This project has been shortlisted for the DPH 2025 Innovation Prize – Best Data Driven Innovation
Team: Meghan Conroy and Jerry Winiczek (CaptureProof)
Outline: CaptureProof is redefining how patient data is generated, analyzed, and acted upon – using just photo or video from any smartphone or tablet. Our HIPAA-compliant platform transforms everyday devices into precision diagnostic tools, enabling asynchronous, AI-powered triage across a spectrum of high-impact conditions including ALS, Parkinson’s, epilepsy, fall risk, post-operative care, and organ viability.
Unlike traditional tools limited to clinic settings or subjective scoring, CaptureProof uses computer vision and machine learning to extract objective motion, symmetry, and biometric data from routine patient videos. No wearables. No sensors. No proprietary hardware.
Across multiple peer-reviewed studies and specialty areas, CaptureProof has delivered measurable impact:
Falls & Aging: In over 500+ patient assessments, Balance Scan automates TUG and Sit-to-Stand analysis with >96% accuracy. Our AI captures gait, sway, cadence, and turning velocity to stratify fall risk remotely. In older adults, 50% showed high-risk metrics – even with no outward symptoms – unlocking massive prevention potential and healthcare savings.
ALS Monitoring: In clinical studies at Temple University, CaptureProof revealed motor and bulbar decline not seen in gold-standard ALSFRS-R or ROADS scores. Video-captured facial asymmetry, slowed sit-to-stand, and reduced word-per-breath preceded clinical detection – enabling earlier intervention. 100% of ALS patients with “normal” scores had measurable motor impairments by video.
Parkinson’s Remote Trials: A Weill Cornell study using CaptureProof to administer a visual-only, short UPDRS at home showed 99.7% of video uploads were usable and correlated tightly with in-clinic ratings. High compliance (96.4% video adherence) and strong inter-rater reliability (ICC ≥ 0.91) proved feasibility for decentralized trials and longitudinal monitoring.
Epilepsy Diagnosis: In a prospective study of 172 patients, asynchronous video review using CaptureProof led to 5.45× diagnostic accuracy, 99% faster time-to-diagnosis, and reduced unnecessary EEG referrals. Clinicians made accurate calls in under 24 hours using patient-captured seizure videos .
Post-Op Recovery: A randomized control trial in knee replacement patients found no difference in clinical outcomes between asynchronous tele-rehabilitation via CaptureProof and traditional PT. However, the digital arm saw a 60% reduction in hospital resource use, high patient satisfaction (mean 4.5/5), and easier access to care – especially among older, mobility-restricted populations.
Liver Transplantation: In France, our platform accurately quantified steatosis from intra-operative donor liver images, achieving a 0.86 correlation with biopsy. This visual AI tool enabled non-invasive, real-time triage – essential in the time-critical transplant setting.
What makes CaptureProof extraordinary is its universality: anyone, anywhere with a phone or tablet can generate high-quality clinical data. Our platform ensures end-to-end encryption, potential for anonymization at capture, and region-specific processing (HIPAA/GDPR/PDPA-compliant). Video becomes not just documentation, but actionable evidence, analyzed by AI models that grow more powerful with every upload through ongoing training.
CaptureProof is already used in hospitals, homes, trials, and low-resource settings. We’ve proven this technology isn’t just novel – it’s scalable, inclusive, and transformative.
