Home / AITHERO
Private Sector
Stage 1, Stage 2

AITHERO

Horus ML
‹ Back to Open Call Winners
AITHERO brings Aitheroscope, an AI medical software that detects asymptomatic atherosclerosis from a single retinal image, to the AI-on-Demand Platform. Aitheroscope reaches over 93% sensitivity, far above the risk calculators used in primary care, and is already used in 21 healthcare centres. In the programme, it will be deployed on European infrastructure through CKS DEV Lite, aligned with the EU AI Act using TAI Scan, and validated with reproducible testing workflows built with the AI Testing Framework. The goal is to make it available on the AIoD Marketplace and prepare a real-world deployment with a private healthcare adopter, helping prevent heart attacks and strokes across Europe.

Background

Cardiovascular disease is the leading cause of death in Europe, and atherosclerosis is its main underlying cause. It develops silently for years, so many people only find out after a heart attack or stroke. In adults aged 35 to 55, around 60% already have atherosclerosis, and about 35% of the population carries it undiagnosed. Primary care relies on risk calculators that detect only around 40% of cases, so most people with the disease are classified as low risk and receive no preventive care. Imaging tests that confirm it, such as vascular ultrasound, are too costly and time-consuming to use for population screening.

Proposed solution

Aitheroscope is an AI-based medical software that detects asymptomatic atherosclerosis from a single retinal image, a quick and non-invasive test that can be performed in primary care with a standard retinal camera. The retina is the only place in the body where blood vessels can be observed directly, and changes in its microvasculature reflect the state of the arteries. Aitheroscope combines convolutional neural networks and Vision Transformers in an ensemble model that analyses these patterns and returns a result in under 10 seconds, with an interpretable report for the clinician. It can work from the image alone or be enhanced with a few clinical variables. Clinically validated and published in European Heart Journal – Digital Health, it reaches over 93% sensitivity and over 85% specificity, with equivalent performance in women and men, far above current risk calculators. Positive cases are referred for confirmation and preventive management. Aitheroscope is already used in 21 healthcare centres across Spain, Mexico and Chile, with more than 12,000 patients screened, and is delivered as a web application or via API, fitting easily into existing clinical workflows.

Objectives

  • Bring population-level screening of asymptomatic atherosclerosis to primary care across Europe, helping prevent heart attacks and strokes
  • Integrate Aitheroscope into the AIoD Platform, deploying it on European infrastructure through CKS DEV Lite
  • Align the product with the EU AI Act using TAI Scan
  • Build reproducible testing workflows with the AI Testing Framework and share a reusable healthcare AI pattern with the community
  • Publish Aitheroscope on the AIoD Marketplace to reach healthcare providers and insurers across Europe
  • Prepare a real-world deployment with a private healthcare adopter in Stage 3

Horus ML

Spain
Spanish healthtech company Horus ML makes cardiovascular screening accessible through Aitheroscope, an AI medical device detecting asymptomatic atherosclerosis from a single retinal image. Combining clinical validation with sovereign AI, it brings hospital-grade prevention to primary care across 21 centers in Europe and Latin America.
Stay informed on AI opportunities
Get updates on new solutions, market trends, and adoption strategies delivered to your inbox.
The DeployAI project has received funding from the Digital Europe Programme (DIGITAL) under grant agreement №. 101146490 - DIGITAL-2022-CLOUD-AI-B-03
Stay informed on AI opportunities
Get updates on new solutions, market trends, and adoption strategies delivered to your inbox.
The DeployAI project has received funding from the Digital Europe Programme (DIGITAL) under grant agreement №. 101146490 - DIGITAL-2022-CLOUD-AI-B-03