Purpose Predictively diagnosing infectious diseases helps in providing better treatment and enhances the prevention and control of such diseases. This study uses actual data from a hospital. A multiple infectious disease diagnostic model (MIDDM) is designed for conducting multi-classification of infectious diseases so as to assist in clinical infectious-disease decision-making. Methods Based on actual hospital medical records of infectious diseases from December 2012 to December 2020, a deep learning model for multi-classification research on infectious diseases is constructed. The data includes 20,620 cases covering seven types of infectious diseases, including outpatients and inpatients, of which training data accounted for 80%, i.e., 16,496 cases, and test data accounted for 20%, i.e., 4124 cases. Through the auto-encoder, data normalization and sparse data densification processing are carried out to improve the model training effect. A residual network and attention mechanism are introduced into the MIDDM model to improve the performance of the model. Result MIDDM achieved improved prediction results in diagnosing seven kinds of infectious diseases. In the case of similar disease diagnosis characteristics and similar interference factors, the prediction accuracy of disease classification with more sample data is significantly higher than the prediction accuracy of disease classification with fewer sample data. For instance, the training data for viral hepatitis, influenza, and hand foot and mouth disease were 2954, 3924, and 3015 respectively and the corresponding test accuracy rates were 99.86%, 98.47%, and 97.31%. There is less training data for syphilis, infectious diarrhea, and measles, i.e., 1208, 575, and 190 respectively and the corresponding test accuracy rates were noticeably lower, i.e., 83.03%, 87.30%, and42.11%. We also compared the MIDDM model with the models used in other studies. Using the same input data, taking viral hepatitis as an example, the accuracy of MIDDM is 99.44%, which is significantly higher than that of XGBoost (96.19%), Decision tree (90.13%), Bayesian method (85.19%), and logistic regression (91.26%). Other diseases were also significantly better predicted by MIDDM than by these three models. Conclusion The application of the MIDDM model to multi-class diagnosis and prediction of infectious diseases can improve the accuracy of infectious-disease diagnosis. However, these results need to be further confirmed via clinical randomized controlled trials.

No abstract

Background Modernizing medical education by using artificial intelligence and other new technologies to improve the clinical thinking ability of medical students is an important research topic in recent years. Prominent medical universities are actively conducting research and exploration in this area. In particular, given the shortage of human resources, the need to maintain social distancing to prevent the spread of the epidemics, and the increase in the cost of medical education, it is critical to harness online learning to promote medical education. A virtual case learning system that uses natural language processing technology to process and present a hospital’s real medical records and evaluate student responses can effectively improve medical students’ clinical thinking abilities. Objective The purpose of this study is to develop a virtual case system, AIteach, based on actual complete hospital medical records and natural language processing technology, and achieve clinical thinking ability improvement through a contactless, self-service, trial-and-error system application. Methods Case extraction is performed on a hospital’s case data center and the best-matching cases are produced through natural language processing, word segmentation, synonym conversion, and sorting. A standard clinical questioning data module, virtual case data module, and student learning difficulty module are established to achieve simulation. Students can view the objective examination and inspection data of actual cases, including details of the consultation and physical examination, and automatically provide their learning response via a multi-dimensional evaluation system. In order to assess the changes in students’ clinical thinking after using AIteach, 15 medical graduate students were subjected to two simulation tests before and after learning through the virtual case system. The tests, which included the full-process case examination of cases having the same difficulty level, examined core clinical thinking test points such as consultation, physical examination, and disposal, and generated multi-dimensional evaluation indicators (rigor, logic, system, agility, and knowledge expansion). Thus, a complete and credible evaluation system is developed. Results The AIteach system used an internal and external double-cycle learning model. Students collect case information through online inquiries, physical examinations, and other means, analyze the information for feedback verification, and generate their detailed multi-dimensional clinical thinking after learning. The feedback report can be evaluated and its knowledge gaps analyzed. Such learning based on real cases is in line with traditional methods of disease diagnosis and treatment, and addresses the practical difficulties in reflecting actual disease progression while keeping pace with recent research. Test results regarding short-term learning showed that the average score (P < 0.01) increased from 69.87 to 85.6, the five indicators of clinical thinking evaluation improved, and there was obvious logical improvement, reaching 47%. Conclusion By combining real cases and natural language processing technology, AIteach can provide medical students (including undergraduates and postgraduates) with an online learning tool for clinical thinking training. Virtual case learning helps students to cultivate clinical thinking abilities even in the absence of clinical tutor, such as during pandemics or natural disasters.

No abstract

No abstract