Platform and device for personalized management of bone marrow suppression based on self-report outcome of patient

Abstract

The invention relates to the field of intelligent management, in particular to a bone marrow suppression personalized management platform and equipment based on self-report outcome of a patient. The functional service layer is used for providing functional services for a medical care terminal and a patient terminal, and the medical care terminal uploads resource information, uploads questionnaires and records an overall chemotherapy appointment list; the patient end can browse resource information, fill in questionnaires and upload self-inspection data; and storing the functional service layer data. And the interaction processing layer is used for realizing a multi-dimensional interaction function, and the interaction function comprises patient and patient health interaction, nurse and patient tracking follow-up visit, man-machine intelligent suggestion, doctor and patient question and answer and storage of data of the interaction processing layer. And the intelligent control layer performs intelligent analysis based on the data of the function service layer, the data of the interaction processing layer and the clinical data of the patient in the hospital database, and provides alarm, reminding and appointment functions for the patient terminal. Therefore, the problems that in the tumor chemotherapy intermission period, a patient is prone to CIM and infection, and the re-admission rate is high are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management, and in particular to a personalized management platform and device for myelosuppression based on patient self-reported outcomes. Background Art

[0002] Chemotherapy is the main treatment for cancer patients and can effectively inhibit tumor recurrence and metastasis. Chemotherapy-induced myelosuppression (CIM) is the most common and most serious adverse reaction, usually manifested as a decrease in blood indicators. CIM is not only a key factor affecting the course and dosage of treatment, but also increases the physiological and psychological burden, infection and risk of readmission, and even endangers the patient's life.

[0003] Studies have shown that 79% of patients have received CIM treatment, and up to 88% of patients believe that it seriously affects their quality of life. In the day-care or outpatient chemotherapy model, the time spent between patients and medical staff is reduced, the patients' compliance with routine blood tests during chemotherapy intervals is low, and its management involves individual differences and dynamic changes, resulting in an incidence rate as high as 67.16%. Even in patients with certain chemotherapy regimens, the incidence of CIM during the first chemotherapy is as high as 90%, and 42% of patients have at least one emergency visit or hospitalization. Therefore, it is crucial to closely monitor the patient's routine blood values ​​during chemotherapy intervals and obtain patient-reported data in real time.

[0004] At present, most studies focus on preventive drugs for CIM, and fewer studies focus on improving patients' compliance with blood routine examinations during the chemotherapy interval to ensure the safety of chemotherapy. Therefore, improving the status of patients' self-management of CIM has become an urgent problem to be solved. Summary of the invention

[0005] In order to solve the problem that patients are prone to CIM, infection and high readmission rate during the interval period of tumor chemotherapy, the present invention provides a personalized management platform for bone marrow suppression based on patient self-reported outcomes.

[0006] In a first aspect, the present invention provides a personalized management platform for myelosuppression based on patient self-reported outcomes, comprising:

[0007] Functional service layer: used to provide functional services for medical and nursing ends and patient ends. The medical and nursing ends upload resource information, upload questionnaires, and record the overall chemotherapy appointment list through the functional service layer. The patient ends browse resource information, fill in questionnaires, and upload self-examination data through the functional service layer.

[0008] The functional service layer is also used to store functional service layer data, which includes: patient resource information browsing data, questionnaire data, chemotherapy record data, and self-examination data;

[0009] Interaction processing layer: used to realize multi-dimensional interaction functions, including: patient-patient health interaction, nurse-patient tracking and follow-up, human-computer intelligent suggestions, and doctor-patient Q&A;

[0010] The interactive processing layer is also used to store interactive processing layer data, which includes: patient-patient interaction data, nurse-patient follow-up data, human-computer interaction data, and doctor-patient question and answer data;

[0011] Intelligent control layer: performs intelligent analysis based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database, and provides alarms, reminders and appointment functions to the patient side according to the analysis results.

[0012] In some embodiments, the medical care end uploads resource information, uploads questionnaires, and records the overall chemotherapy appointment list through the functional service layer, including:

[0013] Medical users upload resource information to the resource information module through the functional service layer. The resource information includes: tumor health education content in the form of text, pictures, audio, video, tumor disease knowledge, symptom management, coping skills, and social resources;

[0014] Medical users edit the content and format of the questionnaire in the questionnaire module and publish it in the questionnaire module;

[0015] The chemotherapy appointment list module presets the patient chemotherapy schedule, and the medical staff records the chemotherapy information in the chemotherapy appointment list module based on whether the patient undergoes chemotherapy on time;

[0016] The patient at the patient end uploads self-examination data through the self-examination data module of the functional service layer.

[0017] In some embodiments, the intelligent control layer includes:

[0018] Intelligent analysis module: Based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database, data is collated, cleaned and analyzed, and a prediction model for the development of CIM symptoms during the interval period of tumor chemotherapy is constructed based on ePROMs, and personalized suggestions and opinions are given according to the prediction results;

[0019] Intelligent alarm module: based on the prediction results of the intelligent analysis module and the preset alarm level, contact and guide the patient for follow-up treatment;

[0020] Intelligent appointment module: calculates the patient's follow-up examination and treatment plan based on the functional service layer data, the interactive processing layer data and the patient's clinical data in the hospital database and informs the patient and medical staff through the interactive processing module;

[0021] Intelligent reminder module: reminds patients and medical staff of examination time based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database.

[0022] In some embodiments, the preset alarm levels are divided into first-level alarm, second-level alarm, third-level alarm and fourth-level alarm;

[0023] The first level alarm is a low-risk warning, indicating that the patient has no abnormal symptoms, and the alarm color is green;

[0024] The second level alarm is a low-risk warning, indicating that the patient's symptoms are mild, and the alarm color is green;

[0025] The third level alarm is a medium-risk warning, indicating that the patient has mild or non-life-threatening symptoms, and the alarm color is yellow;

[0026] The fourth level alarm is a high-risk warning, indicating that the patient has serious symptoms or is even life-threatening, and the alarm color is red.

[0027] In some embodiments, in response to a third level alarm, medical personnel are required to contact the patient within four hours;

[0028] In response to a Level 4 alert, medical staff are required to contact the patient and conduct a clinical assessment within one hour.

[0029] In some embodiments, the personalized recommendations and suggestions include: guiding the patient on activities, rest, diet, and medication.

[0030] In some embodiments, the prediction model for the development of CIM symptoms during the interval period of tumor chemotherapy based on ePROMs is constructed, and its input parameters specifically include:

[0031] Demographic characteristics data: including mean age, median, age range, height, weight, and body mass index;

[0032] Clinical characteristic data: including tumor staging data, chemotherapy regimen data, the proportion of patients with bone metastasis and its distribution characteristics, and comorbid disease data;

[0033] Routine blood test data: including white blood cell count, neutrophil count, red blood cell count, platelet count, hemoglobin concentration, and the changing trend of routine blood test results before and after chemotherapy;

[0034] Symptom characteristic data: including the frequency and severity of common symptoms, and symptom development trends;

[0035] Quality of life data: including the patient's quality of life score, the relationship between quality of life and chemotherapy cycles, bone metastasis, and symptom severity;

[0036] Group characteristic data: including characteristics of patients grouped according to CIM symptom development pattern, chemotherapy cycle response, etc., and intra-group and inter-group difference data;

[0037] Correlation characteristic data: including the correlation data between the patient's demographic characteristics, clinical characteristics, chemotherapy regimen, blood routine test results, symptoms and quality of life scores, and key correlation data;

[0038] Time dynamic characteristic data: including the trend of symptom development and blood routine changes at different time points during the chemotherapy cycle, and symptom data at periodic or specific time points;

[0039] Prognostic feature data: including survival estimates under different treatment options and the relationship between specific CIM symptoms and prognosis.

[0040] In some embodiments, the method of data sorting, cleaning and analysis includes:

[0041] Descriptive statistical analysis: used to understand the basic characteristics of the data;

[0042] Correlation analysis: used to analyze the relationship between independent variables and dependent variables;

[0043] Regression analysis: including linear regression analysis: used to quantify the relationship between independent variables and dependent variables; logistic regression analysis: used to analyze binary data; and multiple regression analysis: used to analyze situations with multiple dependent variables;

[0044] Time series analysis: used to analyze and predict changes in patient symptoms and blood routine results at different time points during the chemotherapy cycle;

[0045] Cluster analysis: used to divide patient groups with similar CIM development characteristics into corresponding groups;

[0046] Principal component analysis: used to reduce the dimensionality of data to reveal the main variation patterns of the data and simplify the analysis of multidimensional data;

[0047] Survival analysis: evaluate the survival time of patients under different treatment options or the time when CIM symptoms appear;

[0048] Factor Analysis: Used to identify latent structures in the data.

[0049] In some embodiments, the doctor-patient Q&A specifically involves the patient and medical staff consulting and answering questions about problems encountered by the patient;

[0050] The patient-patient healthy interaction refers specifically to patients sharing experiences, talking to each other, encouraging each other and supporting each other;

[0051] The nurse-patient follow-up specifically refers to the medical staff providing one-on-one guidance to the patient and conducting regular online follow-up, the follow-up content mainly covers the patient's recent life and medical treatment;

[0052] The human-machine intelligent suggestion specifically refers to the platform giving targeted content push based on the patient's clinical data and the functional service layer data, based on the patient's questions raised, the information browsed, and the patient's status.

[0053] In a second aspect, the present invention proposes an electronic device on which the personalized management platform for bone marrow suppression based on patient self-reported outcomes described in the first aspect is deployed.

[0054] In order to solve the problem that patients are prone to CIM, infection and high readmission rate during the interval of tumor chemotherapy, the present invention has the following advantages:

[0055] The personalized management platform for bone marrow suppression based on patient self-reported outcomes can enhance the self-reported outcomes of tumor chemotherapy patients and promote patient self-management; it can provide comprehensive, precise, and personalized care for the adverse reactions of bone marrow suppression in this special group of chemotherapy patients. It can realize the integrated, intelligent, and standardized management of reminders to recheck blood routine, test reporting, and CIM early warning alarms, create a scientific and efficient CIM management model, improve the mastery of disease knowledge and compliance with rechecking blood routine during chemotherapy intervals, reduce the occurrence of CIM, reduce patient infection and readmission rates, thereby reducing the economic burden on patients' families, improving the work efficiency of medical staff and the quality of life of patients, and solving the current unfavorable situation of CIM self-management. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of the structural framework of the personalized management platform is shown;

[0057] Figure 2 A schematic diagram showing the functions of the medical and patient ends;

[0058] Figure 3 A block diagram of the self-efficacy theory is shown;

[0059] Figure 4 A processing flow chart of the fourth level alarm is shown;

[0060] Figure 5 A diagram showing personalized suggestions and opinions from the intelligent analysis module. DETAILED DESCRIPTION

[0061] The present disclosure will now be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and thus implement the present disclosure, rather than implying any limitation on the scope of the present disclosure.

[0062] As used herein, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to". The term "based on" is to be interpreted as "based at least in part on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment". The term "another embodiment" is to be interpreted as "at least one other embodiment". The orientation or position relationship indicated by the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inner", "outer", "vertical", "horizontal", "lateral", "longitudinal" and the like is based on the orientation or position relationship shown in the accompanying drawings. These terms are mainly for better describing the present application and its embodiments, and are not used to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation. In addition, in addition to being used to indicate an orientation or position relationship, some of the above terms may also be used to indicate other meanings, such as the term "upper" may also be used to indicate a certain dependency or connection relationship in some cases. For those of ordinary skill in the art, the specific meanings of these terms in this application can be understood according to the specific circumstances. In addition, the terms "install", "set", "provided with", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be an internal connection between two devices, elements or components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to the specific circumstances. In addition, the terms "first", "second", etc. are mainly used to distinguish different devices, elements or components (the specific types and structures may be the same or different), and are not used to indicate or imply the relative importance and quantity of the indicated devices, elements or components. Unless otherwise specified, "plurality" means two or more.

[0063] In a first aspect, this embodiment discloses a personalized management platform for myelosuppression based on patient self-reported outcomes, including:

[0064] Functional service layer: used to provide functional services for medical and nursing ends and patient ends. The medical and nursing ends upload resource information, upload questionnaires, and record the overall chemotherapy appointment list through the functional service layer. The patient ends browse resource information, fill in questionnaires, and upload self-examination data through the functional service layer.

[0065] The functional service layer is also used to store functional service layer data, which includes: patient resource information browsing data, questionnaire data, chemotherapy record data, and self-examination data;

[0066] Interaction processing layer: used to realize multi-dimensional interaction functions, including: patient-patient health interaction, nurse-patient tracking and follow-up, human-computer intelligent suggestions, and doctor-patient Q&A;

[0067] The interactive processing layer is also used to store interactive processing layer data, which includes: patient-patient interaction data, nurse-patient follow-up data, human-computer interaction data, and doctor-patient question and answer data;

[0068] Intelligent control layer: performs intelligent analysis based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database, and provides alarms, reminders and appointment functions to the patient side according to the analysis results.

[0069] In this example, a personalized management platform for myelosuppression based on patient self-reported outcomes was proposed, e.g. Figure 1 , Figure 2 As shown, it includes a functional service layer, an interactive processing layer, and an intelligent control layer. In the functional service layer, medical staff can upload resource information, questionnaires, and overall chemotherapy appointment lists through different functional service modules; patients on the patient side can browse resource information, fill out questionnaires, and upload self-examination data through different functional service modules. In the interactive processing layer, multi-dimensional interactions can be achieved, including patient-patient health interactions, nurse-patient tracking and follow-up, human-computer intelligent suggestions, and doctor-patient Q&A. In the intelligent control layer, the intelligent control layer performs intelligent analysis based on the functional service layer data stored in the functional service layer, the interactive processing layer data stored in the interactive processing layer, and the patient clinical data in the externally linked hospital database, and provides patients with intelligent reminders such as alarms, reminders, and appointments based on the analysis results.

[0070] Specifically, Figure 3 As shown in the figure, in the self-efficacy theory, self-efficacy is affected by four sources of information: direct experience, indirect experience, verbal persuasion, and emotional arousal. Direct experience refers to the experience gained from personal experience; indirect experience refers to the role model effect, which increases one's confidence by seeing others' achievements in similar environments; verbal persuasion refers to the use of others' encouragement, suggestions, and other words to enhance one's belief and confidence in achieving success; emotional arousal refers to the physiological conditions or the individual's perception of these physiological conditions, which will affect people's emotions, thereby enhancing the patient's confidence and ability to overcome the disease by enhancing the patient's self-efficacy.

[0071] Specifically, browsing the resource information module of the functional service layer and human-computer interaction of the interactive processing layer can help patients further understand the symptoms and achieve the positive effects of direct experience and verbal persuasion in the self-efficacy theory. Through the interactive processing layer of doctor-patient Q&A, nurse-patient follow-up, and patient-patient health interaction, the positive effects of indirect experience, emotional awakening, and verbal persuasion in the self-efficacy theory can be achieved.

[0072] Specifically, the platform combines ePROMs technology and self-efficacy theory to achieve integrated, intelligent, standardized and personalized management of reminding chemotherapy patients to review their blood routine, test reporting and chemotherapy-induced bone marrow suppression early warning alarm, dynamically monitor the changes in patients' blood routine during the entire chemotherapy period, and automatically generate a visualization map of blood routine results to facilitate observation of individualized treatment and changes in blood routine. It provides intelligent and customized management strategies for CIM of cancer chemotherapy patients, helps meet patients' precise and personalized nursing needs, improves patients' quality of life, supports clinical decision-making, and ensures its availability and effectiveness in actual clinical applications.

[0073] In some embodiments, the medical care end uploads resource information, uploads questionnaires, and records the overall chemotherapy appointment list through the functional service layer, including:

[0074] Medical users upload resource information to the resource information module through the functional service layer. The resource information includes: tumor health education content in the form of text, pictures, audio, video, tumor disease knowledge, symptom management, coping skills, and social resources;

[0075] Medical users edit the content and format of the questionnaire in the questionnaire module and publish it in the questionnaire module;

[0076] The chemotherapy appointment list module presets the patient chemotherapy schedule, and the medical staff records the chemotherapy information in the chemotherapy appointment list module based on whether the patient undergoes chemotherapy on time;

[0077] The patient at the patient end uploads self-examination data through the self-examination data module of the functional service layer.

[0078] In this embodiment, the functional service layer includes a resource information module, a questionnaire module, and a chemotherapy appointment list module. The users of the medical and nursing end, that is, the medical staff, upload the tumor health education content through the resource information module to make it into text, pictures, audio, video, tumor disease knowledge, symptom management, coping skills, social resources and other related information. Patients can browse the relevant information in this module to further understand the scientific knowledge of their own diseases, disease management knowledge, and symptom self-care suggestions.

[0079] Specifically, through the questionnaire module, medical staff can edit the corresponding questionnaire content according to the patient's symptoms, and according to the questionnaire filled out by the patient, the system will promptly process the patient's input results intelligently and feedback information to the patient and the medical staff. The information is mainly in the form of text messages or jump calls. When there are serious symptoms or high warnings, both the patient and the medical staff will display alarms and text message reminders of corresponding colors. For example: some of the patient's answers in the questionnaire indicate a more severe condition, but the patient is unaware of it, and the platform will remind or alarm the patient to take measures.

[0080] Specifically, medical staff monitor the chemotherapy date and whether the patient is undergoing chemotherapy through the chemotherapy appointment list module, pay close attention to the patient's treatment according to the chemotherapy situation, and keep records. Patients can check their chemotherapy arrangements through the chemotherapy appointment list and make preparations in advance. For example, if medical staff monitors that a patient has not undergone chemotherapy on time through the chemotherapy appointment list module, they will contact the patient to understand the actual situation of the patient not undergoing chemotherapy, and make specific arrangements based on the specific situation.

[0081] Specifically, patients can upload their own test records or health records and other information through the self-examination data module. For example: patients upload photos of blood routine results, the platform system can identify the photos, extract key indicators such as white blood cells, neutrophils, hemoglobin and platelets, store these data and send them to the intelligent control layer for analysis. Another example: patients upload their own health records, which can include data such as body temperature, stomatitis, dizziness, skin condition, stool color, etc.; they can also upload chemotherapy medication records, including chemotherapy drug regimens, dosages, usage, frequency, chemotherapy regimens and dosages, etc. The functional service layer will extract these data and send them to the intelligent control layer for analysis.

[0082] In some embodiments, the intelligent control layer includes:

[0083] Intelligent analysis module: Based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database, data is collated, cleaned and analyzed, and a prediction model for the development of CIM symptoms during the interval period of tumor chemotherapy is constructed based on ePROMs, and personalized suggestions and opinions are given according to the prediction results;

[0084] Intelligent alarm module: based on the prediction results of the intelligent analysis module and the preset alarm level, contact and guide the patient for follow-up treatment;

[0085] Intelligent appointment module: calculates the patient's follow-up examination and treatment plan based on the functional service layer data, the interactive processing layer data and the patient's clinical data in the hospital database and informs the patient and medical staff through the interactive processing module;

[0086] Intelligent reminder module: reminds patients and medical staff of examination time based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database.

[0087] In this embodiment, the intelligent control layer includes an intelligent analysis module, an intelligent alarm module, an intelligent reservation module and an intelligent reminder module.

[0088] Specifically, the intelligent analysis module coordinates the patient's clinical data, platform function service layer data and interactive processing layer data, and organizes, clarifies and analyzes the data. It also builds a prediction model for the development of CIM symptoms during the interval of tumor chemotherapy based on ePROMs (electronic patient-reported outcome measures), and gives personalized suggestions and opinions based on the prediction results. The intelligent alarm module provides the patient with the corresponding level of alarm and implements corresponding measures based on the prediction results of the intelligent analysis module and the preset alarm level.

[0089] Specifically, when selecting the neural network architecture, this application constructs a prediction model for the development of CIM symptoms based on classic machine learning algorithms such as nomograms, artificial neural networks, decision trees, support vector machines, random forests, and extreme gradient boosting. By evaluating the area under the receiver operating characteristic curve (AUROC), calibration curve, decision curve analysis (DCA) and a series of methods to test the accuracy, discrimination and clinical practicality of the model, it is finally determined to build a CIM symptom development prediction model based on the LSTM recurrent neural network architecture.

[0090] Specifically, the CIM symptom development prediction model is trained based on the LSTM (Long Short-Term Memory Networks) recurrent neural network architecture, and coordinates the patient's clinical data, platform function service layer data, and interactive processing layer data as input basic parameters. The data is organized, clarified, and analyzed to obtain input parameters, which are finally input into the CIM symptom development prediction model to obtain prediction results, and corresponding processing, personalized suggestions, and opinions are performed based on the prediction results.

[0091] Specifically, the intelligent appointment module calculates the patient's subsequent examination and treatment plan based on the patient's functional service layer data, interactive processing layer data and the patient's clinical data in the hospital database. For example, based on the patient's first chemotherapy time and chemotherapy cycle, the second, third and subsequent chemotherapy times are intelligently calculated. The next chemotherapy time is intelligently reminded the day before the next chemotherapy and the patient is reminded to enter the mini program to confirm the appointment time. After the patient confirms the appointment time, the patient is reminded of the precautions before chemotherapy, the daytime chemotherapy process, and the items needed to be brought, etc.

[0092] Specifically, the intelligent reminder module reminds patients and medical staff to prepare for relevant examinations based on the patient's functional service layer data, interactive processing layer data, and the patient's clinical data in the hospital database. For example: according to the clinical prognosis, remind the patient to recheck the blood routine on the 5th, 12th, and 20th day after chemotherapy, and remind the patient to recheck the blood routine at 20:00 the night before and 14:00 on the day of the recheck. If the patient has rechecked the blood routine, stop the reminder; set the reminder for rechecking the blood routine and the reminder for the next chemotherapy time, and the reminder method is mainly through sending message reminders.

[0093] Specifically, through the intelligent control layer, timely and effective feedback and reminders can be provided to patients, medical staff can be reminded in time to carry out corresponding promotion work and give scientific advice.

[0094] In some embodiments, the preset alarm levels are divided into first-level alarm, second-level alarm, third-level alarm and fourth-level alarm;

[0095] The first level alarm is a low-risk warning, indicating that the patient has no abnormal symptoms, and the alarm color is green;

[0096] The second level alarm is a low-risk warning, indicating that the patient's symptoms are mild, and the alarm color is green;

[0097] The third level alarm is a medium-risk warning, indicating that the patient has mild or non-life-threatening symptoms, and the alarm color is yellow;

[0098] The fourth level alarm is a high-risk warning, indicating that the patient has serious symptoms or is even life-threatening, and the alarm color is red.

[0099] In some embodiments, in response to a third level alarm, medical personnel are required to contact the patient within four hours;

[0100] In response to a Level 4 alert, medical staff are required to contact the patient and conduct a clinical assessment within one hour.

[0101] In this embodiment, the preset alarm levels are divided into 4 levels. The first and second level alarms are low-risk warnings, indicating that the patient's symptoms are mild or have no abnormal symptoms, and the alarm color is green; the third level alarm is a medium-risk warning, indicating that the patient has mild or non-life-threatening symptoms, but early intervention may prevent the progression of symptoms or minimize its damage, and the alarm color is yellow; the fourth level alarm is a high-level warning, indicating that the patient has serious symptoms or even life-threatening, the alarm color is red, and immediate treatment is required.

[0102] Specifically, if the first or second level alarm occurs, only intelligent reminders or encouragement to the patient to continue will be given; if the third level alarm occurs, early intervention is required to prevent the progression of symptoms or minimize its damage, and medical staff are required to contact the patient within 4 hours, understand the patient's breathing situation in detail, and follow up treatment; if the fourth level alarm occurs, immediate intervention is required. The specific flow chart is shown in Figure 4, and medical staff must contact the patient within 1 hour and conduct a clinical evaluation, provide timely professional advice and support, give targeted treatment measures based on the results of the clinical evaluation, and conduct follow-up visits.

[0103] Specifically, in some embodiments, the levels are divided as shown in Table 1.

[0104] Table 1 Level Alarm Threshold Values

[0105] hematology Level I Level II Grade III Level IV <![CDATA[White blood cell count (×10 9 )]]> 3.0-3.9 2.0-2.9 1.0-1.9 <1.0 <![CDATA[Neutrophil count (×10 9 )]]> 1.5-2.0 1.0-1.5 0.5-1.0 <0.5 Hemoglobin (g / L) 95-109 80-94 65-79 <65 <![CDATA[Platelets (×10 9 )]]> 75-99 50-75 25-50 <25

[0106] Specifically, by dividing the alarm levels and formulating corresponding targeted measures for each alarm level, appropriate guidance and suggestions can be provided to patients in a timely and effective manner, reducing the infection and readmission rates of patients.

[0107] In some embodiments, the personalized recommendations and suggestions include: guiding the patient on activities, rest, diet, and medication.

[0108] In this embodiment, according to the prediction results of the CIM symptom development prediction model, personalized suggestions and opinions are given to the patient. Since the conditions of each patient are different, corresponding suggestions are given according to different conditions. Figure 5 As shown in the figure, if the patient's white blood cell count is 2.1 and the neutrophil count is 45, the following suggestions are given to the patient: 1. Take rest; 2. Increase nutrition, eat fish, poultry, meat, eggs, milk, peanuts, etc.; 3. Keep warm, avoid colds, pay attention to personal hygiene, wash hands frequently, and remember to wear a mask when going to crowded places; 4. Take oral white blood cell tablets, check blood routine on time and upload the test results. For another example, if the hemoglobin value is 130, which is within the normal range, the suggestions and opinions are: the hemoglobin result is normal, please continue to maintain it.

[0109] In some embodiments, the prediction model for the development of CIM symptoms during the interval period of tumor chemotherapy based on ePROMs is constructed, and its input parameters specifically include:

[0110] Demographic characteristics data: including mean age, median, age range, height, weight, and body mass index;

[0111] Clinical characteristic data: including tumor staging data, chemotherapy regimen data, the proportion of patients with bone metastasis and its distribution characteristics, and comorbid disease data;

[0112] Routine blood test data: including white blood cell count, neutrophil count, red blood cell count, platelet count, hemoglobin concentration, and the changing trend of routine blood test results before and after chemotherapy;

[0113] Symptom characteristic data: including the frequency and severity of common symptoms, and symptom development trends;

[0114] Quality of life data: including the patient's quality of life score, the relationship between quality of life and chemotherapy cycles, bone metastasis, and symptom severity;

[0115] Group characteristic data: including characteristics of patients grouped according to CIM symptom development pattern, chemotherapy cycle response, etc., and intra-group and inter-group difference data;

[0116] Correlation characteristic data: including the correlation data between the patient's demographic characteristics, clinical characteristics, chemotherapy regimen, blood routine test results, symptoms and quality of life scores, and key correlation data;

[0117] Time dynamic characteristic data: including the trend of symptom development and blood routine changes at different time points during the chemotherapy cycle, and symptom data at periodic or specific time points;

[0118] Prognostic feature data: including survival estimates under different treatment options and the relationship between specific CIM symptoms and prognosis.

[0119] In this embodiment, the input parameters of the CIM symptom development prediction model include data in nine aspects, namely, demographic characteristics, clinical characteristics, blood routine, symptom characteristics, quality of life, grouping characteristics, correlation characteristics, time dynamic characteristics and prognostic characteristics, covering patient data in multiple dimensions, providing a comprehensive data foundation for the CIM symptom development prediction model, making the final prediction results more intelligent and accurate, and also making personalized suggestions and opinions based on the prediction results more practical and closer to the actual situation of different patients.

[0120] In some embodiments, the method of data sorting, cleaning and analysis includes:

[0121] Descriptive statistical analysis: used to understand the basic characteristics of the data;

[0122] Correlation analysis: used to analyze the relationship between independent variables and dependent variables;

[0123] Regression analysis: including linear regression analysis: used to quantify the relationship between independent variables and dependent variables; logistic regression analysis: used to analyze binary data; and multiple regression analysis: used to analyze situations with multiple dependent variables;

[0124] Time series analysis: used to analyze and predict changes in patient symptoms and blood routine results at different time points during the chemotherapy cycle;

[0125] Cluster analysis: used to divide patient groups with similar CIM development characteristics into corresponding groups;

[0126] Principal component analysis: used to reduce the dimensionality of data to reveal the main variation patterns of the data and simplify the analysis of multidimensional data;

[0127] Survival analysis: evaluate the survival time of patients under different treatment options or the time when CIM symptoms appear;

[0128] Factor Analysis: Used to identify latent structures in the data.

[0129] In this embodiment, when the functional service layer data, the interactive processing layer data and the patient clinical data in the hospital database are processed into the input parameters of the CIM symptom development prediction model, the main methods used include descriptive statistical analysis, correlation analysis, regression analysis, time series analysis, cluster analysis, principal component analysis, survival analysis and factor analysis 8 total analysis methods. Through the combined use of these methods, the development characteristics and laws of CIM symptoms of patients during the interval period of tumor chemotherapy can be comprehensively analyzed and revealed.

[0130] Specifically, descriptive statistical analysis is used to understand the basic characteristics of the data, such as mean, median, standard deviation, frequency distribution, etc. It is used to statistically analyze the distribution of patients' gender, age, height, body mass index, blood routine results, symptom scores, quality of life scores and other data.

[0131] Specifically, correlation analysis is used to analyze the relationship between independent variables and dependent variables, which uses Pearson correlation coefficient, Spearman rank correlation coefficient or Kendall's Tau to evaluate the correlation between variables, where independent variables include age, BMI, tumor stage, etc., and dependent variables include CIM symptoms, quality of life score, etc.

[0132] Specifically, linear regression analysis in regression analysis is used to quantify the relationship between independent variables and dependent variables and predict the severity of CIM symptoms; logistic regression analysis is used to analyze binary data, such as whether severe bone marrow suppression occurs; and multivariate regression analysis is used in cases with multiple dependent variables to explore the impact of multiple factors on CIM symptoms.

[0133] Specifically, the goal of time series analysis is to analyze and predict changes in patient symptoms and blood routine results at different time points during the chemotherapy cycle.

[0134] Specifically, cluster analysis methods include K-means, hierarchical clustering and other methods, which can be used to group patients and facilitate the management of the platform.

[0135] Specifically, principal component analysis can reduce the number of features while retaining the main information in the data, such as the main change patterns of blood routine and clinical indicators, which is beneficial to the calculation of the model.

[0136] Specifically, through the combined use of various analysis methods, the functional service layer data, the interactive processing layer data and the patient clinical data in the hospital database can be sorted, cleaned and analyzed, and finally the input parameters of the CIM symptom development prediction model can be obtained.

[0137] In some embodiments, the doctor-patient Q&A specifically involves the patient and medical staff consulting and answering questions about problems encountered by the patient;

[0138] The patient-patient healthy interaction refers specifically to patients sharing experiences, talking to each other, encouraging each other and supporting each other;

[0139] The nurse-patient follow-up specifically refers to the medical staff providing one-on-one guidance to the patient and conducting regular online follow-up, the follow-up content mainly covers the patient's recent life and medical treatment;

[0140] The human-machine intelligent suggestion specifically refers to the platform giving targeted content push based on the patient's clinical data and the functional service layer data, based on the patient's questions raised, the information browsed, and the patient's status.

[0141] In this embodiment, the specific method of multi-dimensional interaction through the interactive processing layer is pointed out, including doctor-patient Q&A, patient-patient interaction data, nurse-patient follow-up data, human-computer interaction data, etc., to achieve multi-faceted and multi-dimensional interactions such as human-to-human interaction, human-computer interaction, and information interaction. Through interaction, the positive effects of verbal persuasion, indirect experience, and emotional awakening in the self-efficacy theory can be achieved.

[0142] In a second aspect, the present invention proposes an electronic device on which the personalized management platform for bone marrow suppression based on patient self-reported outcomes described in the first aspect is deployed.

[0143] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

[0144] In addition, it should be understood that although the present specification is described according to embodiments, not every embodiment contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation methods that those skilled in the art can understand.

Claims

1. A personalized management platform for myelosuppression based on patient self-reported outcomes, characterized by: include: Functional service layer: used to provide functional services for medical and nursing ends and patient ends. The medical and nursing ends upload resource information, upload questionnaires, and record the overall chemotherapy appointment list through the functional service layer. The patient ends browse resource information, fill in questionnaires, and upload self-examination data through the functional service layer. The functional service layer is also used to store functional service layer data, which includes: patient resource information browsing data, questionnaire data, chemotherapy record data, and self-examination data; Interaction processing layer: used to realize multi-dimensional interaction functions, including: patient-patient health interaction, nurse-patient tracking and follow-up, human-computer intelligent suggestions, and doctor-patient Q&A; The interactive processing layer is also used to store interactive processing layer data, which includes: patient-patient interaction data, nurse-patient follow-up data, human-computer interaction data, and doctor-patient question and answer data; Intelligent control layer: performs intelligent analysis based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database, and provides alarms, reminders and appointment functions to the patient side according to the analysis results.

2. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 1, characterized in that: The medical care end uploads resource information, uploads questionnaires, and records the overall chemotherapy appointment list through the functional service layer, including: Medical users upload resource information to the resource information module through the functional service layer. The resource information includes: tumor health education content in the form of text, pictures, audio, video, tumor disease knowledge, symptom management, coping skills, and social resources; Medical users edit the content and format of the questionnaire in the questionnaire module and publish it in the questionnaire module; The chemotherapy appointment list module presets the patient chemotherapy schedule, and the medical staff records the chemotherapy information in the chemotherapy appointment list module based on whether the patient undergoes chemotherapy on time; The patient at the patient end uploads self-examination data through the self-examination data module of the functional service layer.

3. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 1, characterized in that: The intelligent control layer includes: an intelligent analysis module: based on the functional service layer data, the interactive processing layer data and the clinical data of patients in the hospital database, the intelligent analysis module performs data sorting, cleaning and analysis, builds a prediction model for the development of CIM symptoms during the interval period of tumor chemotherapy based on ePROMs, and gives personalized suggestions and opinions based on the prediction results; Intelligent alarm module: based on the prediction results of the intelligent analysis module and the preset alarm level, contact and guide the patient for follow-up treatment; Intelligent appointment module: calculates the patient's follow-up examination and treatment plan based on the functional service layer data, the interactive processing layer data and the patient's clinical data in the hospital database and informs the patient and medical staff through the interactive processing module; Intelligent reminder module: reminds patients and medical staff of examination time based on the functional service layer data, interactive processing layer data and patient clinical data in the hospital database.

4. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 3, characterized in that: The preset alarm levels are divided into first-level alarm, second-level alarm, third-level alarm and fourth-level alarm; The first level alarm is a low-risk warning, indicating that the patient has no abnormal symptoms, and the alarm color is green; The second level alarm is a low-risk warning, indicating that the patient's symptoms are mild, and the alarm color is green; The third level alarm is a medium-risk warning, indicating that the patient has mild or non-life-threatening symptoms, and the alarm color is yellow; The fourth level alarm is a high-risk warning, indicating that the patient has serious symptoms or is even life-threatening, and the alarm color is red.

5. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 4, characterized in that: In response to a Level 3 alert, medical staff are required to contact the patient within four hours; In response to a Level 4 alert, medical staff are required to contact the patient and conduct a clinical assessment within one hour.

6. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 3, characterized in that: The personalized recommendations and suggestions include: guiding the patient's activities, rest, diet, and medication.

7. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 3, characterized in that: The ePROMs-based model for predicting the development of CIM symptoms during the interval period of tumor chemotherapy is constructed, and its input parameters specifically include: Demographic characteristics data: including mean age, median, age range, height, weight, and body mass index; Clinical characteristic data: including tumor staging data, chemotherapy regimen data, the proportion of patients with bone metastasis and its distribution characteristics, and comorbid disease data; Routine blood test data: including white blood cell count, neutrophil count, red blood cell count, platelet count, hemoglobin concentration, and the changing trend of routine blood test results before and after chemotherapy; Symptom characteristic data: including the frequency and severity of common symptoms, and symptom development trends; Quality of life data: including the patient's quality of life score, the relationship between quality of life and chemotherapy cycles, bone metastasis, and symptom severity; Group characteristic data: including characteristics of patients grouped according to CIM symptom development pattern, chemotherapy cycle response, etc., and intra-group and inter-group difference data; Correlation characteristic data: including the correlation data between the patient's demographic characteristics, clinical characteristics, chemotherapy regimen, blood routine test results, symptoms and quality of life scores, and key correlation data; Time dynamic characteristic data: including the trend of symptom development and blood routine changes at different time points during the chemotherapy cycle, and symptom data at periodic or specific time points; Prognostic feature data: including survival estimates under different treatment options and the relationship between specific CIM symptoms and prognosis.

8. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 3, characterized in that: The method for data sorting, cleaning and analysis comprises: Descriptive statistical analysis: used to understand the basic characteristics of the data; Correlation analysis: used to analyze the relationship between independent variables and dependent variables; Regression analysis: including linear regression analysis: used to quantify the relationship between independent variables and dependent variables; logistic regression analysis: used to analyze binary data; and multiple regression analysis: used to analyze situations with multiple dependent variables; Time series analysis: used to analyze and predict changes in patient symptoms and blood routine results at different time points during the chemotherapy cycle; Cluster analysis: used to divide patient groups with similar CIM development characteristics into corresponding groups; Principal component analysis: used to reduce the dimensionality of data to reveal the main variation patterns of the data and simplify the analysis of multidimensional data; Survival analysis: evaluate the survival time of patients under different treatment options or the time when CIM symptoms appear; Factor Analysis: Used to identify latent structures in the data.

9. The personalized management platform for myelosuppression based on patient self-reported outcomes according to claim 1, characterized in that: The doctor-patient Q&A refers specifically to consultation and answering of questions between patients and medical staff regarding problems encountered by patients; The patient-patient healthy interaction refers specifically to patients sharing experiences, talking to each other, encouraging each other and supporting each other; The nurse-patient follow-up specifically refers to the medical staff providing one-on-one guidance to the patient and conducting regular online follow-up, the follow-up content mainly covers the patient's recent life and medical treatment; The human-machine intelligent suggestion specifically refers to the platform giving targeted content push based on the patient's clinical data and the functional service layer data, based on the patient's questions raised, the information browsed, and the patient's status.

10. An electronic device, characterized in that: The electronic device is deployed with the personalized management platform for bone marrow suppression based on patient self-reported outcomes as described in any one of claims 1 to 9.

Images