High-Dimensional Projected Clustering for Learner Competency Analysis in Medical Training Programs

Abstract

Technology-enhanced learning (TEL) has been employed in healthcare systems to deliver case-based practical knowledge and prognostic assessments for healthcare professionals. Each medical specialty requires a unique set of skills, knowledge, and experience to ensure optimal patient care, necessitating the identification and assignment of the best medical professionals. We propose a recommendation system that facilitates the selection and assembly of top medical professionals for life-threatening cases on the basis of assessments conducted on a TEL platform for critical care skills. The key contribution of this work is the process of forming a team of the most qualified medical professionals for a critical care case. The system helps learners refine their skills by identifying areas that need improvement and providing timely assistance to struggling learners. A variant of content-based filtering is employed by leveraging the projected clustering (PROCLUS) algorithm on assessment data. The recommendation system identifies clusters of similar performance patterns among healthcare professionals on the basis of key features. Additionally, weak learners deficient in crucial healthcare areas are identified, and the model recommends the most qualified professionals for specific critical care cases.

Keywords

• Content-based filtering
• critical care skill development
• healthcare professional selection
• learner competency analysis
• medical training programs
• performance pattern identification