PROJECT 01 / Healthcare AI · Predictive analytics
PharmaBeacon.
Anticipating drug shortages with machine learning, historical FDA-related data, and explainable risk signals.
AIS xTech Challenge · Florida International University
Team member
Hackathon prototype
The problem
Drug supply disruptions create uncertainty for healthcare organizations. PharmaBeacon explores whether historical data and quantitative signals can help organizations anticipate potential shortages.
The approach
The prototype combines historical FDA-related data, predictive modeling, and risk rankings. A hospital-oriented dashboard brings together shortage risk, historical analysis, explainable signals, and data provenance.
Technical focus
The work spans Python, data engineering, statistical modeling, and machine learning. Making a risk signal understandable and traceable is part of the product concept, alongside prediction.
Outcome
Our team received second place at the AIS xTech Challenge 2026 at FIU. The result is a hackathon prototype, rather than a clinically validated decision-making system.
Limitations & next questions
Prototype evaluation does not establish clinical effectiveness. Further work would need to examine data quality, time-based validation, missed shortages, and the usefulness of signals in a real healthcare workflow.