Sample data application mode
Product Description
Based on the effective association between sample information, clinical data and data (genome, transcriptome, proteome data, etc.) generated by the multi-omics platform, relying on the network real-time management and effective integration between different sample databases brought by the big data platform of sample database, applying the analysis and optimization methods of big data and artificial intelligence to conduct structured processing and in-depth mining of a large number of clinical unstructured data, Promote the transformation of the research model of major diseases and promote the research process of translational medicine (see Figure 11 for details).
Figure 11 Big data integration and utilization architecture
(1) The clinical application scenario big data platform can be effectively applied to clinical scenarios to solve the problems of traditional clinical data search such as low efficiency, limited search methods, and strong professional technical requirements. The design based on the big data architecture can effectively optimize traditional retrieval problems, and can be used in multiple modes such as full-text search, structured search, segmentation search, fuzzy search, and composite search, and can achieve second-level search; Based on the big data platform, a large number of electronic medical records can be used as analysis samples on the basis of standard data production, combined with clinical knowledge, to abstract disease feature fields for modeling, and combined with professional clinical knowledge base, to provide comprehensive auxiliary decision-making services.
(2) It is difficult to find the scientific research ideas of clinical samples in scientific research application scenarios, many unstructured fields of medical records need a lot of manpower, and the processed data need time and energy to be converted and processed before analysis. This series of problems perplex the scientific research work of clinicians. The big data platform can be used to support the whole process of clinical scientific research (discovery of scientific research inspiration, preliminary research and verification, scientific research project approval, delimitation of target population, establishment of observation indicators, data collection, final statistical analysis and article writing, etc.) to help clinical doctors carry out scientific research more efficiently.
(3) Management application scenarios can unify the statistical caliber of data and effectively centralize the classification methods of multi-service systems through the big data platform; Form quality comparison of alliance hospitals; To form effective means for the construction of key disciplines and the development of disease types; Control and evaluate data quality problems; Promote the improvement of the overall medical quality control plan; Establish a unified index of samples; Unified master data management; Effective use of clinical electronic medical record data and scientific research data for medical research.
(4) In addition to providing data and application support for sample libraries, medical institutions, scientific research, etc., the big data platform for donor service scenarios will also be of great significance in sample donor services. Based on the sample, clinical, follow-up, scientific research and other information provided by the donor, a set of intelligent guidance service mode can be formed, and artificial intelligence technology can be used to identify the diseases that the donor may have and provide medical guidance. Combining the big data platform disease knowledge map, data mining guidance algorithm, semantic analysis and other tools, so that donors can get guidance services. By establishing the knowledge base of relevant specialized diseases, symptoms and treatment methods, using the question generator and the medical record classifier, the connection between the knowledge base content and the question content and the recommended hospital is established to realize intelligent recommendation.
(5) Drug research scene In the current era, the quality management of clinical trials is directly related to the drug research and development process and disease prevention strategies. Relying on the big data platform and referring to professional standard data sets and centralized unified coding, clinical trials and clinical research can be carried out with high quality and efficiency. Through the big data platform, based on the inclusion and exclusion conditions of the project, and in combination with the clinical scenario, the subjects suspected to meet the inclusion conditions are found, and the clinical trial enrollment is accelerated. Based on the historical project verification report and quality control rules of the big data platform, it can intelligently predict the possible risks of clinical trials, and use the big data technology to carry out quality evaluation, find out in real time the quality problems such as protocol non-compliance, data inconsistency, AE/SAE omission, and remind the research team to rectify. At the same time, prior to the on-site quality control of CRA, quality controller and inspector, based on the data model calculation, priority should be given to the production of quality control analysis reports to inform the quality control personnel of the risk points, so as to facilitate the quality control personnel to carry out targeted inspections.
(6) Teaching application scenario The disease atlas based on big data is the third tool for clinical medical teaching in addition to textbooks and literature. Statistical analysis of massive real data of a disease can present the real distribution of disease characteristics, including age distribution, common symptoms, gender ratio, common inspection and test methods, etc. The revealing of the logical relationship between the data of each node of big data helps to interpret the disease information in depth. We should combine theory with clinical practical cases to assist clinical teaching.
(2) It is difficult to find the scientific research ideas of clinical samples in scientific research application scenarios, many unstructured fields of medical records need a lot of manpower, and the processed data need time and energy to be converted and processed before analysis. This series of problems perplex the scientific research work of clinicians. The big data platform can be used to support the whole process of clinical scientific research (discovery of scientific research inspiration, preliminary research and verification, scientific research project approval, delimitation of target population, establishment of observation indicators, data collection, final statistical analysis and article writing, etc.) to help clinical doctors carry out scientific research more efficiently.
(3) Management application scenarios can unify the statistical caliber of data and effectively centralize the classification methods of multi-service systems through the big data platform; Form quality comparison of alliance hospitals; To form effective means for the construction of key disciplines and the development of disease types; Control and evaluate data quality problems; Promote the improvement of the overall medical quality control plan; Establish a unified index of samples; Unified master data management; Effective use of clinical electronic medical record data and scientific research data for medical research.
(4) In addition to providing data and application support for sample libraries, medical institutions, scientific research, etc., the big data platform for donor service scenarios will also be of great significance in sample donor services. Based on the sample, clinical, follow-up, scientific research and other information provided by the donor, a set of intelligent guidance service mode can be formed, and artificial intelligence technology can be used to identify the diseases that the donor may have and provide medical guidance. Combining the big data platform disease knowledge map, data mining guidance algorithm, semantic analysis and other tools, so that donors can get guidance services. By establishing the knowledge base of relevant specialized diseases, symptoms and treatment methods, using the question generator and the medical record classifier, the connection between the knowledge base content and the question content and the recommended hospital is established to realize intelligent recommendation.
(5) Drug research scene In the current era, the quality management of clinical trials is directly related to the drug research and development process and disease prevention strategies. Relying on the big data platform and referring to professional standard data sets and centralized unified coding, clinical trials and clinical research can be carried out with high quality and efficiency. Through the big data platform, based on the inclusion and exclusion conditions of the project, and in combination with the clinical scenario, the subjects suspected to meet the inclusion conditions are found, and the clinical trial enrollment is accelerated. Based on the historical project verification report and quality control rules of the big data platform, it can intelligently predict the possible risks of clinical trials, and use the big data technology to carry out quality evaluation, find out in real time the quality problems such as protocol non-compliance, data inconsistency, AE/SAE omission, and remind the research team to rectify. At the same time, prior to the on-site quality control of CRA, quality controller and inspector, based on the data model calculation, priority should be given to the production of quality control analysis reports to inform the quality control personnel of the risk points, so as to facilitate the quality control personnel to carry out targeted inspections.
(6) Teaching application scenario The disease atlas based on big data is the third tool for clinical medical teaching in addition to textbooks and literature. Statistical analysis of massive real data of a disease can present the real distribution of disease characteristics, including age distribution, common symptoms, gender ratio, common inspection and test methods, etc. The revealing of the logical relationship between the data of each node of big data helps to interpret the disease information in depth. We should combine theory with clinical practical cases to assist clinical teaching.
