TrialBridge
The problem and why we chose it
Finding a relevant clinical trial requires comparing a patient’s diagnosis, biomarkers, treatment history, medicines and laboratory results against lengthy eligibility criteria. Patients and caregivers often have photographs of paper reports, while trial information is scattered across technical registry records.
We chose this problem to make that information easier to understand and turn it into useful questions for the treating oncologist.
Our solution
TrialBridge converts report and prescription images into an editable patient profile using Gemma 4. Users review the extracted information and its supporting quotes before searching for trials.
The application retrieves recruiting studies with recruiting or not-yet-recruiting sites in India from ClinicalTrials.gov. It filters candidates by age and sex, performs an initial screening, and checks the remaining trials against their individual eligibility rules.
Results include explanations, supporting patient facts, unresolved questions, Indian hospital locations and trial contacts. Users can also request a detailed check of trials set aside during the initial screening.
What makes it useful
- Starts from report photographs rather than requiring a structured medical record.
- Explains individual eligibility rules instead of providing only an overall result.
- Checks medication-related restrictions and considers stated dates and washout periods.
- Offers Tamil and Hindi translations of patient-facing explanations.
- Generates contact-message drafts with copy, WhatsApp and email options.
- Produces a printable summary for discussion with an oncologist.
- Includes three synthetic sample patients for demonstration.
- Supports light and dark themes and reduced-motion preferences.
Built with
AI tools
Prize Categories Entered
- Submitted
Best Use of Gemma 4
- Submitted

Best Open-Source AI Project