With the development of artificial intelligence, machine learning has been widely used in diagnosis of diseases. It is crucial to conduct diagnostic test accuracy studies and evaluate the performance of models reasonably to improve the accuracy of diagnosis. For machine learning-based diagnostic test accuracy studies, this paper introduces the principles of study design in the aspects of target conditions, selection of participants, diagnostic tests, reference standards and ethics.
ObjectiveThe purpose of this study was to translate the U-CEP scale into Chinese, and evaluate the reliability and validity of the Chinese version of the U-CEP, in order to provide a measurement and evaluation tool for clinical epidemiology education and research. MethodsThe U-CEP scale was translated and adapted using the Brislin translation model. A nationwide survey of clinicians was conducted using the Chinese version of the U-CEP. Item analysis, reliability analysis, and validity analysis were performed using SPSS 26.0 software. ResultsThe discriminant validity analysis showed that except for item 4, the critical value (CR) of the other twenty-four items differed significantly between high and low groups (P<0.01), with CR values ranging from 2.902 to 14.609. The ITCs of the 25 items were all positive, with 5 items having an ITC<0.15(20%), 2 items having ITC≥0.15~0.20 (8%), 6 items having ITC≥0.20~0.40 (24%) and 12 items having ITC≥0.40 (48%). In terms of reliability, the overall Cronbach’s α coefficient of the Chinese version of the U-CEP was 0.80, with Cronbach’s α coefficient ranging from 0.752 to 0.805 when deleting each item one by one. The test-retest reliability was 0.848 (P<0.001). The alternative-form reliability was 0.838 (P<0.001). In terms of validity, expert analysis showed that the content validity of the Chinese version of the U-CEP was good. The construct validity analysis showed that the cumulative contribution rate of the 25 items was 57.50%. No respondent scored full marks or zero marks, indicating that no ceiling or floor effects were found. There were statistically significant differences in the total scores among clinicians with different educational backgrounds or with or without systematic learning of relevant knowledge (P<0.05). ConclusionThe Chinese version of the U-CEP has good reliability and validity, as well as good cultural adaptability. It can effectively assess a physician's knowledge of clinical epidemiology.
ObjectiveBased on the requirements of the era of big medical data and discipline development, this study aimed to enhance the clinical research capabilities of medical postgraduates by exploring and evaluating some teaching innovations. MethodsA research-oriented clinical research design course was developed for postgraduate students, focusing on enhancing their clinical research abilities. Innovative teaching content and methods were implemented, and a questionnaire survey was conducted to assess the effectiveness of the teaching innovations among clinical medical master's students. ResultsA total of 699 clinical medical master's students completed the survey questionnaire. 94% of students expressed satisfaction with the course, 96% believed that the relevant knowledge covered in the course met the requirements of clinical research, 94% felt that their research capabilities had improved after completing the course, and 99% believed that the course helped them publish academic papers and complete their master's theses. ConclusionStudents recognized the teaching innovations in the course, which stimulated their initiative and enthusiasm for learning, improved the teaching quality of the course, and enhanced the research capabilities of the students.