CN-122000034-A - Occupational pneumoconiosis screening and management system based on multisource data fusion

CN122000034ACN 122000034 ACN122000034 ACN 122000034ACN-122000034-A

Abstract

The invention discloses a screening and management system for occupational pneumoconiosis based on multi-source data fusion, which relates to the technical field of occupational disease prevention and treatment, and comprises a multi-source data acquisition module, a data fusion module, a screening diagnosis module, a scheme making module and a follow-up management module; the invention integrates image data, physiological data, basic clinical data, biomarker detection data and treatment process data through a multi-source data acquisition module, classifies, cleans and standardizes the multi-source data through a data fusion module, extracts characteristics to construct a multi-dimensional data model, builds a classification diagnosis model according to multi-center prospective queue research data, and completes the health state and illness classification of the examined person, thereby breaking the limitation of single data dimension, providing comprehensive basis for diagnosis classification by utilizing the complementarity of the multi-source data, enabling the classification standard to have more scientificity and universality through the queue research data support, and laying a solid foundation for the establishment of an intervention scheme.

Inventors

XUAN JIE
XU HAIMING
DU JIA
GAO YANG
LIU XIUFANG
LIU MEILING
ZHU JIA
Tan Qingliu
YAO XIAOMEI

Assignees

宁夏回族自治区第五人民医院(宁夏回族自治区国家矿山医疗救护中心)

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (9)

1.A multi-source data fusion-based occupational pneumoconiosis screening and management system, comprising: The multi-source data acquisition module acquires multi-source data of a person to be detected through an imaging detection device, an intelligent wearing sensing device, a clinical information input unit and a treatment record port, wherein the multi-source data comprises image data, physiological data, basic clinical data, biomarker detection data and treatment process data; The data fusion module is used for preprocessing the acquired multi-source data, extracting core characteristics of imaging, physiological indexes and clinical information, constructing a multi-dimensional data model and outputting a standardized characteristic data set; the screening diagnosis module relies on multi-center prospective queue research data based on a standardized characteristic data set, divides the research queues according to etiology, imaging performance and disease course stage by stage, builds a classification diagnosis model and completes screening and disease classification of the pneumoconiosis of the person to be tested; The scheme making module is used for making a layered intervention scheme covering drug treatment, respiratory training, exercise rehabilitation and psychological dispersion based on the disease grading result and combining with the grouping characteristics of the research queue; And the follow-up management module is used for acquiring current multi-source data and symptom change information of the person to be tested according to a fixed period, and calling historical multi-source data of the person to be tested and a currently executed layering intervention scheme to adjust the layering intervention scheme.
2. The professional pneumoconiosis screening and managing system based on multi-source data fusion according to claim 1 is characterized in that scanning of the whole lung area of a tested person is completed through imaging detection equipment in the multi-source data acquisition module, image data of X-ray and chest low-dose HRCT are obtained, physiological data of heart rate, respiratory rate and blood oxygen saturation of the tested person in a daily state are acquired through intelligent wearable sensing equipment, basic clinical data of age, occupation, dust collection years, past medical history and family history of the tested person and biomarker detection data of inflammatory factors and lung function indexes are input through a clinical information input unit, and therapeutic process data of drug use and rehabilitation training execution conditions of the tested person are input through a therapeutic record port.
3. The professional pneumoconiosis screening and management system based on multi-source data fusion according to claim 1, wherein in the data fusion module, collected multi-source data are classified and arranged, corresponding sources and types of the data are marked, the data are divided into image data, physiological index data, clinical information data and treatment process data, the classified data are cleaned, invalid data are removed, the cleaned valid data are standardized, data formats and units are unified, the characteristics of various data are extracted, a multi-dimensional data model is built after all the characteristics are integrated, a standardized characteristic data set containing all core characteristics is output, and the weight ratio of various characteristics in the multi-dimensional data model is determined through a multi-source characteristic fusion weight formula.
4. The professional pneumoconiosis screening and management system based on multi-source data fusion according to claim 3, wherein in the data fusion module, a multi-source feature fusion weight formula is as follows: , wherein, Is the first The fusion weight of the class feature, Is the first Contribution of class characteristics to diagnosis of pneumoconiosis, Is the first The data integrity of the class feature, Is the total category number of the feature.
5. The screening and managing system for professional pneumoconiosis based on multi-source data fusion according to claim 1, wherein in the screening diagnosis module, historical case data of multi-center prospective queue research is obtained, the historical case data are grouped according to dividing criteria of etiology, imaging performance and disease course stage, meanwhile, a classifying diagnosis model base frame is built, the grouped historical case data are input into the classifying diagnosis model base frame to complete training of a classifying diagnosis model, a standardized feature data set is input into the trained classifying diagnosis model, a disease classification index is calculated, and screening and disease classification of a person to be tested are completed based on the disease classification index.
6. The system for screening and managing occupational pneumoconiosis based on multi-source data fusion according to claim 5, wherein in the screening diagnosis module, a calculation formula of the trained classified diagnosis model is as follows: , wherein, Is the index of the grading of the illness state, Is the first The fusion weight of the class feature, Is the first The quantized values of the class feature, The total category number of the feature; Based on the disease grading index, the pneumoconiosis screening and disease grading are completed when When it is determined as healthy people In case of pneumoconiosis, stage I is determined Stage II pneumoconiosis is determined at 50 when 75, When it is judged as pneumoconiosis stage III And judging that the pneumoconiosis is stage IV, wherein the pneumoconiosis stages I-II are classified into early pneumoconiosis and the pneumoconiosis stages III-IV are classified into late pneumoconiosis.
7. The system for screening and managing occupational pneumoconiosis based on multi-source data fusion according to claim 1, wherein in the scheme making module, a layered intervention scheme is combined with disease grading and research queue grouping characteristics to make a pneumoconiosis I period, wherein the scheme comprises regular lung function monitoring, basic anti-inflammatory treatment and mild respiratory training, a pneumoconiosis II period, wherein the scheme comprises reinforced anti-infective treatment, normal respiratory support and moderate exercise rehabilitation, a pneumoconiosis III period, wherein the scheme comprises lung rehabilitation training, psychological dredging and complication prevention, a pneumoconiosis IV period, wherein the scheme comprises reinforced complication prevention, long-term respiratory support and palliative treatment, and wherein the scheme comprises regular physical examination related contents for a high-risk occupational population of pneumoconiosis, wherein the scheme comprises dust protection training, regular physical examination and respiratory function exercise of corresponding occupations in combination with occupational exposure characteristics.
8. The system for screening and managing occupational pneumoconiosis based on multi-source data fusion according to claim 1, wherein in the follow-up management module, current multi-source data and symptom change information of a person to be tested are obtained according to a fixed period, a hierarchical intervention scheme of historical multi-source data and current execution of the person to be tested is called, a scheme adjustment requirement index is calculated through an intervention scheme adjustment evaluation formula, and an existing hierarchical intervention scheme is adjusted according to the scheme adjustment requirement index.
9. The system for screening and managing occupational pneumoconiosis based on multi-source data fusion according to claim 8, wherein in the follow-up management module, an intervention scheme adjustment evaluation formula is: , wherein, The demand index is adjusted for the solution, The influence coefficient representing the index is calculated, And The quantized values of lung function indexes and inflammatory factor indexes in the current multi-source data and the historical multi-source data are respectively obtained; the existing intervention scheme is adjusted according to the scheme adjustment demand index when When the existing intervention scheme is adjusted, the adjustment mode comprises the steps of increasing and decreasing the dosage of the medicine, adjusting the intensity of rehabilitation training and adding new complication prevention and treatment measures, when When the existing intervention scheme is maintained.

Description

Occupational pneumoconiosis screening and management system based on multisource data fusion Technical Field The invention relates to the technical field of occupational disease prevention and treatment, in particular to a screening and management system for occupational pneumoconiosis based on multi-source data fusion. Background Professional pneumoconiosis is a systemic disease caused by long-term inhalation of productive mineral dust by workers, is highly developed by practitioners in industries such as coal mining, building construction, metal processing and the like, has hidden and irreversible pathological progress, has no obvious clinical symptoms in early stage, often causes diffuse fibrosis of lung tissues during diagnosis, is easy to cause chronic obstructive pulmonary disease, phthisis and other diseases, seriously threatens the life health of the practitioners, and becomes a key attention topic in the field of prevention and treatment of the professional diseases. However, the existing pneumoconiosis screening and diagnosing means rely on single imaging detection or lung function test, effective integration of multi-source data such as image data, physiological indexes, clinical information and the like is not realized, the health state of a person to be tested cannot be comprehensively reflected, the grading diagnosis lacks the support of historical case data, the grading standard subjectivity is strong, the intervention scheme mostly adopts a standardized flow, the individual content is formulated without combining with disease grading, the follow-up management only depends on symptom observation, the comparison analysis and quantitative adjustment basis of the multi-source data are lacked, and the dynamic optimization of the intervention strategy is difficult. Disclosure of Invention The invention aims to make up the defects of the prior art and provides a screening and management system for occupational pneumoconiosis based on multi-source data fusion, which integrates image data, physiological data, basic clinical data, biomarker detection data and treatment process data through a multi-source data acquisition module, classifies, cleans and standardizes the multi-source data through a data fusion module, the characteristics are extracted to construct a multidimensional data model, a classification diagnosis model is built by relying on multi-center prospective queue research data, and the health state and illness state classification of the examined person are completed, so that the limitation of single data dimension is broken, the complementarity of multi-source data is utilized to provide comprehensive basis for diagnosis classification, the classification standard is more scientific and universal through queue research data support, and a solid foundation is laid for the establishment of an intervention scheme. The invention provides a professional pneumoconiosis screening and management system based on multi-source data fusion, which comprises the following steps: The multi-source data acquisition module acquires multi-source data of a person to be detected through an imaging detection device, an intelligent wearing sensing device, a clinical information input unit and a treatment record port, wherein the multi-source data comprises image data, physiological data, basic clinical data, biomarker detection data and treatment process data; The data fusion module is used for preprocessing the acquired multi-source data, extracting core characteristics of imaging, physiological indexes and clinical information, constructing a multi-dimensional data model and outputting a standardized characteristic data set; the screening diagnosis module relies on multi-center prospective queue research data based on a standardized characteristic data set, divides the research queues according to etiology, imaging performance and disease course stage by stage, builds a classification diagnosis model and completes screening and disease classification of the pneumoconiosis of the person to be tested; The scheme making module is used for making a layered intervention scheme covering drug treatment, respiratory training, exercise rehabilitation and psychological dispersion based on the disease grading result and combining with the grouping characteristics of the research queue; And the follow-up management module is used for acquiring current multi-source data and symptom change information of the person to be tested according to a fixed period, and calling historical multi-source data of the person to be tested and a currently executed layering intervention scheme to adjust the layering intervention scheme. Further, in the multi-source data acquisition module, scanning of the whole lung area of the detected person is completed through an imaging detection device, image data of X-ray and chest low-dose HRCT are acquired, physiological data of heart rate, respiratory rate and blood oxygen saturation of the dete