Data-driven comprehensive evaluation of internal medicine residency training
Jin Haijiao
Zhang Minfang
Fang Yan
Gu Jiaqi
Si Yu
Hu Huiming
Feng Zhiying
Ni Zhaohui
Abstract：Objective Background The current standardized training evaluation system for resident physicians has limitations.This study aimed to construct a comprehensive evaluation system for internal medicine residents,incorporating full-cycle,full-process and full-data,to predict their employment destination and provide support for training optimization and talent management.Methods Data from 127 internal medicine residents at Ren Ji Hospital,Shanghai Jiao Tong University School of Medicine were collected from 2017 to 2023.Model parameters were adjusted to optimize prediction performance.A 5-fold cross-validation method was used to assess the model's robustness and generalization ability.Additionally,feature importance analysis was performed to identify the most influential factors affecting employment destination.Results The model achieved an overall accuracy of 91.3％on the entire dataset.The 5-fold cross-validation results demonstrated the model's stability and consistency.Feature importance analysis revealed that academic achievements,such as the number of published SCI papers,awards or honors,and fund support from the National Natural Science Foundation were key factors in predicting employment destination.Conclusion This study constructs a comprehensive evaluation system for internal medicine residents to predict their employment destination and identify key influencing factors.Data-driven evaluation provides a new perspective for educational reform and talent management.Future research can expand data dimensions and optimize the model to improve prediction performance.
Keywords：standardized residency trainingresident evaluationemployment outcomes
Publication Date：2025-05-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 361-365 )