Clinical information processing system based on single disease

Abstract

This application discloses a clinical information processing system based on a single disease, including a clinical information acquisition module for retrieving corresponding clinical information based on the patient's identity information for a single disease and generating a clinical data report. The clinical information includes the patient's first basic information, first physical examination information, and first treatment information, all automatically uploaded by a server. A treatment information matching module is used to match the first basic information and first physical examination information from the clinical data report with corresponding second treatment information using a trained deep learning neural network. A treatment information comparison module is used to mark the first treatment information as correct when the error between the first and second treatment information is within a preset range. Compared with existing technologies, this application improves the efficiency of quality management evaluation by automatically uploading information and automatically generating evaluation information using neural networks, thereby reducing workload.

Description

Technical Field

[0001] This application relates to the field of single-disease processing technology in medical systems, and more particularly to a clinical information processing system based on a single disease. Background Technology

[0002] A single disease is a specific illness that does not produce complications. Current single-disease quality management relies on clinical information about the disease, such as medical quality indicators like diagnosis and treatment, for quality evaluation. Traditional quality management evaluation involves verbal or written management and processing of medical diseases, but it fails to rigorously control quality during and after treatment, nor can it continuously improve medical treatment techniques or evaluate whether physicians' treatment behaviors conform to standards and rationality. This results in low treatment efficiency and unreliable quality.

[0003] To address the aforementioned issues, existing technologies employ a method of generating disease-specific lectures in real-time based on patients' clinical information, thereby providing accurate data for quality management evaluation. However, when using this method for quality management evaluation, it has been found that some clinical information, such as laboratory reports, requires doctors to manually drag and drop, increasing their workload. Furthermore, disease-specific quality management evaluation relies on specialists treating that disease to accurately assess physicians' diagnostic and treatment techniques and behaviors, further increasing the specialists' workload. The quality management evaluation also consumes a significant amount of time, resulting in low efficiency. Summary of the Invention

[0004] The technical problem to be solved by the embodiments of this application is to provide a clinical information processing system based on a single disease, which reduces workload and improves the efficiency of quality management and evaluation.

[0005] To address the aforementioned problems, embodiments of this application provide a clinical information processing system based on a single disease, comprising:

[0006] The clinical information collection module is used to collect the identity information of patients with a single disease, and to search for the clinical information corresponding to the single disease patient in the database based on the identity information, and generate a clinical data report; wherein, the clinical information includes the first basic information, the first physical examination information, and the first diagnosis and treatment information of the single disease patient, the first basic information including age information and gender information, which is automatically uploaded by the first server, the first physical examination information is automatically uploaded by the second server, and the first diagnosis and treatment information is automatically uploaded by the third server;

[0007] The diagnosis and treatment information matching module is used to match the first basic information and the first physical examination information in the clinical data report with the corresponding second diagnosis and treatment information through a trained deep learning neural network.

[0008] The diagnosis and treatment information comparison module is used to compare the first diagnosis and treatment information with the second diagnosis and treatment information, and when the error between the first diagnosis and treatment information and the second diagnosis and treatment information is within a preset range, the first diagnosis and treatment information is marked as the correct diagnosis and treatment information.

[0009] Furthermore, the diagnostic information matching module includes a training unit and a matching unit;

[0010] The training unit is used to collect multiple second basic information and multiple second physical examination information corresponding one-to-one with multiple single disease patients. For any information combination formed by any second basic information and any second physical examination information, after presetting the corresponding second diagnosis and treatment information, each information combination is input into the deep learning neural network to be trained for matching training until each information combination is matched with the corresponding second diagnosis and treatment information.

[0011] The matching unit is used to match the first basic information and the second physical examination information in the clinical data report with the corresponding second diagnosis and treatment information through a trained deep learning neural network.

[0012] Furthermore, the matching unit is also used for:

[0013] When the first basic information and the first physical examination information in the clinical data report do not match the corresponding second diagnostic information through the trained deep learning neural network, the first basic information and the first physical examination information are sent to the smart terminal as a combination to be confirmed. The diagnostic information fed back by the smart terminal based on the combination to be confirmed is then used as the second diagnostic information corresponding to the combination to be confirmed. Each combination to be confirmed is then input into the deep learning neural network to be trained for matching training until the combination to be confirmed is matched with the corresponding second diagnostic information.

[0014] Furthermore, it also includes:

[0015] The model building module is used to build a three-dimensional clinical information model based on multiple sets of second basic information, multiple sets of second physical examination information, and multiple sets of second diagnostic and treatment information; wherein, the three-dimensional clinical information model includes an X-axis, a Y-axis, and a Z-axis, where the X-axis represents the second basic information, the Y-axis represents the second physical examination information, and the Z-axis represents the second diagnostic and treatment information.

[0016] Furthermore, the diagnostic information comparison module includes:

[0017] An information comparison unit is used to compare the first diagnostic information with the second diagnostic information;

[0018] The first processing unit is configured to mark the first diagnostic information as correct diagnostic information when the error between the first diagnostic information and the second diagnostic information is within a preset range.

[0019] The second processing unit is used to mark the clinical data report and the corresponding second diagnostic information as a group of information to be confirmed and send them to a preset smart terminal when the error between the first diagnostic information and the second diagnostic information exceeds a preset range.

[0020] Furthermore, the clinical information also includes the patient's consultation type information for the single disease, which is automatically uploaded through the first server.

[0021] Furthermore, the first server connects to the registration system of the authorized medical institution, the second server connects to the physical examination system of the authorized medical institution, and the third server connects to the diagnosis and treatment system of the authorized medical institution.

[0022] Implementing the embodiments of this application has the following beneficial effects:

[0023] This application provides a single-disease-based clinical information processing system that reduces doctors' workload by automatically uploading information and automatically generates evaluation information using neural networks, thereby further reducing workload while improving the efficiency of quality management evaluation. Attached Figure Description

[0024] Figure 1 is a schematic diagram of the structure of a single-disease-based clinical information processing system provided in an embodiment of this application;

[0025] Figure 2 is a schematic diagram of the diagnosis and treatment information matching module;

[0026] Figure 3 is a schematic diagram of the diagnostic information comparison module;

[0027] Figure 4 is a flowchart illustrating a single-disease-based clinical information processing system provided in another embodiment of this application;

[0028] Figure 5. Three-dimensional clinical information model diagram. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] Referring to Figure 1, which is a schematic diagram of the structure of a single-disease-based clinical information processing system provided in an embodiment of this application, it includes:

[0031] The clinical information collection module 101 is used to collect the identity information of patients with a single disease, and to search for the corresponding clinical information in the database based on the identity information to generate a clinical data report.

[0032] The clinical information includes the first basic information, first physical examination information, and first diagnosis and treatment information of patients with a single disease. The first basic information includes age and gender information, which is automatically uploaded by the first server. The first physical examination information is automatically uploaded by the second server, and the first diagnosis and treatment information is automatically uploaded by the third server.

[0033] In this embodiment, authorized medical institutions and regions can achieve interconnection based on standardized semantic interoperability through the server. In this case, each medical visit is not limited to a single medical institution, but includes the patient's medical records at all medical institutions throughout their life. Clinical information transcends the boundaries of medical institutions and regions.

[0034] In this embodiment, age information can be uniformly divided into age segments such as neonatal period, infancy, early childhood, preschool age, school age, adolescence, youth, middle age, and old age. The specific age segment division can be determined by medical experts after discussion.

[0035] In this embodiment, the identity information may include, but is not limited to, the user's name and ID card information. Clinical information also includes the patient's treatment type information for a single disease. The treatment type information is automatically uploaded through the first server. Treatment type information includes outpatient, inpatient, and health check-up information.

[0036] As a preferred example of this embodiment, the first server connects to the registration system of an authorized medical institution. When the basic information and consultation type information of a single-disease patient are input into the registration system, the consultation type information and basic information of the single-disease patient are automatically uploaded to the database. The second server connects to the physical examination system of an authorized medical institution. When the physical examination system, such as a physical examination center, generates physical examination information for a single-disease patient, this physical examination information is automatically uploaded to the database through the second server. The third server connects to the diagnosis and treatment system of an authorized medical institution. When a doctor makes a diagnosis and treatment opinion for a single-disease patient through the diagnosis and treatment system, such as the selection of medication, the diagnosis and treatment system automatically uploads the diagnosis and treatment opinion as diagnosis and treatment information to the database through the third server. This automatic information uploading operation by the server eliminates the need for doctors to manually drag and drop and upload test reports.

[0037] As another preferred example of this embodiment, the first server, the second server, and the third server are all connected to the registration system, the physical examination system, and the diagnosis and treatment system. The first server, the second server, and the third server can all upload basic information, consultation type information, physical examination information, and diagnosis and treatment information. Through the above methods, even if one server fails, information can still be uploaded through other servers.

[0038] The diagnosis and treatment information matching module 102 is used to match the first basic information and the first physical examination information in the clinical data report with the corresponding second diagnosis and treatment information through a trained deep learning neural network.

[0039] In this embodiment, as shown in FIG2, the diagnosis and treatment information matching module 102 includes a training unit 201 and a matching unit 202.

[0040] The training unit 201 is used to collect multiple second basic information and multiple second physical examination information that correspond one-to-one with multiple single disease patients. For any combination of second basic information and any second physical examination information, after pre-setting the corresponding second diagnosis and treatment information, each information combination is input into the deep learning neural network to be trained for matching training until each information combination is matched with the corresponding second diagnosis and treatment information.

[0041] In this embodiment, the second diagnostic information refers to the expert group's diagnostic opinion on the combination of any second basic information and any second physical examination information. For example, if the expert group diagnoses a single-disease patient in the preschool age group based on the disease problems reflected in each physical examination indicator in their second physical examination information, they can determine the treatment plan required for patients in this age group when the aforementioned physical examination indicators are present; this is the second diagnostic information. The samples of the second basic information and the second physical examination information represent all samples that can be collected in the current state. When the sample size of the second basic information and the second physical examination information is sufficiently large, the training effect of the deep learning neural network will be better, and the subsequent information comparison will be more accurate.

[0042] The matching unit 202 is used to match the first basic information and the first physical examination information in the clinical data report with the corresponding second diagnosis and treatment information through a trained deep learning neural network.

[0043] In this embodiment, the matching unit 202 is further configured to, when the first basic information and the first physical examination information in the clinical data report do not match the corresponding second diagnostic and treatment information through the trained deep learning neural network, send the first basic information and the first physical examination information as a combination to be confirmed to the smart terminal, and take the diagnostic and treatment information fed back by the smart terminal based on the combination to be confirmed as the second diagnostic and treatment information corresponding to the combination to be confirmed, and then input each combination to be confirmed into the deep learning neural network to be trained for matching training until the combination to be confirmed matches the corresponding second diagnostic and treatment information.

[0044] Since the accuracy of information comparison depends on the number of training samples, insufficient samples may lead to matching failures when matching diagnostic information. In such cases, the first basic information and the first physical examination information that failed to match are sent as a combination to be confirmed to the smart terminal. This notifies the expert group to formulate corresponding second diagnostic information for this combination. Then, using this combination and its corresponding second diagnostic information, the deep learning neural network is further trained to optimize the information comparison.

[0045] The diagnosis and treatment information comparison module 103 is used to compare the first diagnosis and treatment information with the second diagnosis and treatment information, and when the error between the first diagnosis and treatment information and the second diagnosis and treatment information is within a preset range, the first diagnosis and treatment information is marked as the correct diagnosis and treatment information.

[0046] In this embodiment, the preset range is set so that the medication information in the first and second treatment information must be the same, such as the type and dosage of medication, while other aspects such as dietary recommendations may have some errors. If the medication information is the same, the first treatment information is marked as the correct treatment information, and the doctor who provided the first treatment information is given a positive evaluation.

[0047] In this embodiment, as shown in FIG3, the diagnosis and treatment information comparison module 103 includes:

[0048] The information comparison unit 301 is used to compare the first diagnosis and treatment information with the second diagnosis and treatment information.

[0049] The first processing unit 302 is used to mark the first diagnostic information as correct diagnostic information when the error between the first diagnostic information and the second diagnostic information is within a preset range.

[0050] The second processing unit 303 is used to mark the clinical data report and the corresponding second diagnostic information as a group of information to be confirmed and send them to a preset smart terminal when the error between the first diagnostic information and the second diagnostic information exceeds a preset range.

[0051] In this embodiment, since multiple treatments may exist for the same disease, when the medication information in the first treatment information differs from the second treatment information, the clinical data report corresponding to the first treatment information and the second treatment information are sent to the smart terminal. This notifies the expert group to determine whether the first treatment information in the clinical data report is feasible. If feasible, the first treatment information is added to the second treatment information as an option for subsequent treatment, based on a positive evaluation from the doctor who provided the first treatment information. Otherwise, a negative evaluation is given.

[0052] Furthermore, as shown in Figure 4, another embodiment of this application provides a flowchart of a clinical information processing system based on a single disease, including:

[0053] The model building module 104 is used to build a three-dimensional clinical information model based on multiple second basic information, multiple second physical examination information, and multiple second diagnostic and treatment information. The three-dimensional clinical information model includes an X-axis, a Y-axis, and a Z-axis, where the X-axis represents the second basic information, the Y-axis represents the second physical examination information, and the Z-axis represents the second diagnostic and treatment information.

[0054] In this embodiment, the three-dimensional clinical information model can be shown in Figure 5. Since any spatial location in the three-dimensional space corresponds to a specific clinical data report, a complete and three-dimensional clinical data report record is formed, which can be used to comprehensively reflect the overall picture of the clinical information content.

[0055] This application provides a single-disease-based clinical information processing system, including a clinical information acquisition module for collecting the identity information of patients with a single disease, searching for corresponding clinical information in a database based on the identity information, and generating a clinical data report. The clinical information includes the patient's first basic information, first physical examination information, and first treatment information, all automatically uploaded by a server. A treatment information matching module uses a trained deep learning neural network to match the first basic information and first physical examination information from the clinical data report with corresponding second treatment information. A treatment information comparison module compares the first treatment information with the second treatment information, and marks the first treatment information as correct when the error between the two is within a preset range. Compared with existing technologies, this embodiment reduces the workload of doctors by automatically uploading information and automatically generates evaluation information using a neural network, thereby further reducing workload while improving the efficiency of quality management evaluation.

[0056] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. A single-disease-based clinical information processing system characterized by comprising: The application comprises: a clinical information collection module for collecting identity information of a single-disease patient, searching for clinical information corresponding to the single-disease patient in a database according to the identity information, and generating a clinical data report; wherein the clinical information comprises first basic information, first physical examination information and first diagnosis and treatment information of the single-disease patient, the first basic information comprises age information and gender information, is automatically uploaded by a first server, the first physical examination information is automatically uploaded by a second server, and the first diagnosis and treatment information is automatically uploaded by a third server; the first server accesses a registration system of an authorized medical institution, the second server accesses a physical examination system of the authorized medical institution, and the third server accesses a diagnosis and treatment system of the authorized medical institution; a diagnosis and treatment information matching module for matching corresponding second diagnosis and treatment information through a trained deep learning neural network by using the first basic information and the first physical examination information in the clinical data report; a diagnosis and treatment information comparison module for comparing the first diagnosis and treatment information with the second diagnosis and treatment information, and marking the first diagnosis and treatment information as correct diagnosis and treatment information when an error between the first diagnosis and treatment information and the second diagnosis and treatment information is within a preset range; the diagnosis and treatment information matching module comprises a matching unit; the matching unit is configured to, when the first basic information and the first physical examination information in the clinical data report are not matched with corresponding second diagnosis and treatment information through a trained deep learning neural network, send the first basic information and the first physical examination information as a to-be-confirmed combination to an intelligent terminal, and input each to-be-confirmed combination into a deep learning neural network to be trained for matching training until the to-be-confirmed combination is matched with corresponding second diagnosis and treatment information after the intelligent terminal feeds back diagnosis and treatment information based on the to-be-confirmed combination as the second diagnosis and treatment information corresponding to the to-be-confirmed combination.

2. The single case-based clinical information processing system according to claim 1, wherein, the diagnosis and treatment information matching module further comprises a training unit; the training unit is configured to collect a plurality of second basic information and a plurality of second physical examination information corresponding to a plurality of single-disease patients, preset corresponding second diagnosis and treatment information for an information combination formed by any one of the second basic information and any one of the second physical examination information, and input each information combination into a deep learning neural network to be trained for matching training until each information combination is matched with corresponding second diagnosis and treatment information. the matching unit is further configured to match corresponding second diagnosis and treatment information through a trained deep learning neural network by using the first basic information and the second physical examination information in the clinical data report.

3. The single case-based clinical information processing system according to claim 2, wherein, The application further comprises: a model establishment module for establishing a three-dimensional clinical information model according to a plurality of second basic information, a plurality of second physical examination information and a plurality of second diagnosis and treatment information; wherein the three-dimensional clinical information model comprises an X axis, a Y axis and a Z axis, the X axis is the second basic information, the Y axis is the second physical examination information, and the Z axis is the second diagnosis and treatment information.

4. The single case-based clinical information processing system according to claim 1, wherein, the diagnosis and treatment information comparison module comprises: an information comparison unit for comparing the first diagnosis and treatment information with the second diagnosis and treatment information. The first processing unit is configured to mark the first diagnosis and treatment information as correct diagnosis and treatment information when an error between the first diagnosis and treatment information and the second diagnosis and treatment information is within a preset range. The second processing unit is configured to mark the clinical data report and the corresponding second diagnosis and treatment information as a to-be-confirmed information group and send the to-be-confirmed information group to a preset intelligent terminal when the error between the first diagnosis and treatment information and the second diagnosis and treatment information exceeds the preset range.

5. The single case-based clinical information processing system according to any one of claims 1 to 4, wherein The clinical information further comprises visit type information of the single-disease patient, and the visit type information is automatically uploaded by the first server.

Images