Cardiovascular intelligence from mammography

Breast cancer is feared.
Heart disease is still underestimated.

Every mammogram may hold a cardiovascular signal.

About 1 in 8 women in the US will develop breast cancer in her lifetime. Cardiovascular disease remains the No. 1 killer of women and causes about 1 in 3 female deaths each year. The screening infrastructure already exists. The missing layer is cardiovascular intelligence.

The Problem

Current risk tools miss too many women.

Standard cardiovascular risk models were calibrated on populations that underrepresent women. Important signals exist in routine imaging, but are rarely translated into prevention.

69%
Silent risk
Of women who experienced sudden cardiac death in a large prospective cohort, 69% had no prior history of cardiac disease.
Circulation, 2003
~50%
Underestimated risk
The Framingham Risk Score has been shown to underestimate coronary heart disease risk by approximately 50% in older women, compared to 8% in older men.
PLOS ONE, 2012
~40M
Missed scale
Approximately 40 million screening mammograms are performed each year in the United States, yet cardiovascular signals in breast imaging are rarely converted into structured prevention.
National Breast Cancer Foundation
The Insight

Routine mammograms can reveal more than breast cancer risk.

Breast arterial calcification is visible on mammography and has been associated with cardiovascular outcomes. Recent AI-based studies show that automated BAC quantification can predict cardiovascular morbidity and mortality and add prognostic value beyond traditional risk tools.

PrevisionDX is building on this foundation by developing a composite approach that combines BAC, breast density, and AI-derived tissue features rather than relying on a single imaging marker.

Read the published evidence
The CCRI

Toward a composite cardiovascular risk index from routine mammography.

The CCRI is being developed as a multi-biomarker, explainable, longitudinal framework. Designed to extend the value of existing screening, not to replace clinical judgment.

01 / STEP
Routine mammogram
Already part of women's preventive care. No additional imaging required.
02 / STEP
Imaging biomarkers
BAC, breast density, and AI-derived tissue features extracted from the existing exam.
03 / STEP
Composite index
Designed to integrate multiple cardiovascular signals into one explainable framework.
04 / STEP
Clinical workflow
Built for future integration across radiology, primary care, and cardiology.
Multi-biomarker
Not BAC alone.
Explainable
Designed to show which imaging signals contributed.
Longitudinal
Built to track change as women return for screening.

No new scan. No new radiation. Built for existing mammography workflows.

Where We Are

In development with real-world clinical data.

Algorithm
Algorithm development
PrevisionDX is developing the CCRI engine, beginning with BAC detection and expanding toward multi-biomarker fusion.
Data
Retrospective cohort
A partnered health system cohort includes tens of thousands of women, hundreds of thousands of mammograms, cardiovascular outcomes, and long-term EHR follow-up.
Validation
External validation pathway
PrevisionDX is preparing external validation with US academic and clinical collaborators.
Where We Stand

BAC detection is emerging. Composite cardiovascular intelligence is the next step.

Today's market is beginning to recognize the cardiovascular value of mammography. Existing tools focus on BAC detection, breast density, or breast cancer risk.

PrevisionDX is developing a composite cardiovascular framework built from outcomes data and designed for longitudinal risk intelligence.

Team

Built by imaging AI and healthcare operators.

Founder & CEO
David Amsel

Healthcare operator with experience building and scaling regulated healthcare businesses, including multi-state care delivery, federal nutrition programs, compliance-heavy operations, payer-facing services, and custom healthcare workflow software.

Co-Founder & CTO
Orna Bregman-Amitai

Computer vision and medical imaging AI leader with decades of experience across healthcare imaging, applied AI, and product development. Former senior leader at Zebra Medical Vision and inventor on multiple granted US patents.

Clinical & Analytics Lead
Yarden Aronson

Biomedical engineer with experience in experimental research, data analysis, and technical workflow development. Co-authored published research at Tel Aviv University.

Contact

Let's talk.

PrevisionDX welcomes conversations with imaging platforms, health systems, research collaborators, clinical leaders, and strategic partners.

Email PrevisionDX