Reply: Humanitarian Facial Recognition for Rare Craniofacial Malformations

Abstract

To the Editors: Many thanks to Umutoni and Dadson1 for their comments on our article dedicated to the use of facial recognition algorithms in humanitarian conditions.2 The algorithms we develop are the result of academic research and will be available free of charge to medical professionals as a mobile application. We expect the first prototype early 2025. Collaborations with charities such as Operation Smile, Smile Train, and Mercy Ships will improve patient care and constitute great proofs of concepts for the device. Furthermore, such collaboration will contribute to augment the datasets on which we train the algorithms, within the framework of European Union regulations on the secondary use of medical data for research purposes.3 Diagnostic wandering is a critical issue in low- to middle-income countries and also in high-income countries such as France,4 where delays before diagnosis exceed 5 years in 25% of cases of rare diseases. In this context, algorithms such as Artificial Intelligence for Dysmorphology (AIDY) deserve an international diffusion with applications extending from congenital disorders to acquired conditions with facial expression, such as cardiac, dermatologic, or endocrinologic diseases. Furthermore, AIDY toolkits can also help to better characterize the facial phenotype within a syndrome, for instance, regarding genotype/phenotype correlations5; establish similarity maps between conditions; and assess facial feature modifications during growth or before/after a medical or a surgical treatment.6 Artificial intelligence–based tools relying on massive multimodal datasets built over years in rare disease reference centers concentrate and order tremendous amounts of fundamental knowledge and clinical experience. These innovative tools allow us to diffuse the expertise of specialists with unique training to the international community of health professionals, contribute to improve patient care, standardize treatment plans by providing quantitative result assessment, and ensure accelerated referral to dedicated centers. Patients with rare diseases currently face challenging situations in many confict zones or regions confronted with natural disasters, where their management is often considered as a secondary priority. The recent Russian attack on Ohmatdyt National Specialized Children's Hospital, the largest pediatric hospital in Ukraine, on July 8, 2024, injured and killed health professionals and patients, and destroyed facilities critical for the management of rare diseases. In this perspective, digital tools such as AIDY, which allow the securing and sharing of exclusive clinical knowledge, are crucial to keep high standards of care even in the most adverse conditions. DISCLOSURE The authors have no financial interest to declare in relation to the content of this article.

Ähnliche Arbeiten