Precision management system for diabetic foot based on artificial intelligence

Abstract

The present invention discloses an artificial intelligence-based accurate management system for diabetic foot, including a medical record management subsystem, a big data language model subsystem, a medical and nursing end, and a user end; the medical record management subsystem is used to obtain the new medical record of the patient input in real time by the medical and nursing end and record the time of the new medical record, wherein the new medical record of the patient contains the words "diabetic foot"; the medical record management subsystem sends the new medical record of the patient, the medical records of other historical patients, and the matching degree instruction to the big data language model subsystem, the medical record management subsystem obtains the single patient matching result of the big data language model subsystem, and the medical record management subsystem records the most matching medical record of each other patient. The present invention realizes the automatic tracking and follow-up of diabetic foot, improves the accuracy of tracking and follow-up, and realizes the accurate management of diabetic foot.

Description

Technical Field

[0001] The present invention relates to the technical field of diabetic foot management, and in particular to an artificial intelligence-based precise management system for diabetic foot. Background Art

[0002] With the rapid development of artificial intelligence, the development of artificial intelligence in the medical field is becoming increasingly closely integrated with disease diagnosis. With the continuous advancement of technology, the application of artificial intelligence in medical imaging, pathology, genetics and other fields has gradually increased, providing new possibilities for the diagnosis and treatment of diseases.

[0003] Especially in medical imaging, AI can help doctors discover and diagnose diseases by analyzing large amounts of medical imaging data through deep learning algorithms. For example, AI can automatically detect and identify tumors, lesions, and other abnormalities in CT scans, MRIs, and X-ray images. In addition, AI can also provide more accurate prognostic assessments and treatment recommendations through analysis of imaging data.

[0004] However, artificial intelligence in the treatment of diabetic foot mostly focuses on the graded diagnosis of diabetic foot images, such as grading the severity of the disease by the size of the wound in the image data. This is very dependent on image data, and there is a lack of image data for a large number of historical medical records, which cannot be used as historical data. At the same time, the treatment of diabetic foot is a long process, and there is a high possibility of recurrence. Follow-up of diabetic foot treatment is very necessary. At present, follow-up relies on the experience of medical staff, and the effect is not very ideal. Summary of the invention

[0005] To this end, it is necessary to provide an AI-based precise management system for diabetic foot to solve the problem that existing diabetic foot requires imaging as a basis for judgment and the follow-up after treatment is not ideal.

[0006] To achieve the above-mentioned purpose, the present invention provides an artificial intelligence-based accurate management system for diabetic foot, including a medical record management subsystem, a big data language model subsystem, a medical terminal and a user terminal;

[0007] The medical record management subsystem is used to obtain the new medical records of patients input by the medical staff in real time and record the time of the new medical records, wherein the new medical records of patients contain the words "diabetic foot";

[0008] The medical record management subsystem sends the patient's new medical record, other patients' historical medical records and matching degree instructions to the big data language model subsystem. The medical record management subsystem obtains the single patient matching result of the big data language model subsystem. The medical record management subsystem records the most matching medical record of each other patient.

[0009] The medical record management subsystem sends the most matching medical record of each other patient and the next medical record and disease progression instruction of the patient to the big data language model subsystem, and the medical record management subsystem obtains the disease progression result returned by the big data language model subsystem;

[0010] The medical record management subsystem obtains the next medical record corresponding to the positive result in the disease progression result, counts the concentrated time period of the next medical record, sends the next medical record and feature control instructions of the concentrated time period to the big data language model subsystem, and the medical record management subsystem obtains the disease progression feature returned by the big data language model subsystem;

[0011] The medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, sends the disease progression characteristics and corresponding options to the user end after the sending time arrives, and the medical record management subsystem obtains the option result returned by the user end;

[0012] The medical record management subsystem sends the option result, the next medical record in the concentrated time period and the matching degree to the big data language model subsystem to obtain the option matching result;

[0013] The medical record management subsystem sends the option result and the patient's new medical record to the medical care end when the option matching result does not match.

[0014] Furthermore, the medical record management subsystem includes the steps of:

[0015] The medical record management subsystem sends the medical order and summary control instructions corresponding to the next medical record in a concentrated time period to the big data language model subsystem. The medical record management subsystem obtains the medical order summary returned by the big data language model subsystem and sends the medical order summary to the user end.

[0016] Furthermore, the medical record management subsystem only sends a prompt message to the user terminal when the option matching results match.

[0017] Furthermore, the medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, including: the medical record management subsystem uses the time with the largest number of medical records in the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time.

[0018] Furthermore, the medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, including: the medical record management subsystem uses the middle time of the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time.

[0019] Furthermore, the medical record management subsystem further comprises the following steps when the option matching result does not match:

[0020] The medical record management subsystem sends the medical instructions corresponding to the next medical record in a concentrated time period to the medical care end.

[0021] Furthermore, the medical end includes a medical identity code, and the patient's new medical record corresponds to the medical identity code of the medical end. When the option matching result does not match, the medical record management subsystem sends the option result and the patient's new medical record to the medical end corresponding to the medical identity code of the patient's new medical record.

[0022] Furthermore, the medical record management subsystem further includes the steps of sending the disease progression characteristics and corresponding options to the user end after the sending time arrives:

[0023] If the medical record management subsystem does not obtain the option result returned by the user end after the predetermined waiting time has passed, the patient information in the new medical record of the patient is sent to the medical care end.

[0024] Furthermore, the medical record management subsystem sends the patient's new medical record, other historical patient medical records and matching degree instructions to the big data language model subsystem, including the following steps:

[0025] The medical record management subsystem sends new patient medical records, historical medical records of other patients, and matching degree instructions to the big data language model subsystem within a preset time period.

[0026] Furthermore, the big data language model subsystem is a GPT big data language model.

[0027] Different from the existing technology, the above technical solution can match patients with diabetic foot with similar conditions by matching new cases of diabetic foot with the medical records of historical patients. Then, a big data language model is used to summarize the medical characteristics of other patients. These characteristics can be sent to patients at a specific time to register their self-recovery status, and then the big data language model is used to match them again with other historical patient medical records. If they match, the prognosis is developing in a normal direction, so there is no need for too much intervention, which can reduce the energy investment of medical staff. If there is no match, there may be a poor prognosis, and the medical staff will be prompted. The medical staff can manually check and decide whether to communicate further with the patient. In this way, automatic tracking and follow-up of diabetic foot is achieved, the accuracy of tracking and follow-up is improved, and accurate management of diabetic foot is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the structure of the precise management system for diabetic foot described in the specific implementation method;

[0029] Figure 2 This is a method flow chart of the precise management system for diabetic foot described in the specific implementation method;

[0030] Figure 3 A method flow chart of another embodiment.

[0031] Description of reference numerals:

[0032] 1. Precision management system for diabetic foot;

[0033] 10. Medical record management subsystem;

[0034] 11. Big data language model subsystem;

[0035] 12. Medical care end;

[0036] 13. User end. DETAILED DESCRIPTION

[0037] In order to explain the technical content, structural features, achieved objectives and effects of the technical solution in detail, the following is a detailed description in conjunction with specific embodiments and accompanying drawings.

[0038] Reference to "embodiment" herein means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The term "embodiment" appearing in various places in the specification does not necessarily refer to the same embodiment, nor does it particularly limit its independence or association with other embodiments. In principle, in the present application, as long as there is no technical contradiction or conflict, the various technical features mentioned in the embodiments can be combined in any way to form a corresponding implementable technical solution.

[0039] Unless otherwise defined, the technical terms used in this document have the same meanings as those generally understood by those skilled in the art to which this application belongs; the use of relevant terms in this document is only for describing specific embodiments and is not intended to limit this application.

[0040] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships may exist, for example, A and / or B, which means: A exists, B exists, and A and B exist at the same time. In addition, the character " / " in this article generally indicates that the objects before and after are in a logical relationship of "or".

[0041] In the present application, terms such as “first” and “second” are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship of quantity, priority or sequence between these entities or operations.

[0042] Without further limitations, in this application, the words "include", "comprises", "has" or other similar expressions used in the sentences are intended to cover non-exclusive inclusion. These expressions do not exclude the presence of additional elements in the process, method or product including the elements, so that the process, method or product including a series of elements may include not only those limited elements, but also other elements not explicitly listed, or also include elements inherent to such process, method or product.

[0043] Similar to the understanding in the Examination Guidelines, in this application, expressions such as "greater than", "less than", "exceed" and the like are understood to exclude the number itself; expressions such as "above", "below", "within" and the like are understood to include the number itself. In addition, in the description of the embodiments of this application, "multiple" means more than two (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups", "multiple times", etc., unless otherwise clearly and specifically limited.

[0044] In the description of the embodiments of the present application, space-related expressions used, such as "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "vertical", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicate the orientation or position relationship based on the orientation or position relationship shown in the specific embodiments or drawings, and are only for the convenience of describing the specific embodiments of the present application or facilitating the reader's understanding, and do not indicate or imply that the referred device or component must have a specific position, a specific orientation, or be constructed or operated in a specific orientation, and therefore cannot be understood as a limitation on the embodiments of the present application.

[0045] Unless otherwise expressly specified or limited, in the description of the embodiments of the present application, the terms such as "install", "connect", "connect", "fix", "set", etc. used should be understood in a broad sense. For example, the "connection" can be a fixed connection, a detachable connection, or an integrated setting; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For technicians in the technical field to which the present application belongs, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0046] See also Figures 1 to 3This embodiment provides an artificial intelligence-based precise management system for diabetic foot 1, including a medical record management subsystem 10, a big data language model subsystem 11, a medical terminal 12, and a user terminal 13. The medical record management subsystem 10 is used to manage, store, and send medical records. The big data language model subsystem 11 is provided with a big data language model based on artificial intelligence, and can answer big data language based on input questions; the medical terminal 12 is used to provide for doctors or hospital staff, and the user terminal 13 is provided for users and patients.

[0047] First, in step S201, the medical record management subsystem obtains the new medical record of the patient input in real time by the medical staff and records the time of the new medical record, wherein the new medical record of the patient contains the words "diabetic foot". That is, the medical record management subsystem first obtains the new medical record related to diabetic foot.

[0048] Then, in step S202, the medical record management subsystem sends the patient's new medical record, other historical patient medical records, and the matching degree instruction to the big data language model subsystem. The medical record management subsystem obtains the single patient matching result of the big data language model subsystem, and the medical record management subsystem records the most matching medical record of each other patient. The matching degree instruction is used to instruct the big data language model to match the patient's new medical record with other historical patient medical records. It can be to output the matching degree value or to find the most matching medical record, and finally find a most matching medical record from each other patient. Of course, the other patient medical records here can be pre-screened as medical records related to diabetic foot, which can reduce the number of irrelevant medical records to be matched.

[0049] In step S203, the medical record management subsystem sends the most matching medical record of each other patient, the next medical record of the patient, and the disease progression instruction to the big data language model subsystem, and the medical record management subsystem obtains the disease progression result returned by the big data language model subsystem. The disease progression instruction is used to instruct the big data language model to find the changes in the two medical records, such as finding the disease change and the corresponding disease name, where the change can be a color change, a size change, etc. The disease progression result can be a lighter color, a darker color, a smaller wound, a larger wound, etc.

[0050] In step S204, the medical record management subsystem obtains the next medical record corresponding to the result of positive progress in the disease progress result, counts the concentrated time period of the next medical record, sends the next medical record of the concentrated time period and the feature control instruction to the big data language model subsystem, and the medical record management subsystem obtains the disease progress feature returned by the big data language model subsystem. Positive progress indicates that the disease is developing in a good direction, and a method of pre-setting matching words for screening can be adopted, such as shallowing as a matching word, and if shallowing is matched, it is positive progress. The medical record matched in this way is the result of the disease getting better. The concentrated time period can be the time period corresponding to the medical record whose number of medical records is greater than a certain value or the percentage value of the total medical records, such as the next medical record is concentrated between 20 days and 30 days. The feature control instruction is used by the big data language model subsystem to filter out the characteristics of the disease, which indicate the good condition of the disease, such as the wound becomes smaller and the color becomes lighter, which is recorded as the disease progress feature, and the feature control instruction can be to filter out similar disease features in the medical record.

[0051] In step S205, the medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, and the sending time is equal to the time when the new medical record time exceeds the concentrated time period. After the sending time arrives, the disease progression characteristics and corresponding options are sent to the user end, and the medical record management subsystem obtains the option results returned by the user end. The options here can be yes or no, and the patient can choose according to his or her own recovery situation. In step S206, the medical record management subsystem sends the option results and the next medical record and matching degree instructions of the concentrated time period to the big data language model subsystem to obtain the option matching results. In this way, when matching through the big data language model, it is determined whether the user's recovery situation matches the results of other medical records. Matching means good recovery. In step S207, the medical record management subsystem sends the option results and the patient's new medical record to the medical care end when the option matching results do not match, so that the medical staff is reminded when there is a mismatch, and the medical staff can manually screen and determine whether manual intervention is required, thereby tracking after treatment.

[0052] The present invention can match patients with diabetic foot with similar conditions by matching new cases of diabetic foot with the medical records of historical patients. Then, the medical records of other patients are summarized with a big data language model. These characteristics can be sent to the patient at a specific time to register the self-recovery situation, and then matched again with other historical patient medical records by the big data language model. If it is matched, the prognosis develops in a normal direction, and there is no need for too much intervention, which can reduce the energy investment of medical staff. If it does not match, the medical staff is prompted, and the medical staff can decide whether to communicate further with the patient after manual review. In this way, automatic tracking and follow-up of diabetic foot is achieved, the accuracy of tracking and follow-up is improved, and accurate management of diabetic foot is achieved.

[0053] Furthermore, if Figure 3 As shown, when the option matching result matches, the medical record management subsystem includes the following steps: Step S301 The medical record management subsystem sends the medical order and summary control instruction corresponding to the next medical record in the concentrated time period to the big data language model subsystem, and Step S302 The medical record management subsystem obtains the summary of the medical order returned by the big data language model subsystem, and sends the summary of the medical order to the user end. The big data language model subsystem can summarize the medical order to obtain a summarized medical order. If a large number of continuing treatments appear in the medical order, the medical order of continuing treatment can be summarized, and the medical order is directly sent to the user end by the management system. The user can use the medical order for follow-up treatment, so that there is no need for doctors to operate the medical order, which greatly improves efficiency.

[0054] Furthermore, the medical record management subsystem 10 only sends a prompt message to the user terminal 13 when the option matching result matches. When the option matching result matches, it means that the disease progression is similar to the medical record with good results before, and the disease progression is good. The user terminal 13 can be prompted with a good prognosis alone without sending it to the medical terminal 12, thereby reducing the workload of the medical terminal 12.

[0055] In some embodiments, the medical record management subsystem 10 calculates the sending time according to the concentrated time period and the new medical record time, including: the medical record management subsystem 10 uses the time with the largest number of medical records in the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time. Statistics can be performed on a daily basis to calculate which day has the largest number of medical records. If the largest number of medical records is one month apart, this time is used as the sending time. In this way, the time interval when the system collects user data is the same as the time interval with the largest number of medical records, and the time of prognosis tracking can be more consistent with the actual situation.

[0056] In some embodiments, the time with the largest number of medical records mentioned above may also deviate, so the middle time may also be used. At this time, the medical record management subsystem 10 calculates the sending time based on the concentrated time period and the new medical record time, including: the medical record management subsystem 10 uses the middle time of the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time. If the time period is 7 days before and after, the middle time is the 4th day in the middle. Of course, the number of medical records corresponding to the concentrated time period here can be greater than a preset value, and the time corresponding to these large numbers of medical records is recorded as the concentrated time period.

[0057] Furthermore, when the option matching result does not match, the medical record management subsystem 10 further includes the following steps: the medical record management subsystem 10 sends the medical advice corresponding to the next medical record in the concentrated time period to the medical care terminal 12. The medical advice here can be used as a reference for doctors, and the development of the disease in the past can also be known, so that doctors can make more accurate judgments.

[0058] In order to realize the association between doctors and patients, the medical terminal 12 includes a medical identification code, that is, an identification code, and the new medical record of the patient corresponds to the medical identification code of the medical terminal 12. When the option matching result does not match, the medical record management subsystem 10 sends the option result and the new medical record of the patient to the medical terminal 12 corresponding to the medical identification code of the new medical record of the patient. In this way, the information can be sent to the doctor corresponding to the medical terminal 12 who filled in the medical record last time, and the doctor can view the recent situation of the patient from the option result, so as to decide whether to actively contact the patient for further examination.

[0059] Furthermore, after the medical record management subsystem 10 sends the disease progression characteristics and corresponding options to the user terminal 13 after the sending time arrives, the medical record management subsystem 10 also includes the step of: if the medical record management subsystem 10 does not obtain the option result returned by the user terminal 13 after the predetermined waiting time, the patient information in the new medical record of the patient is sent to the medical care terminal 12. In this way, if the patient (user) does not perform a system return visit operation, the medical care terminal 12 can know the patient's timeout situation, and can check and decide whether to return the visit to achieve post-treatment tracking.

[0060] In order to reduce the work intensity of the system, the medical record management subsystem 10 sends the patient's new medical record, other historical patient medical records and matching degree instructions to the big data language model subsystem 11, including the following steps: the medical record management subsystem 10 sends the patient's new medical record, other historical patient medical records and matching degree instructions to the big data language model subsystem 11 in a preset time period. The preset time period can be a time period with low system work intensity, such as early morning, so that system processing resources can be concentrated for processing.

[0061] Furthermore, the big data language model subsystem 11 is a GPT big data language model. In some embodiments, it can also be an artificial intelligence big data model such as BERT or Wenxinyiyan.

[0062] It should be noted that, although the above embodiments have been described in this article, the patent protection scope of the present invention is not limited thereby. Therefore, based on the innovative concept of the present invention, changes and modifications made to the embodiments described herein, or equivalent structures or equivalent process changes made using the contents of the present invention specification and drawings, directly or indirectly applying the above technical solutions to other related technical fields, are all included in the patent protection scope of the present invention.

Claims

1. The precise management system for diabetic foot based on artificial intelligence is characterized by: It includes the medical record management subsystem, the big data language model subsystem, the medical and nursing end, and the user end; The medical record management subsystem is used to obtain the new medical records of patients input by the medical staff in real time and record the time of the new medical records, wherein the new medical records of patients contain the words "diabetic foot"; The medical record management subsystem sends the patient's new medical record, other patients' historical medical records and matching degree instructions to the big data language model subsystem. The medical record management subsystem obtains the single patient matching result of the big data language model subsystem. The medical record management subsystem records the most matching medical record of each other patient. The medical record management subsystem sends the most matching medical record of each other patient and the next medical record and disease progression instruction of the patient to the big data language model subsystem, and the medical record management subsystem obtains the disease progression result returned by the big data language model subsystem; The medical record management subsystem obtains the next medical record corresponding to the positive result in the disease progression result, and uses a pre-set matching word for screening. The matching word is shallower. If shallower is matched, it means positive progress. The concentrated time period of the next medical record is counted, and the next medical record and feature control instructions of the concentrated time period are sent to the big data language model subsystem. The medical record management subsystem obtains the disease progression feature returned by the big data language model subsystem. The concentrated time period is the time period corresponding to the number of medical records being greater than a certain value; The medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, sends the disease progression characteristics and corresponding options to the user end after the sending time arrives, and the medical record management subsystem obtains the option result returned by the user end; The medical record management subsystem calculates the sending time according to the concentrated time period and the new medical record time, including: the medical record management subsystem uses the time with the largest number of medical records in the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time; or the medical record management subsystem uses the time in the middle of the concentrated time period as the interval time, and adds the interval time to the new medical record time as the sending time; the medical record management subsystem instructs the option result and the next medical record of the concentrated time period and the matching degree to the big data language model subsystem to obtain the option matching result; The medical record management subsystem sends the option result and the patient's new medical record to the medical care end when the option matching result does not match; The medical record management subsystem includes the steps of: The medical record management subsystem sends the doctor's order and summary control instruction corresponding to the next medical record in the concentrated time period to the big data language model subsystem. The big data language model subsystem summarizes the doctor's order and obtains a summarized doctor's order. The medical record management subsystem obtains the doctor's order summary returned by the big data language model subsystem and sends the doctor's order summary to the user end. The medical record management subsystem further comprises the following steps when the option matching result does not match: The medical record management subsystem sends the medical instructions corresponding to the next medical record in a concentrated time period to the medical care end.

2. The artificial intelligence-based precise management system for diabetic foot according to claim 1, characterized in that: The medical record management subsystem only sends a prompt message to the user end when the option matching results match.

3. The artificial intelligence-based precise management system for diabetic foot according to claim 1, characterized in that: The medical end includes a medical identity code, and the patient's new medical record corresponds to the medical identity code of the medical end. When the option matching result does not match, the medical record management subsystem sends the option result and the patient's new medical record to the medical end corresponding to the medical identity code of the patient's new medical record.

4. The artificial intelligence-based precise management system for diabetic foot according to claim 1, characterized in that: The medical record management subsystem further includes the following steps after sending the disease progression characteristics and corresponding options to the user end after the sending time arrives: If the medical record management subsystem does not obtain the option result returned by the user end after the predetermined waiting time has passed, the patient information in the new medical record of the patient is sent to the medical care end.

5. The artificial intelligence-based precise management system for diabetic foot according to claim 1, characterized in that: The medical record management subsystem sends the patient's new medical record, other historical patient medical records and matching degree instructions to the big data language model subsystem, including the following steps: The medical record management subsystem sends new patient medical records, historical medical records of other patients, and matching degree instructions to the big data language model subsystem within a preset time period.

6. The artificial intelligence-based precise management system for diabetic foot according to claim 1 is characterized by: The big data language model subsystem is a GPT big data language model.

Images