Blood tumor recognition early warning management system based on artificial intelligence

Abstract

The invention relates to the technical field of tumor examination management, and discloses a blood tumor identification early warning management system based on artificial intelligence, which comprises a switch for providing a hardware support layer, monitoring equipment and wireless equipment, the data storage module, the data acquisition module, the data analysis module, the examination early warning module, the management module, the clinical decision module and the user interface layer are used for examination management. And the data acquisition module is used for acquiring laboratory detection data, iconography examination data and symptom and sign data and feeding back the acquired data to the data analysis module. According to the method, the judgment of triggering early warning is performed on the patient by adopting the multi-modal detection indexes, so that potential patients with hematologic tumors can be identified at the first time, the misdiagnosis risk is reduced, and the effects of timely warning, early diagnosis and early treatment are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor examination management, and in particular to a blood tumor recognition and early warning management system based on artificial intelligence. Background Art

[0002] In solid tumors such as liver cancer and gastric cancer, laboratory indicator tumor series (AFP, CA72-4, etc.) can identify potential tumors early, attract patients' attention, and achieve the purpose of early detection, early diagnosis, and early treatment. However, as of now, there are no relevant early identification indicators in the blood tumor system: lymphoma patients often wait until the lymph nodes are swollen to a certain extent and undergo lymph node biopsy before a clear diagnosis can be made; myeloma patients have a misdiagnosis rate of more than 50%, and will first go to the nephrology, orthopedics and other departments for treatment due to the first symptoms, delaying the disease; some leukemia patients fail to pay attention to the abnormal cell morphology and proportion in the early blood routine test. Once diagnosed, the disease is often critical and the prognosis is poor. Obviously, early identification of blood tumors and timely triggering of early warning systems play a vital role in achieving early detection, early diagnosis, and early treatment of blood tumor patients.

[0003] With the rapid development of artificial intelligence (AI) technology, it has shown great potential in the medical field, especially in the diagnosis and management of blood tumors. The AI-based blood tumor identification and early warning management system is expected to effectively improve the timeliness and accuracy of diagnosis, optimize treatment plans, and enhance monitoring capabilities, thereby improving the quality of life and treatment effects of patients.

[0004] The existing patent discloses a blood system malignant tumor screening system for the target population (publication number CN111524594A), including a data acquisition module, which is used to obtain blood routine test sheet images that meet the target population, identify blood routine test indicators, age and gender in the blood routine test sheet images; screening prediction module, which is used to predict the blood system malignant tumor evaluation index prediction value of the corresponding person to be screened based on the blood system malignant tumor screening model. The technology disclosed in this patent lacks the ability to diagnose and warn patients of early lymphoma, multiple myeloma and leukemia, and it is difficult to quickly help patients and hematologists understand the various test indicators. Summary of the invention

[0005] The main technical problem solved by the present invention is to provide a blood tumor identification and early warning management system based on artificial intelligence, which solves the problem of high misdiagnosis rate in the early stage.

[0006] To solve the above technical problems, according to one aspect of the present invention, more specifically, a blood tumor identification and early warning management system based on artificial intelligence is characterized in that it includes a switch, a monitoring device and a wireless device for providing a hardware support layer, and a data storage module, a data acquisition module, a data analysis module, a review and early warning module, a management module, a clinical decision module and a user interface layer for inspection management;

[0007] The data acquisition module is used to collect medical indicators of the lymphoma inspection system and the ultrasound system, medical indicators of the multiple myeloma inspection system, and medical indicators of the leukemia inspection system;

[0008] The data analysis module is used to determine whether the patient has a risk of lymphoma in the initial test according to the erythrocyte sedimentation rate, β2 microglobulin, lactate dehydrogenase in the test system, and the abnormal increase in lymph node size in the ultrasound system; determine whether the patient has a risk of multiple myeloma in the initial test according to the abnormal increase in hemoglobin, serum calcium, globulin, erythrocyte sedimentation rate in the test system, and whether creatinine obviously exceeds the range; determine whether the patient has a risk of leukemia in the initial test according to the abnormal increase in white blood cells, hemoglobin, platelets, and erythrocyte sedimentation rate in the test system;

[0009] The review and warning module issues a warning of the presence of lymphoma, multiple myeloma or leukemia to the patient and the hematologist based on the analysis results of the data analysis module, and provides data information of abnormal indicators to the patient and the hematologist;

[0010] The user interface layer includes a doctor-side interface and a patient-side interface.

[0011] Furthermore, the data analysis module determines the probability coefficient of lymphoma risk in the patient's initial test based on the abnormal increments of erythrocyte sedimentation rate, β2 microglobulin and lactate dehydrogenase in the test system and the abnormal increment of lymph node size in the ultrasound system, then:

[0012] ;

[0013] In the formula, represents the probability coefficient of the patient having lymphoma in the initial test, Indicates an abnormal increase in the erythrocyte sedimentation rate in serum. Indicates an abnormal increase in serum β2 microglobulin. Indicates an abnormal increase in serum lactate dehydrogenase. Indicates abnormal increase in lymph node size in ultrasound system;

[0014] when When the patient is initially tested, it means that the patient has symptoms of lymphoma. At this time, the review and warning module needs to issue a lymphoma warning information to the patient and the hematologist.

[0015] Furthermore, the data analysis module determines the probability coefficient of the presence of multiple myeloma in the patient's initial test based on the abnormal increments of hemoglobin, serum calcium, globulin, and erythrocyte sedimentation rate in the test system, then:

[0016] ;

[0017] In the formula, represents the probability coefficient of the patient having multiple myeloma in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in serum calcium in the blood serum. Indicates an abnormal increase in globulin in serum. Indicates an abnormal increase in serum creatinine;

[0018] when When the patient is initially tested, it means that the patient has symptoms of multiple myeloma. At this time, the review and warning module needs to issue a warning information about the presence of multiple myeloma to the patient and the hematologist.

[0019] Furthermore, the data analysis module determines the probability coefficient of the presence of leukemia symptoms in the patient's initial test based on the abnormal increments of white blood cells, hemoglobin, platelets, and erythrocyte sedimentation rate in the test system, then:

[0020] ;

[0021] In the formula, represents the probability coefficient of the patient having leukemia in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in white blood cells in serum. Indicates an abnormal increase in platelets in serum. Indicates an abnormal increase in abnormal cells in the serum;

[0022] when When , it means that the patient in the initial test has symptoms of leukemia. At this time, the review and warning module needs to issue leukemia warning information to the patient and the hematologist.

[0023] Furthermore, the clinical decision module is used to automatically generate diagnostic suggestions based on the warning information provided by the review and warning module, and ultimately the attending physician determines whether the judgment of the review and warning module is correct after the attending physician verifies and confirms it.

[0024] Furthermore, the user interface layer is used to display clinical decision recommendations, review warning status, indicator status of the inspection system, and indicator status of the ultrasound system.

[0025] Furthermore, the user interface layer is also used in the data visualization report generation module and the health education and advice module;

[0026] The data visualization report generation module is used to generate a medical report and pathological examination data of the patient, and the medical report and pathological examination data are archived for review and retrieval;

[0027] The health education and advice module automatically searches the cloud big data for the precautions required for the condition based on the condition determined by the clinical decision module.

[0028] Beneficial effects:

[0029] 1. The present invention can determine whether it is necessary to issue an early warning to the attending physician based on the data indicators of the testing system and the indicators of the ultrasound system. This can assist the attending physician in quickly identifying the patient's potential blood tumor risk and improve the efficiency and accuracy of medical treatment.

[0030] 2. The present invention provides interfaces to both the attending physician and the patient, and can provide the patient with automatically retrieved health education and medical dietary advice in a targeted manner, which can assist the attending physician in quickly understanding the patient's medical condition and improve the efficiency and accuracy of medical treatment.

[0031] 3. The present invention uses multimodal detection indicators to determine whether patients have triggered early warnings, which helps to identify potential patients with blood tumors in the first place, reduce the risk of misdiagnosis, and play a role in early detection, early diagnosis, and early treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flow chart of the present invention;

[0033] Figure 2 It is a schematic diagram of the structure of the present invention. DETAILED DESCRIPTION

[0034] In order to make the technical solution of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0035] Example 1

[0036] like Figure 1 - Figure 2As shown, according to one aspect of the present invention, a blood tumor recognition and early warning management system based on artificial intelligence is provided, including a switch, a monitoring device and a wireless device for providing a hardware support layer, and a data storage module, a data acquisition module, a data analysis module, a review and early warning module, a management module, a clinical decision module and a user interface layer for inspection management; the data acquisition module is used to collect medical indicators of the inspection system and the ultrasound system of lymphoma, the inspection system medical indicators of multiple myeloma, and the inspection system medical indicators of leukemia; the data analysis module is used to collect the medical indicators of the inspection system according to the erythrocyte sedimentation rate, β2 microglobulin, lactate dehydrogenase in the inspection system, and according to the medical indicators of the ultrasound system The system can determine whether the patient has a risk of lymphoma in the initial examination based on the abnormal increase in lymph node size; determine whether the patient has a risk of multiple myeloma in the initial examination based on the abnormal increase in hemoglobin, serum calcium, globulin, and erythrocyte sedimentation rate in the testing system, and whether the creatinine is significantly above the range; determine whether the patient has a risk of leukemia in the initial examination based on the abnormal increase in white blood cells, hemoglobin, platelets, and erythrocyte sedimentation rate in the testing system; and determine whether an early warning needs to be issued to the attending physician based on the data indicators of the testing system, data indicators, and indicators of the ultrasound system, which can assist the attending physician in quickly understanding the patient's condition and improve the efficiency and accuracy of medical treatment.

[0037] The review and early warning module issues early warnings of lymphoma, multiple myeloma or leukemia to patients and hematologists based on the analysis results of the data analysis module, and provides data information of abnormal indicators to patients and hematologists; the user interface layer includes the doctor-side interface and the patient-side interface. By using multimodal detection indicators to determine whether patients trigger early warnings, it helps to identify potential patients with blood tumors in the first place, reduce the risk of misdiagnosis, and play a role in timely warning, early diagnosis, and early treatment.

[0038] Example 2

[0039] The data analysis module determines the probability coefficient of lymphoma risk in the patient's initial test based on the abnormal increments of erythrocyte sedimentation rate, β2 microglobulin and lactate dehydrogenase in the test system and the abnormal increment of lymph node size in the ultrasound system, then:

[0040] ;

[0041] In the formula, represents the probability coefficient of the patient having lymphoma in the initial test, Indicates an abnormal increase in the erythrocyte sedimentation rate in serum. Indicates an abnormal increase in serum β2 microglobulin. Indicates an abnormal increase in serum lactate dehydrogenase. Indicates abnormal increase in lymph node size in ultrasound system;

[0042] when When the patient is initially tested, it means that there is a potential risk of lymphoma. At this time, the review and warning module needs to issue a lymphoma warning information to the patient and the hematologist.

[0043] Among them, patient X was monitored and warned of lymphoma. (Abnormal increase = the mass of red blood cells exceeding the normal range ÷ the mass of red blood cells within the normal range × 100%, if it is within the normal range, take 0.1%).

[0044] Abnormal increase of β2 microglobulin in patient X's serum (Abnormal increase = the mass of β2 microglobulin exceeding the normal range ÷ the mass of β2 microglobulin within the normal range × 100%, if it is within the normal range, 0.1% is taken).

[0045] Abnormal increase of lactate dehydrogenase in patient X's serum (Abnormal increase = the mass of lactate dehydrogenase exceeding the normal range ÷ the mass of lactate dehydrogenase within the normal range × 100%, if it is within the normal range, take 0.1%).

[0046] Abnormal lymph node size increment in the patient's X-ray ultrasound system (Abnormal increase = the volume of lymph nodes exceeding the normal range ÷ the volume of lymph nodes within the normal range × 100%, and 0.1% is taken if it is within the normal range), then:

[0047] ;

[0048] From the above calculation, it can be known that patient X has lymphoma characteristics in the initial test. At this time, the review warning module needs to issue lymphoma warning information to the patient and the hematologist. The lymphoma warning information data is shown in Table 1.

[0049] Table 1: Lymphoma early warning information statistics

[0050]

[0051] From the data in Table 1 above, we can see that when the sample data implemented increases infinitely, there will be an obvious warning dividing line for indicator C, and this dividing line is 64%.

[0052] Then, when When the patient's blood test shows lymphoma characteristics, it is necessary to issue an early warning of the presence of lymphoma to the patient or hematologist.

[0053] Example 3

[0054] The data analysis module determines the probability coefficient of the presence of multiple myeloma in the patient's initial test based on the abnormal increments of hemoglobin, serum calcium, globulin, and erythrocyte sedimentation rate in the test system, so:

[0055] ;

[0056] In the formula, represents the probability coefficient of the patient having multiple myeloma in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in serum calcium in the blood serum. Indicates an abnormal increase in globulin in serum. Indicates an abnormal increase in serum creatinine;

[0057] when When the patient is initially tested, it means that the patient has symptoms of multiple myeloma. At this time, the review and warning module needs to issue a warning information about the presence of multiple myeloma to the patient and the hematologist.

[0058] In addition, abnormal increase = mass exceeding the normal range ÷ mass within the normal range × 100%, and if it is within the normal range, it is taken as 0.1%.

[0059] Among them, a preliminary test warning for multiple myeloma was performed on patient Y. The abnormal increase in serum hemoglobin was taken , the abnormal increase in serum calcium in serum , abnormal increase in serum globulin , abnormal increase in serum creatinine , then we have:

[0060] ;

[0061] From the above calculation, we can know that the probability coefficient of multiple myeloma in the initial test of patient Y is Then the patient X in the initial test has the characteristics of multiple myeloma. At this time, the review and warning module needs to issue a warning information of multiple myeloma to the patient and the hematologist. The warning information data of multiple myeloma is shown in Table 2.

[0062] Table 2: Multiple myeloma early warning information statistics

[0063]

[0064] It can be seen from the data in Table 2 above that when the sample data implemented increases infinitely, there will be an obvious warning dividing line for the judgment indicator G, and this dividing line is 71%.

[0065] Then, when When the patient's blood test shows the characteristics of multiple myeloma, it is necessary to issue an early warning of the presence of multiple myeloma to the patient or hematologist.

[0066] Example 4

[0067] The data analysis module determines the probability coefficient of leukemia in the patient's initial test based on the abnormal increments of white blood cells, hemoglobin, platelets, and erythrocyte sedimentation rate in the test system, so:

[0068] ;

[0069] In the formula, represents the probability coefficient of the patient having leukemia in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in white blood cells in serum. Indicates an abnormal increase in platelets in serum. Indicates an abnormal increase in abnormal cells in the serum;

[0070] when When , it means that the patient in the initial test has symptoms of leukemia. At this time, the review and warning module needs to issue leukemia warning information to the patient and the hematologist.

[0071] Among them, a preliminary leukemia test warning was conducted on patient Z. , abnormal increase of white blood cells in serum , abnormal increase of platelets in serum , the abnormal increase of abnormal cells in serum , then we have:

[0072] ;

[0073] From the above calculation, it can be known that the patient Z in the initial test has the characteristics of leukemia. At this time, the review and warning module needs to issue leukemia warning information to the patient and the hematologist. The leukemia warning information data is shown in Table 3.

[0074] Table 3: Leukemia early warning information statistics

[0075]

[0076] It can be seen from the data in Table 3 above that when the sample data implemented increases infinitely, there will be an obvious warning dividing line for the judgment indicator M, and this dividing line is 67%.

[0077] Then, when When the patient's blood test shows the characteristics of leukemia, it is necessary to issue an early warning of the presence of leukemia to the patient or hematologist.

[0078] Example 5

[0079] The clinical decision-making module is used to automatically generate diagnostic suggestions based on the warning information provided by the review and warning module, and the attending physician will ultimately determine whether the judgment of the review and warning module is correct. The user interface layer is used to display clinical decision suggestions, review and warning status, indicator status of the inspection system, and indicator status of the ultrasound system. The user interface layer is also used for the data visualization report generation module and the health education and advice module; the data visualization report generation module is used to generate the patient's medical report and pathological examination data, which are used for archiving for review and retrieval; the health education and advice module automatically retrieves the necessary precautions for the condition from the cloud big data based on the condition determined by the clinical decision module.

[0080] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A blood tumor recognition and early warning management system based on artificial intelligence, characterized in that: It includes switches, monitoring equipment and wireless equipment for providing hardware support layer, as well as data storage module, data acquisition module, data analysis module, review and early warning module, management module, clinical decision module and user interface layer for inspection management; The data acquisition module is used to collect medical indicators of the lymphoma inspection system and the ultrasound system, medical indicators of the multiple myeloma inspection system, and medical indicators of the leukemia inspection system; The data analysis module is used to determine whether there is an indication of lymphoma in the initial test of the patient based on the erythrocyte sedimentation rate, beta 2 microglobulin, lactate dehydrogenase in the test system and the abnormal increase in lymph node size in the ultrasound system; Determine whether the patient has multiple myeloma indications in the initial test based on abnormal increases in hemoglobin, serum calcium, globulin, and creatinine in the test system; Determine whether the patient has leukemia indications in the initial test based on the abnormal increase of white blood cells, hemoglobin, platelets, and abnormal cells in the test system; The review and warning module issues a blood tumor warning of lymphoma, multiple myeloma or leukemia to the patient and the hematologist based on the analysis results of the data analysis module, and provides data information of abnormal indicators to the patient and the hematologist; The user interface layer includes a doctor-side interface and a patient-side interface.

2. The artificial intelligence-based blood tumor recognition and early warning management system according to claim 1, characterized in that: The data analysis module determines the probability coefficient of the presence of lymphoma in the patient's initial test based on the abnormal increments of erythrocyte sedimentation rate, β2 microglobulin and lactate dehydrogenase in the test system and the abnormal increment of lymph node size in the ultrasound system, then: ； In the formula, represents the probability coefficient of the patient having lymphoma in the initial test, Indicates an abnormal increase in the erythrocyte sedimentation rate in serum. Indicates an abnormal increase in serum β2 microglobulin. Indicates an abnormal increase in serum lactate dehydrogenase. Indicates abnormal increase in lymph node size in ultrasound system; when When the patient is initially tested, it means that the patient has symptoms of lymphoma. At this time, the review and warning module needs to issue a lymphoma warning information to the patient and the hematologist.

3. The blood tumor recognition and early warning management system based on artificial intelligence according to claim 1 is characterized by: The data analysis module determines the probability coefficient of the presence of multiple myeloma in the patient's initial test based on the abnormal increments of hemoglobin, serum calcium, globulin, and creatinine in the test system, then: ； In the formula, represents the probability coefficient of the patient having multiple myeloma in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in serum calcium in the blood serum. Indicates an abnormal increase in globulin in serum. Indicates an abnormal increase in serum creatinine; when When the patient is initially tested, it means that the patient has symptoms of multiple myeloma. At this time, the review and warning module needs to issue a warning information about the presence of multiple myeloma to the patient and the hematologist.

4. The artificial intelligence-based blood tumor recognition and early warning management system according to claim 1, characterized in that: The data analysis module determines the probability coefficient of leukemia in the initial test of the patient based on the abnormal increments of white blood cells, hemoglobin, platelets, and abnormal cells in the test system, then: ； In the formula, represents the probability coefficient of the patient having leukemia in the initial test, Indicates an abnormal increase in hemoglobin in serum. Indicates an abnormal increase in white blood cells in serum. Indicates an abnormal increase in platelets in serum. Indicates an abnormal increase in abnormal cells in the serum; when When , it means that the patient in the initial test has symptoms of leukemia. At this time, the review and warning module needs to issue leukemia warning information to the patient and the hematologist.

5. The blood tumor recognition and early warning management system based on artificial intelligence according to claim 1 is characterized by: The clinical decision module is used to automatically generate diagnostic suggestions based on the warning information provided by the review and warning module, and ultimately the attending physician determines whether the judgment of the review and warning module is correct after the attending physician verifies and confirms it.

6. The artificial intelligence-based blood tumor recognition and early warning management system according to claim 5 is characterized by: The user interface layer is used to display clinical decision recommendations, review warning status, indicator status of the inspection system, and indicator status of the ultrasound system.

7. The artificial intelligence-based blood tumor recognition and early warning management system according to claim 5 is characterized by: The user interface layer is also used in the data visualization report generation module and the health education and advice module; The data visualization report generation module is used to generate a medical report and pathological examination data of the patient, and the medical report and pathological examination data are archived for review and retrieval; The health education and advice module automatically searches the cloud big data for the precautions required for the condition based on the condition determined by the clinical decision module.

Images