fairfolio
Saved Projects

ThermaScreen: Affordable AI-Based Thermal Signal Conditioning and Area-Location Analysis for Surgical Site Monitoring

CSEF · 2026 Medicine & Physiology (Track 2) (Senior Division)

Guhan Senthil

Overview

ThermaScreen aims to develop and validate an affordable AI-based spatial thermal imaging system for preventive home-based inflammatory screening, with surgical wound monitoring as its primary clinical application. This project tests the hypothesis that a low-cost 32×32 pixel infrared thermal sensor (target cost ~$60), enhanced through AI-driven signal conditioning and spatial area–location analysis, can detect localized inflammation associated with surgical site wounds. A prototype was engineered for non-invasive thermal screening with integrated remote monitoring capability. Unlike conventional systems which operate at larger stand-off distances, ThermaScreen uses a controlled close-proximity scans (roughly at a safe ~4 cm from the wound surface), improving spatial resolution, minimizing environmental interference, and enhancing temperature sensitivity within this compact field of view. AI-based denoising and baseline model was created to generate conditioned ΔT (temperature deviation) maps, followed by connected-component spatial analysis to determine area of inflamed zone, geometry, and location relative to the surgical site. The reporting system allows for home-based monitoring by patients and real-time clinician review. Validation of ThermaScreen was conducted in three phases. A Synthetic Data Simulation Study (N > 100,000 samples) validated algorithmic performance with controlled random synthetic constraints verified for spatial analysis with a sensitivity of >99% and a specificity of >99%. A Laboratory Thermal Validation Study (N=41) using a controlled physical heating element (heated blanket) to simulate skin with conditioning and spatial analysis achieved sensitivity of 97% and specificity of 92%. A Controlled Clinical Trial on real surgical wounds (N=30) demonstrated sensitivity of 89% and specificity of 90%. The trial was conducted through blind testing between the surgeon’s judgment of the wound and the system’s prediction. The system additionally accurately predicted potential infections before clear visible signs appeared. All human subject procedures were conducted under approval with anonymized data handling. Chronic wound care costs the U.S. healthcare system $28–50+ billion annually, with surgical site infections contributing an additional $3–10 billion (NIH). These findings demonstrate that affordable AI-enabled spatial thermal imaging can support remote surgical wound monitoring and enable earlier clinical intervention beyond traditional healthcare settings. By significantly lowering cost barriers to medical-grade thermal systems, ThermaScreen demonstrates scalable potential for applications including oncology screening, vascular assessment, diabetic foot ulcer monitoring, and remote triage.

Competition history

  • CSEF 2026 Medicine & Physiology (Track 2) (Senior Division) · Entry S-21-26

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

Source: California Science & Engineering Fair public projects

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google