AI-assisted case-based learning and flipped classroom to improve clinical decision-making: a randomized controlled trial in reproductive medicine

Clinical Snapshot

Key Findings

• P Value: Statistical significance claimed but specific p-values not provided

• Effect Size: Not reported in abstract

• Primary Outcome: Theoretical knowledge test scores, Mini-CEX scores, and OSCE scores

• Nnt Or Sensitivity: Not calculable from available data

• Confidence Interval: Not reported

Clinical Application

Requires access to specialised educational platforms and technology infrastructure Potentially applicable to Australian O&G training programs, though would need adaptation to local curriculum requirements and RANZCOG competency frameworks Obstetrics and gynecology residents in reproductive medicine training programs

Abstract

BACKGROUND: Efficient training of reproductive medicine clinicians is critical in the context of declining global fertility and increasing infertility. Traditional lecture‑based instruction often fails to sufficiently develop clinical decision‑making skills within limited residency rotations. Innovative strategies that integrate artificial intelligence (AI) with case‑based learning (CBL) and flipped classroom (FC) formats may enhance clinical reasoning, however rigorous evidence in reproductive medicine education remains limited. METHODS: We conducted a randomized controlled trial involving 50 obstetrics and gynecology residents at the First Hospital of Jilin University. Participants were randomly assigned to an AI‑assisted CBL+FC group or a traditional lecture control group. The AI‑assisted CBL+FC group completed pre‑class interactive case work with virtual standardized patients on the DoctorU platform and case analyses on the Superstar Learning platform, followed by interactive in‑class discussions. Primary outcomes included post‑course theoretical knowledge tests, Mini‑Clinical Evaluation Exercise (Mini‑CEX), and Objective Structured Clinical Examination (OSCE) scores. Secondary outcomes assessed learner motivation, clinical thinking, self‑directed learning, and perceived course effectiveness using a 5‑point Likert scale. RESULTS: Baseline characteristics were comparable between groups. After the intervention, the AI‑assisted CBL+FC group achieved significantly higher theoretical test scores than the control group. The AI‑assisted CBL+FC group also demonstrated superior overall clinical competence in Mini‑CEX assessments and higher OSCE total scores. Participants in the AI‑assisted CBL+FC group reported greater improvements in learning motivation, clinical reasoning, self‑directed learning, and perceived course effectiveness. CONCLUSIONS: The AI‑assisted CBL+FC instructional model significantly enhances theoretical knowledge, clinical decision‑making skills, and learner engagement among reproductive medicine residents. This blended learning model offers an efficacious and generalizable methodology for training practitioners to address the evolving clinical requirements within contemporary fertility care.