Idea
Automated platform for real-time lesion analysis in coronary angiography improving diagnostic consistency and clinical workflow efficiency.
Research Paper
Core Innovation
This paper introduces ODySSeI, integrating deep learning models with a novel Pyramidal Augmentation Scheme to improve lesion detection and segmentation robustness. It uniquely estimates lesion severity without quantitative coronary angiography, directly computing key metrics from predicted lesion geometry, enabling fast and accurate analysis across diverse datasets.
Why It Matters
Coronary artery disease diagnosis via invasive angiography is currently subjective and variable, leading to inconsistent clinical decisions. ODySSeI standardizes lesion assessment with automated, accurate analysis, reducing interpretation time and variability. This scalable solution enhances clinical workflows and supports timely, reproducible decision-making across diverse healthcare settings.
Market Size (TAM)
$2–10B TAM for cardiovascular imaging AI; $500M–$1B SAM from hospitals and imaging centers. Driven by rising cardiovascular disease prevalence and demand for diagnostic automation.
Potential Customers & Pain Points
- Cardiology departments – Need consistent and fast lesion assessment
- Hospitals – Require scalable tools for coronary disease diagnosis
- Medical imaging centers – Face variability in angiography interpretation
- Healthcare providers – Demand real-time clinical decision support.
Business Model
Subscription-based SaaS platform with tiered pricing for hospitals and imaging centers; potential licensing for integration with medical device manufacturers.
Competitive Landscape
- HeartFlow FFRct
- QAngio XA
- Medis QCA
- Canon Cardiovascular AI
Implementation Challenges
- Regulatory approval for clinical use
- Integration with existing hospital imaging systems
- Clinician trust and adoption of AI tools
- Data privacy and security compliance
Validation Strategy
- Conduct multi-center clinical trials to validate diagnostic accuracy and workflow impact
- Obtain regulatory clearances (FDA
- CE marking)
- Pilot deployments in partner hospitals for real-world feedback
- Iterate product based on clinician usability studies
Research Paper Overview
ODySSeI: An Open-Source End-to-End Framework for Automated Detection, Segmentation, and Severity Estimation of Lesions in Invasive Coronary Angiography Images
Summary
ODySSeI automates lesion detection, segmentation, and severity estimation in invasive coronary angiography images, reducing subjective interpretation and variability. It uses deep learning with a novel Pyramidal Augmentation Scheme for robustness across diverse patient cohorts and offers real-time processing with high accuracy. The framework is open-source and accessible via a web interface, supporting scalable and reproducible clinical decision-making.