Analyzing Patterns in Anesthesiology Residents' Exam Performance Using Data Mining Techniques.

Abstract

Background: Residency is a critical period in the development of medical professionals. It provides hands-on training and exposure to various medical specialties, enabling residents to improve their skills and achieve expertise in their chosen field.

Objectives: This study aimed to extract frequent patterns in annual and board examination performance among anesthesiology residents by analyzing results from the department's weekly exams.

Methods: This cross-sectional study was conducted in the Department of Anesthesiology, Critical Care, and Pain Medicine (DACCPM) from September 2022 to June 2023. Weekly intra-group exams were administered at the university's electronic exam center for residents in their first to fourth years (CA-1 to CA-4), with a total of 61 participants. Learner grades were categorized as excellent (A), good (B), average (C), poor (D), and inferior (E). The Apriori algorithm was employed to extract frequently repeated patterns in these exams and compare them with results from the final national examination.

Results: A total of 24 exams were conducted, with all 61 residents participating. The most frequent patterns, identified with a minimum support of 0.41, revealed that residents generally achieved average scores in exam 7 and very poor scores in exams 1 and 5. The study found a statistically significant relationship between residents' scores in in-training examinations (ITEs) and their national examination performance.

Conclusions: Analyzing residents' exam performance using frequent pattern recognition can help identify their strengths and weaknesses. Faculty members can utilize these insights to better plan curricula and enhance the quality of education.