HYBRID OPENAI AI SYSTEM FOR MEDICAL ACADEMIC WORKFLOW OPTIMIZATION
Abstract
Medical education institutions often face challenges due to semi-manual and fragmented academic workflows, particularly in enrollment, academic evaluation, and student progression tracking. This study aimed to assess current academic workflows and develop an AI-enhanced framework to optimize enrollment processing, evaluation, and student progression monitoring. Using a descriptive-developmental mixed-methods design, data were gathered from 212 respondents through surveys, semi-structured interviews, document analysis, and workflow observation. Results revealed that existing processes remain predominantly manual, leading to repeated document verification, encoding redundancy, and limited system integration.The study proposes an AI-assisted architecture incorporating automated information extraction using OCR, rule-based validation, deficiency detection, and intelligent recommendation generation. The proposed system further leverages the hybrid AI approach for the system flow using OpenAI: Rule-based engine + Predictive ML + LLM. Whereas, Random forest prediction model was utilized to maximize reusing student’s historical data. The developed framework demonstrates strong potential to reduce processing time, minimize errors, enhance evaluation quality, and enable real-time student progression tracking while maintaining data security. This research contributes to a practical model for AI-driven digital transformation in medical education environments.
