Healthtech funding
DiagFit's €6.5 million bet on the data you already collect
DiagFit closes a €6.5 million Series A to deploy an AI platform that combines imaging, health records, and wearable data into a single risk score for chronic diseases. The company claims its model outperforms standard screening by 12 percentage points.
Paris-based healthtech startup DiagFit has closed a €6.5 million Series A funding round, the company announced today. The round was co-led by Bpifrance, through its French Tech Accélération programme, and Heartcore Capital. Existing angel investors, including former Doctolib CTO Jean-Charles Samuelian-Werve, joined in.
The company plans to use the fresh capital to push its proprietary AI diagnostic platform into more French hospitals and clinics. It's also looking to double its engineering team to 40 people and kick off clinical trials necessary for CE marking under the European Medical Device Regulation (MDR).
Platform and use case
At the heart of DiagFit's tech is a multi-modal machine learning pipeline that ingests and fuses three types of data: medical imaging, think X-rays, CT scans, and ultrasound, structured electronic health records including lab results and vitals, and constant streams from consumer wearables like smartwatches and fitness bands. The ensemble model churns out risk scores and diagnostic suggestions for five target chronic conditions: type 2 diabetes, hypertension, chronic kidney disease, non-alcoholic fatty liver disease, and early-stage heart failure.
According to the company, its system hits an area under the ROC curve of 0.91 when detecting pre-diabetic states. That comes from a retrospective validation study on 12,000 anonymized patient records from two French university hospital centers. Against the same cohort, the model beat standard screening tools such as the Finnish Diabetes Risk Score, or FINDRISC, by 12 percentage points.
Investor rationale
“What sets DiagFit apart is its pragmatic approach to data integration. They are not chasing a single biomarker or imaging modality, they combine everything a hospital already collects into a single, clinically actionable score. This matches the reality of primary care, where physicians rarely have a complete picture on one device,”said Marie Lefèvre, a partner at Heartcore Capital, in a statement.
Bpifrance's bet comes via its French Tech Accélération programme, which targets deeptech startups at the Series A stage that are focused on industrial scalability.
Market context
The global AI health diagnostics market is projected to hit $34.3 billion by 2028, per recent estimates. Chronic diseases account for 74% of all deaths worldwide, according to the World Health Organization, fueling demand for earlier and cheaper detection tools. DiagFit squares off against a crowded field that includes French startup Incepto and US-based Viz.ai, but it tries to stand out with its multi-modal, primary-care-first angle rather than focusing strictly on radiology or cardiology.
The company was founded in 2022 by Dr. Amina Belkacem, CEO and former cardiologist at Hôpital Européen Georges-Pompidou, and Thomas Roussel, the CTO and an ex-DeepMind research engineer. They previously raised a €1.2 million seed round in early 2023.
DiagFit aims to submit its first device for CE marking in early 2026, targeting a commercial launch across three European countries by late 2026.