Feasibility phase

Every year around 500,000 children worldwide are born with genetic bone dysplasia, which is associated with over 500 different genes. Due to their rarity, even experienced clinicians have never seen many of these disorders. For more than half of patients, it therefore takes around five to seven years before a diagnosis is made, or the cause is never identified. A causal therapy is therefore not possible. On the other hand, there are current developments in effective drug therapies that require the earliest possible diagnosis.

Accelerating Diagnosis of Rare Genetic Bone Diseases with AI

The Bone2Gene project aims to use innovative artificial intelligence to revolutionise access to effective treatment methods and change the diagnostics market. During the funding period, experts from the fields of big data, AI, medical genetics and paediatric endocrinology from the Universities of Bonn and Magdeburg will work together. They will build a comprehensive database and train the Bone2Gene AI prototype to recognise and distinguish patterns in medical images. Bone2Gene aims to shorten the time to diagnosis and reduce the overall cost of the diagnostic process. Bone2Gene also aims to provide a simple screening tool for the recruitment of patients for clinical trials.

Conceptual phase

About one in a thousand children is affected by a rare bone disease (skeletal dysplasia). The majority of known skeletal dysplasias are already associated with pathogenic variants of various genes. Clinical diagnosis usually requires the identification of patterns in X-ray images of the skeleton. However, due to their large number and rarity, even experienced clinicians have never seen many of the disorders. Traditional approaches are rather subjective and very time-consuming and can lead to delayed or even incorrect diagnoses. Rare disease patients wait an average of five years and receive at least one misdiagnosis before receiving a confirmed diagnosis – a process often referred to as a ‘diagnostic odyssey’.

The overarching goal of Bone2Gene’s artificial intelligence is to recognise the characteristic imaging patterns of diseases and to support clinicians in the diagnostic process. Bone2Gene would shorten the diagnostic odyssey, reducing the overall cost of the diagnostic process for the healthcare system. Its unique selling point is the detection of various skeletal dysplasias, which both represents a competitive advantage over the competition and offers interesting exit strategies. In the upcoming exploratory phase a concept for the best possible use of AI for the analysis of skeletal dysplasia is to be developed. To this end an IP strategy will be developed, the market and competition will be analysed and an implementation plan for the subsequent feasibility phase will be drawn up.

Detailed information about all phase 4 projects