Aidoc, the New York-based clinical AI developer, has announced the submission of its CARE1 foundation model for FDA review, marking a milestone in the company’s multiyear development of the Clinical AI Reasoning Engine (CARE) framework, which aims to enhance diagnostic accuracy and efficiency in healthcare.

CARE1, built on extensive multimodal training using millions of medical exams, is designed for use in CT imaging and aims to support clinicians in addressing critical diagnostic challenges by reducing delays, improving workflow, and enhancing patient outcomes. While still under investigation, preliminary internal tests have shown promising results across multiple pathologies.

"Eighteen months ago, we embarked on an ambitious journey to redefine how clinical AI is developed," said Michael Braginsky, Aidoc’s CTO. "The CARE framework will enable high-performance AI capable of detecting a wide range of conditions in real time, setting a new standard for patient care."
Aidoc’s solutions are used in over 1,000 medical centers worldwide and include 17 FDA-cleared AI algorithms.

According to Aidoc, CARE1 leverages a data set of unprecedented scale, reflecting real-world clinical environments, to ensure consistency and reliability. Significant financial investment has also been pivotal to the development of CARE1, allowing for comprehensive data processing, training, and testing. Aidoc notes that the model uses only a portion of its vast data set, suggesting room for future advancements.

The company has already submitted its first real-world derivative of the CARE1 Foundation Model for FDA 510(k) review. The initial application leveraged CARE1 in Aidoc's retrained rib fractures triage module with the addition of a pretraining layer.

"We decided to start this shift by transforming the rib fractures triage module to be CARE1-enabled, allowing us to begin learning how to measure, validate and ultimately bring foundation model-enabled algorithms to production in clinical settings," added Braginsky. "This marks a pivotal shift in AI development at Aidoc. We've fully committed to foundation model technology because it's clear this approach surpasses anything we've seen before. At the same time, we are steadfast in ensuring this transition is done with precision, focusing on safety, rigorous validation and a comprehensive understanding of this groundbreaking model's behavior to ensure it delivers meaningful clinical impact."

Back to HCB News

Artificial Intelligence Homepage