aidy.eu

AIDY

Artificial intelligence for rare-disease diagnosis. From a simple front and profile photo, the AIDY app directs patients with rare diseases toward the right care pathway, with a geolocated referral.

Visit aidy.eu →

The problem

5 years of diagnostic wandering

According to the WHO, there are between 6,000 and 8,000 different rare diseases, affecting about 300 million people worldwide — about 3 million in France. 80% are genetic in origin and often affect children. On average, patients wait 5 years and see 8 doctors before getting a diagnosis.

750+

genetic syndromes detectable with distinctive facial features

90%

average detection accuracy

50K+

face images analyzed in the database

How it works

Photo analysis in seconds

Secure sign-in via the e-CPS portal or by invitation. The user photographs the face from the front and in profile in good lighting; the algorithm compares facial traits against AIDY's database and suggests a list of likely syndromes ranked by probability, with a geolocated referral to expert centers.

GDPR-compliant, hosted on HDS-certified (Hébergeur de Données de Santé) French servers; photos are not stored on AIDY's servers, only on the user's phone.

Medical device pending certification — does not replace a medical diagnosis.

Research

Synthetic faces and privacy

Teaching and diagnosing rare diseases relies on observing characteristic facial traits — but sharing these images creates a dilemma between transmitting knowledge and protecting patients. AIDY is developing a synthetic-face-generation method: from a small number of real diagnosed-patient photos, an AI learns the disease's characteristic traits and then generates hundreds of synthetic face variations — faithfully reproducing the phenotypic traits without reproducing any real patient's identity.

Team

Who's behind AIDY

Antoine Ferry

President of the association

Roman Hossein Khonsari

Surgeon, researcher at Institut Imagine

Nicolas Garcelon

Research Director, Institut Imagine

Olivier Lienhard

CTO

Quentin Hennocq

Researcher

Thomas Bongibault

Researcher

Ahmed Zaiter

Researcher

Publications

Research and innovation

  • Evaluation of a New Inclusive Next-Generation Synthetic Face Tool for Dysmorphology.

    Benichou L, Breton L, Garcelon N, et al., Khonsari RH. 2026. American Journal of Medical Genetics.

  • Next Generation Phenotyping and Synthetic Faces in Coffin Siris Syndrome.

    Hennocq Q, Lienhard O, Rao D, et al., Khonsari RH. 2024. Clinical Genetics.

  • Humanitarian Facial Recognition for Rare Craniofacial Malformations.

    Hennocq Q, Bongibault T, Garcelon N, Khonsari RH. 2024. Plastic & Reconstructive Surgery – Global Open.

  • Artificial intelligence-based diagnosis in fetal pathology using external ear shapes.

    Hennocq Q, Garcelon N, Bongibault T, et al., Khonsari RH. 2024. Prenatal Diagnosis.

  • AI-based diagnosis and phenotype–genotype correlations in syndromic craniosynostoses.

    Hennocq Q, Paternoster G, Collet C, et al., Khonsari RH. 2024. Journal of Cranio-Maxillofacial Surgery.

  • Next generation phenotyping for diagnosis and phenotype–genotype correlations in Kabuki syndrome.

    Hennocq Q, Willems M, Amiel J, et al., Khonsari RH, Garcelon N. 2024. Scientific Reports (Nature).

  • AI-based diagnosis in mandibulofacial dysostosis with microcephaly using external ear shapes.

    Hennocq Q, Bongibault T, Marlin S, et al., Khonsari RH. 2023. Frontiers in Pediatrics.

  • An automatic facial landmarking for children with rare diseases.

    Hennocq Q, Bongibault T, Bizière M, et al., Khonsari RH. 2023. American Journal of Medical Genetics.