Advancing medical affair capabilities and insight generation through machine learning techniques.

Area of Science:

• * Data Science and Artificial Intelligence in Pharmaceuticals
• * Machine Learning Applications in Healthcare

Background:

• * Pharmaceutical companies increasingly use machine learning (ML) for research, drug development, and medical affairs.
• * AI tools like chatbots are mature in other industries but underutilized in pharma for healthcare professional (HCP) engagement.
• * Pharmacists can play a key role in developing and implementing these technologies.

Purpose of the Study:

• * To develop an AI-powered tool to optimize the analysis of unsolicited medical information (MI) from HCPs.
• * To enhance medical affairs efficiency and provide actionable intelligence for targeted content delivery.

Main Methods:

• * Development of a Python-coded tool named MUFASA (Medical Information Data Uses For AI Semantic Analysis).
• * Utilizes Sentence Transformer library, clustering (HDBSCAN), and visualization techniques.
• * Applies AI to analyze unsolicited MI data for semantic understanding and theme discovery.

Main Results:

• * MUFASA enhances medical affairs through semantic search, cluster analysis, and visualization.
• * Improves efficiency in handling MI and Medical Science Liaison (MSL) cases via 3D vector mapping and clustering.
• * Facilitates staff training, ensures response consistency, and mitigates compliance risks.
• * HDBSCAN algorithm identifies actionable themes from large datasets; visualizations support evidence-based decisions.

Conclusions:

• * Significant opportunities exist at the intersection of healthcare and data science for pharmaceutical companies.
• * Leveraging ML for analyzing unsolicited inquiries from HCPs can optimize medical affairs processes.
• * Harnessing machine learning techniques on abundant data can drive innovation and improve engagement.