# AutoChamber: an FDA-Designated Breakthrough AI Add-on to Coronary Artery Calcium and Lung Cancer Screening CT Scans to Flag Patients at High Risk of Atrial Fibrillation, Stroke and Heart Failure

> **NIH NIH R43** · HEARTLUNG CORPORATION · 2024 · $294,023

## Abstract

Project Summary/Abstract
AutoChamber is part of an initiative spearheaded by HeartLung.AI, in collaboration with a team of esteemed
physician researchers from leading US academic institutions. The AI tool is designed to opportunistically
screen existing chest CT scans stored in hospital PACS or new scans obtained for any medical evaluations
such as coronary artery calcium scan, lung cancer screening or diagnostic work ups following accidents or
pneumonia and such, to detect asymptomatic enlarged cardiac chambers and thick left ventricular wall without
any X-ray contrast enhanced agent. The human eye cannot distinguish between the inside of cardiac
chambers and the cardiac wall without contrast enhanced agent whereas, the AutoChamber AI which was
trained using contrast-enhanced cardiac CT scans can accurately detect and measure cardiac chambers
volume and left ventricular wall mass. The project underscores a significant stride towards addressing the
unmet needs in early detection and prevention of heart failure and stroke, potentially saving lives, and reducing
healthcare costs. It aligns with the mission of fostering innovative solutions that offer opportunistic diagnoses
using existing CT scans or CT scans obtained for other reasons unrelated to cardiovascular disease. Over 76
million CT scans were performed in the United States in 2019, an estimated 230 CT procedures per 1,000
people with chest CT accounting for 12.7 million that can be used for AutoChamber™ screening.
AutoChamber™ is the newest AI technology that has received a Medical Device Breakthrough Designation
from the FDA as recently as August 30, 2023.
Building upon the encouraging results gleaned from the MESA database, the project aims to extend its
research to the Framingham Heart Study database, the largest prospective cohort study of cardiovascular
diseases with a rich repository of thousands of CT scans to further validate the performance of
AutoChamber™.
The project delineates three specific aims:
Aim 1: To apply AutoChamber™ on CAC scans from the Framingham database and compare the cardiac
chamber volume measurements with those obtained from contrast-enhanced MRI.
Aim 2 To evaluate the performance of AutoChamber™ in predicting incidents of atrial fibrillation and stroke,
benchmarking against existing standards like the CHARGE-AF Score and CHADS-VASc Score.
Aim 3 To assess the capability of AutoChamber™ in detecting prevalent left ventricular hypertrophy and
asymptomatic left ventricular dysfunction, and predicting incidents of heart failure, using parameters like left
ventricular and left ventricular wall volumetry.

## Key facts

- **NIH application ID:** 10922647
- **Project number:** 1R43HL174376-01
- **Recipient organization:** HEARTLUNG CORPORATION
- **Principal Investigator:** MORTEZA NAGHAVI
- **Activity code:** R43 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $294,023
- **Award type:** 1
- **Project period:** 2024-09-27 → 2025-09-26

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10922647

## Citation

> US National Institutes of Health, RePORTER application 10922647, AutoChamber: an FDA-Designated Breakthrough AI Add-on to Coronary Artery Calcium and Lung Cancer Screening CT Scans to Flag Patients at High Risk of Atrial Fibrillation, Stroke and Heart Failure (1R43HL174376-01). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/10922647. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*