Accurately Inferring Missing Factor VIII Concentrate Dosing Information from Health-Record Data in Patients with Haemophilia A

This poster, presented at the ISTH 2026 Congress (International Society on Thrombosis and Haemostasis), showcases research conducted within the PHEMS project to address missing treatment information in electronic health records (EHRs) for patients with haemophilia A. The study addresses the challenge of missing treatment information by developing a methodology to infer factor VIII dosing from routinely collected clinical data. Using data from children with haemophilia A treated at Erasmus MC Sophia Children's Hospital, the approach was evaluated for its ability to reconstruct treatment histories and generate more complete datasets for research. By improving the quality and usability of EHR data, this work supports the development of privacy-preserving machine learning models for rare diseases and contributes to the broader PHEMS mission of advancing federated paediatric health data research across Europe.