Turn Anatomy Into Intelligence

We build AI-based digital twins of patient anatomy to predict device performance and optimize interventions in a personalized manner.

Request Demo

Clinical Problem

Complex vascular devices are often optimized for average patients.

Modern vascular care involves complex anatomy and high-stakes decisions, yet clinicians and researchers often work with limited geometric or physiological insight.

Patient variation matters, and device optimization should be personalized to each patient. We make it possible for computational models to run on individual 3D models of patient anatomy delivering profound insight.

See the Approach

How It Works

A clear workflow from scan to insight

CT, MRI, and vascular imaging data are ingested as the starting point for a patient-specific modelling workflow.

We identify lesions that are of most interest to the surgeon and ensure we standardize that segment of the scan to suit our predictive capabilities.

We map these preprocessed images into a Graph Neural Network (GNN) that can learn the geometry of each patient's anatomy and transcribe it into a digital twin. We then simulate device performance in this digital twin to predict outcomes.

Results are organized into outputs that help teams study anatomy, flow behavior, and treatment questions with greater clarity. Our focus is to provide a third eye to the surgeon to help them make informed decisions.

Benefits for Surgeons

Our Main Advantages

Decrease in OR Time

We work on preventing complications within the Operating Room by informing them of the personalized risks associated with each patient's anatomy. This allows more efficient planning.

Prediction of Risk Factors

With more insights derived prior to the procedure, surgeons can make more informed decisions and reduce the risk of complications.

Improved Device Production

Consumers can now be confident with the device they are using with, and companies can now produce devices that are more personalized to each patient.

Our First Case Study

EVAR: Preventing TIa Endoleaks with Digital Twins

Endovascular aneurysm repair depends on vessel geometry, landing zones, and flow behaviour.

Digital twins help teams study how individual anatomy may influence seal quality, device fit, and the risk of complications such as type Ia endoleaks before intervention.

Digital Twins to Reduce Sealing Failures

We use an AI Digital Twin to extract the proximal neck of the aneurysm and demonstrate which regions are less likely to seal properly. Surgeons can account for this on initial insertion, preventing wasted time on intraoperative adjustments.

Discuss This Case