PETRA (Pedal Elevated Temperature Risk Assessment) is a thermal-AI screening tool designed to predict diabetic foot ulcers 4–5 weeks before ulceration.
PETRA is a thermal-AI screening tool designed to predict diabetic foot ulcer risk 4–5 weeks before visible ulceration, enabling earlier intervention in low-resource clinical settings.
I co-founded and developed PETRA through field-based work in India, integrating biomedical engineering with global health design principles. The system uses a smartphone-mounted thermal camera to capture plantar temperature maps and applies calibration-free contralateral asymmetry analysis to detect clinically significant hotspots (≥2.2 °C) while explicitly restricting analysis to anatomically valid foot regions.
I designed and managed the core technical pipeline, including thermal data processing, left–right foot segmentation, hotspot detection, and model validation workflows. The system is being developed in close collaboration with clinicians across India to ensure clinical relevance, usability, and real-world feasibility.
PETRA is currently advancing toward expanded clinical testing, peer-reviewed publication, and regulatory alignment as a low-cost, high-sensitivity triage tool intended for frontline and primary care settings.
The goal is simple but urgent: reduce preventable amputations by giving healthcare workers an early-warning system that functions reliably in real-world conditions—without calibration, specialized infrastructure, or expert interpretation.
