation, and the points not falling within the common region are used as the key data for clinical drug evaluation.

[0012] Furthermore, 、 is calculated by the following formula: ; ; In the above formula, represents the maximum value of BMI corresponding to the age of the subject patient, represents the minimum value of BMI corresponding to the age of the subject patient, represents the average value of the similarity values between all diagnostic reports and the standard diagnostic report within the j-th interval time.

[0013] Furthermore, for the subject patients judged to have medium safety, reduce the dosage of the clinical drug trial in the next interval time. The dosage of the clinical drug trial that the subject patient needs to reduce in the next interval time is calculated by the following formula: ; In the above formula, represents the dosage of the clinical drug trial that the subject patient judged to have medium safety needs to reduce in the interval time, represents the dosage of the clinical drug trial that the subject patient judged to have medium safety needs to conduct in the first interval time.

[0014] Further, in step S2, obtain the key information of the drug to be clinically tested, denoted as the first key information; use natural language processing technology to extract professional terms from the historical case information in the first collected information to form multiple second key information; use a semantic similarity calculation method to calculate the semantic similarity between each second key information and the first key information to form multiple similarity values, perform clustering analysis on the multiple similarity values to form K clusters, calculate a distance threshold according to all the similarity values within the cluster where the minimum similarity value is located, screen out the clusters whose distance from the center point of the cluster where the minimum similarity value is located does not exceed the distance threshold, and use the patients corresponding to all the similarity values within the screened clusters as the test patients; the distance threshold is calculated by the following formula: ; In the above formula, represents the distance threshold, represents the abscissa of the center point of the cluster where the minimum similarity value is located, y represents the ordinate of the center point of the cluster where the minimum similarity value is located, represents the abscissa of the i-th point within the cluster where the minimum similarity value is located, represents the ordinate of the i-th point within the cluster where the minimum similarity value is located, and I represents the total number of points within the cluster where the minimum similarity value is located.

[0015] An automatic patient information processing system includes a data collection module, a data processing module, and a data screening module. The data collection module includes a first collection module and a second collection module. The data processing module includes a first processing module and a second processing module. The first collection module is connected to the first processing module. The second collection module is connected to the second processing module. The second collection module is connected to the data screening module.

[0016] Compared with the prior art, the beneficial effects of the present invention are: According to the first acquisition module and the first processing module, the present invention screens the most suitable test patients for the drug to be clinically tested. During the clinical drug test on the test patients, according to the second acquisition module and the second processing module, multiple interval times are set, and multiple set times are also provided within each interval time. It is required that each test patient conducts a medical interview and obtains a BMI value at each set time to obtain a corresponding diagnostic report. At the same time, a standard diagnostic report is set according to the drug being clinically tested. The diagnostic report after the medical interview of each test patient is compared and analyzed with the standard diagnostic report to obtain a similarity value. At each interval time, the safety level of each test patient is judged according to the similarity value and the BMI value obtained within this interval time. For the test patients judged to have a low safety level, their clinical drug tests are stopped. Finally, the data processed in the second processing module is classified by the data screening module into regular data and key data for consideration in the subsequent clinical trial evaluation stage of this drug. The present invention can ensure the accuracy of the clinical drug test results while guaranteeing the safety of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a block diagram of the system of the present invention.

[0018] Figure 2 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0019] The technical solution of the present invention will be clearly described below in conjunction with the description of the drawings. Obviously, the described embodiments are not all embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0020] As Figure 1 shown, the present invention provides an automatic patient information processing system, including a data acquisition module, a data processing module, and a data screening module. The data acquisition module includes a first acquisition module and a second acquisition module. The data processing module includes a first processing module and a second processing module. The first acquisition module is connected to the first processing module, the second acquisition module is connected to the second processing module, and the second processing module is connected to the data screening module.

[0021] The first acquisition module is used to acquire patient information during the stage of screening test subjects. The first processing module is used to process the patient information acquired by the first acquisition module. The first processing module obtains multiple test subjects whose symptoms match those treated by the clinical drug to be tested. The second acquisition module is used to acquire patient information of the test subjects screened by the first processing module. The second processing module is used to process the patient information acquired by the second acquisition module. The second processing module is used to judge the safety of the test subjects during the drug clinical trial stage. The data screening module screens out routine data and endpoint data according to the processing data of the second processing module throughout the test, providing guarantee for the subsequent effectiveness evaluation stage of the clinical drug.

[0022] As Figure 2 shown, the present invention also provides an automatic patient information processing method, including the following steps: S1. Obtain the first acquisition information. The specific method is as follows: Set a set of patients to be screened, which includes a large number of patients. Acquire the patient information of each patient in the set of patients to be screened, which is the first acquisition information. The first acquisition information includes basic information and historical case information. The basic information includes name, gender, home address, mobile phone number, etc.

[0023] S2. According to the first acquisition information, obtain multiple test subjects whose symptoms match those treated by the clinical drug to be tested. The specific method is as follows: Obtain the key information of the clinical drug to be tested, which is set as the first key information. The first key information includes information such as the name, efficacy, ingredients, and indications of the clinical drug to be tested; Use natural language processing technology to extract professional terms from the historical case information in the first acquisition information to form multiple second key information. The professional terms extracted from the historical case information of each patient form a key information. The second key information includes terms related to diseases, drug names, drug ingredients, drug indications, etc. in the patient's historical cases; Use the semantic similarity calculation method to calculate the semantic similarity between each second key information and the first key information to form multiple similarity values. Perform clustering analysis on the multiple similarity values to form K clusters. According to all the similarity values within the cluster where the smallest similarity value is located, calculate the distance threshold. Screen out the clusters whose distance from the center point of the cluster where the smallest similarity value is located does not exceed the distance threshold. The patients corresponding to all the similarity values within the screened clusters are used as test subjects; The distance threshold is calculated by the following formula: ; In the above formula, represents the distance threshold, represents the abscissa of the center point of the cluster where the smallest similarity value is located, y represents the ordinate of the center point of the cluster where the smallest similarity value is located, represents the abscissa of the i-th point within the cluster where the smallest similarity value is located, Denote the ordinate of the $i$-th point in the cluster where the minimum similarity value is located, and $I$ represents the total number of points in the cluster where the minimum similarity value is located.

[0024] S3. Obtain the second collection information. The specific method is as follows: During the clinical drug trial for each subject patient, a medical interview needs to be conducted at regular intervals. The corresponding diagnostic report and the BMI value (standard body mass index) of the subject patient are obtained. The diagnostic report includes multiple test indicators, and each test indicator is associated with the indication of the clinical drug. At each set time, the patient information of each subject patient is collected, which is the second collection information. The second collection information includes the diagnostic report and the BMI value.

[0025] S4. Set $J$ interval times during the clinical drug trial. Each interval time contains multiple set times, and each interval time contains at least three set times. After each interval time, based on all the second collection information obtained within that interval time, calculate a safety judgment value once, and judge the safety of each subject patient after that interval time according to the safety judgment value.

[0026] The specific method is as follows: Set a standard diagnostic report. The multiple indicators included in the standard diagnostic report correspond to the various test indicators in the diagnostic report after the medical interview at each set time. The indicators in the standard diagnostic report are set according to human medical standards.

[0027] The safety judgment value of each subject patient after the $j$-th interval time is calculated according to the following formula: ; ; ; In the above formula, Denote the safety judgment value of each subject patient after the $j$-th interval time, Denote the first weight value, Denote the second weight value, Denote the detection error coefficient of the BMI value, Denote the total number of set times included in each interval time, $n$ takes values from 1 to $N$, Denote the average value of the maximum and minimum BMI values corresponding to the age of the subject patient, Denote the similarity value between the diagnostic report corresponding to the $n$-th set time and the standard diagnostic report, Denote the Similarity value between the diagnostic report corresponding to the $n$-th set time and the standard diagnostic report; Denote the BMI value of the subject patient before the clinical drug trial, Indicates the similarity value between the historical case information of the subject patient and the standard diagnostic report, where j takes values from 1 to J.

[0028] 、 、 The specific method obtained is as follows: Using the GloVe algorithm, the corresponding diagnostic report is compared and analyzed with the standard diagnostic report to obtain the similarity between the two diagnostic reports, and this similarity is the similarity value.

[0029] The levels for judging the safety of the subject patient during the drug clinical trial phase include high safety, medium safety, and low safety. The specific method for judging the safety of each subject patient after the j-th interval time according to the safety judgment value is as follows: Set a first safety threshold and a second safety threshold, where the first safety threshold is less than the second safety threshold. The first safety threshold is denoted as and the second safety threshold is denoted as . Compare with 、 specifically as follows: When , it is judged that the safety level of the subject patient is high safety.

[0030] When , it is judged that the safety level of the subject patient is medium safety.

[0031] When , it is judged that the safety level of the subject patient is low safety.

[0032] 、 It is calculated through the following formula: ; ; In the above formula, represents the maximum value of the BMI corresponding to the age of the subject patient, represents the minimum value of the BMI corresponding to the age of the subject patient, represents the average value of the similarity values between all diagnostic reports and the standard diagnostic report within the j-th interval time.

[0033] For the subject patients judged to have low safety, stop the clinical drug trial for these subject patients. For the subject patients judged to have medium safety, reduce the dosage of the clinical drug trial in the next interval time. The dosage of the clinical drug trial that the subject patient needs to reduce in the next interval time is calculated through the following formula: ; In the above formula, Indicates the dosage reduction of the clinical drug trial required for the test subjects judged to have medium safety within the interval time, Indicates the dosage of the clinical drug trial that the test subjects judged to have medium safety need to undergo within the first interval time.

[0034] S5. Screen out the similarity values between the diagnostic reports obtained by asking each test subject at each set time and the standard diagnostic report. For each test subject, establish a change curve with the set time as the abscissa and the similarity value as the ordinate in the same coordinate system. Each change curve encloses a change curve area with the abscissa. For the change curve area corresponding to the test subject who undergoes the interrogation at the last set time, select the common area of these change curve areas. The points falling within the common area are used as the regular data for the clinical drug evaluation, and the points not falling within the common area are used as the key data for the clinical drug evaluation. The data screening module outputs the regular data and the key data, which helps to better utilize the non-conventional data in the subsequent clinical drug efficacy evaluation stage, improving the efficiency and accuracy of the drug efficacy evaluation.

[0035] The present invention selects the most suitable test subjects for the drug to be clinically tested. During the process of conducting the clinical drug trial on the test subjects, multiple interval times are set, and multiple set times are also provided within each interval time. It is required that each test subject undergoes an interrogation and obtains the BMI value at each set time to obtain the corresponding diagnostic report. At the same time, a standard diagnostic report is set according to the drug being clinically tested. The diagnostic report after the interrogation of each test subject is compared and analyzed with the standard diagnostic report to obtain the similarity value. At each interval time, the safety level of each test subject is judged based on the similarity value and the BMI value obtained within that interval time. For the test subjects judged to have low safety, their clinical drug trials are stopped; finally, by classifying the data processed in the second processing module into regular data and key data, it is used for the consideration in the subsequent clinical trial evaluation stage of the drug. The present invention can ensure the accuracy of the clinical drug trial results while guaranteeing the safety of the patients.

[0036] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than a limitation on the protection scope of the present invention. Any simple modification or equivalent replacement of the technical solution of the present invention by those of ordinary skill in the art does not depart from the essence and scope of the technical solution of the present invention.

Claims

1. An automatic patient information processing method, characterized in that, It includes the following steps: S1. Obtain the first collection information, where the first collection information includes historical case information; S2. According to the historical case information and the key information of the drug to be clinically tested, screen multiple subject patients who match the drug to be clinically tested; S3. Obtain the second collection information; S4. According to the second collection information, judge the safety level of each subject patient. The safety levels judged by the second processing module include low safety level, medium safety level, and high safety level. For the subject patients judged to have a low safety level, stop the clinical drug trial; S5. Screen the information of each subject patient in the clinical drug trial stage into regular data and key data.

2. The automatic patient information processing method according to claim 1, wherein Set J interval times in the clinical drug trial stage. Set multiple set times within each interval time. Each subject patient has a consultation at each set time to obtain the corresponding diagnostic report and BMI value. The second collection information includes the diagnostic report and BMI value.

3. The automatic patient information processing method according to claim 2, characterized in that, In step S4, according to the second acquisition information obtained within each interval time, calculate the corresponding safety judgment value and set the first safety threshold and the second safety threshold , where the first safety threshold is less than the second safety threshold, and compare each calculated safety judgment value with 、 Specifically, it is as follows: When the safety judgment value is, it is judged that the safety level of the subject patient is high safety; When Safety judgment value the safety level of the subject patient is judged to be medium safety; When the safety judgment value is reached, it is determined that the safety level of the subject patient is low safety.

4. The automatic patient information processing method according to claim 3, wherein, The safety judgment value is calculated by the following formula: ； In the above formula, represents the safety judgment value of each subject patient after the j-th interval time, where j ranges from 1 to J, represents the first weight value, represents the second weight value, represents the detection error coefficient of the BMI value, represents the total number of set times included in each interval time, where n ranges from 1 to N, represents the average value of the maximum and minimum BMI values corresponding to the age of the subject patient, represents the similarity value between the diagnostic report corresponding to the n-th set time and the standard diagnostic report, represents the similarity value between the diagnostic report corresponding to the [n]-th set time and the standard diagnostic report.

5. The automatic patient information processing method according to claim 4, wherein , Calculated according to the following formula: ； ； In the above formula, represents the BMI value of the subject patient before the clinical drug trial, represents the similarity value between the historical case information of the subject patient and the standard diagnostic report.

6. The automatic patient information processing method according to claim 4, wherein, In step S5, the specific method for screening is as follows: Screen out the similarity values between the diagnostic reports of each subject patient's consultation at each set time and the standard diagnostic report. For each subject patient, establish a change curve with the set time as the abscissa and the similarity value as the ordinate in the same coordinate system. Each change curve and the abscissa enclose a change curve area. Select the common area of the change curve area corresponding to the subject patient who has the consultation at the last set time. The points falling within this common area are used as the regular data for clinical drug evaluation, and the points not falling within the common area are used as the key data for clinical drug evaluation.

7. A method for automatically processing patient information according to claim 4, characterized in that, , which is calculated by the following formula: ； ； In the above formula, represents the maximum value of BMI corresponding to the age of the subject patient, represents the minimum value of BMI corresponding to the age of the subject patient, represents the average value of the similarity values between all diagnostic reports and the standard diagnostic report within the j-th interval time.

8. The automatic patient information processing method according to claim 3, characterized in that For the subject patients judged to have a medium safety level, reduce the dosage of the clinical drug trial in the next interval time. The dosage of the clinical drug trial that the subject patient needs to reduce in the next interval time is calculated by the following formula: ； In the above formula, represents the dosage reduction of the clinical drug trial required for the test subjects judged to have medium safety within the interval, represents the dosage of the clinical drug trial that the test subjects judged to have medium safety need to undergo within the first interval.

9. The automatic patient information processing method according to claim 1, characterized in that In step S2, obtain the key information of the drug to be clinically tested, denoted as the first key information; use natural language processing technology to extract professional terms from the historical case information in the first collection information to form multiple second key information; Adopt a semantic similarity calculation method to calculate the semantic similarity between each second key information and the first key information to form multiple similarity values. Perform clustering analysis on the multiple similarity values to form K clusters. According to all the similarity values within the cluster where the minimum similarity value is located, calculate the distance threshold. Screen out the clusters whose distance from the center point of the cluster where the minimum similarity value is located does not exceed the distance threshold. The patients corresponding to all the similarity values within the screened clusters are used as subject patients; the distance threshold is calculated by the following formula: ； In the above formula, represents the distance threshold, represents the abscissa of the center point of the cluster where the minimum similarity value is located, y represents the ordinate of the center point of the cluster where the minimum similarity value is located, represents the abscissa of the i-th point in the cluster where the minimum similarity value is located, represents the ordinate of the i-th point in the cluster where the minimum similarity value is located, and I represents the total number of points in the cluster where the minimum similarity value is located.

10. An automatic patient information processing system, characterized in that, It is carried out using the automatic patient information processing method described in any one of claims 1-9, including a data collection module, a data processing module, and a data screening module. The data collection module includes a first collection module and a second collection module. The data processing module includes a first processing module and a second processing module. The first collection module is connected to the first processing module. The second collection module is connected to the second processing module. The second collection module is connected to the data screening module.

Images