Deep learning model management for coronary heart disease early warning research

Coronary Heart Disease (CHD) is one of the common diseases that threaten people's health and life. To facilitate the CHD early warning research, the deep learning based methods have drawn much attention. However, the literature mostly focuses on how to establish and optimize the CHD early warning models, while overlooking the training data-model-experimental results modeling lifecycle management. Aiming to promote the early warning research of CHD, we contribute a data management system integrated the CHD patient data with the deep learning model data. In the system, a deep learning model version tree is established to represent the relationship between the models. Tracking-Ancestors algorithm and Find-Specified-Ancestor algorithm are designed to conduct the lineage management of the deep learning model. Considering the big data characteristics of the patient data and deep learning model data, we compare the query response time and select MongoDB as the DBMS for the Pdmdims (Patient Data & Deep Learning Model Data Integrated Management System). The research results show that Pdmdims can provide an effective integrated data management platform for CHD early warning researchers.