Skeletal dysplasias are rare diseases that collectively affect a large number of people. The latest Nosology and classification of genetic skeletal disorders includes a total of 771 distinct diseases associated with 552 genes.
The clinical diagnosis of skeletal disorders usually requires the identification of different patterns in skeletal radiographs. However, the large number of these disorders and their rarity mean that even experienced clinicians may have only encountered a small fraction of them. Referring to X-ray catalogues is a conventional approach to tackle this issue, but it can be both time-consuming and subjective, and may potentially result in incorrect or delayed diagnoses.
Bone2Gene AI aims to identify the unique imaging patterns linked to various bone disorders and support clinicians in the diagnostic process. It would shorten the diagnostic odyssey and reduce the difficulties that patients with rare bone diseases and their families endure.
Learn moreBone2Gene is under development and so far has over 2300 X-rays from over 1000 individuals with 30 different bone disorders, and these numbers are growing.
The performance of the preliminary Bone2Gene classifier across 11 distinct classes (10 disorders plus a class with no bone disorder).