Skip to main content

Double funding success for Oxford team to improve detection and risk classification of myeloproliferative neoplasms

Blood Cancer UK and Cancer Research UK recognise the need and impact of artificial intelligence approaches developed by Professors Daniel Royston and Jens Rittscher for early detection and assessment of these blood cancers

Cancer Research UK and Blood Cancer UK have awarded funding to a multidisciplinary team from the University of Oxford. These awards will help to advance the AI-based methods and predict the progression of myeloproliferative neoplasms (MPNs) more accurately.

MPNs are a group of closely related disorders of the bone marrow affecting around 5000 people every year in the UK. Patients with MPNs are at higher risk of developing leukaemia, especially those with a subtype called myelofibrosis (the most severe) where this develops in >10% of patients.

Because the treatment strategy varies depending on the MPN subtype, accurate assessment of MPN type at diagnosis is crucial for optimal treatment selection. In addition to mutational and blood count analysis, morphological analysis of a bone marrow biopsy is a key component for classification. Unfortunately, this is highly subjective, reliant on qualitative observations and there is great variability even when it is done by expert haematopathologists.

There is unmet clinical need for a more accurate method for diagnosing MPN from a bone marrow biopsy. The team, led by Professor Daniel Royston (Radcliffe Department of Medicine and Oxford University Hospitals NHS Foundation Trust), have already developed artificial intelligence approaches to help pathologists extract quantitative data from scanned images of bone marrow biopsies. These algorithms will enable more accurate and reliable classification of MPN type.


Refinement of fibrosis assessment in bone marrow biopsies using the Continuous Indexing of Fibrosis (DOI 10.1038/s41375-022-01773-0) the group developed recently


With the new funding, the team now wish to refine and validate these methods with the aim of integrating them into existing NHS pathology workflows to bring about earlier diagnosis of MPNs in the clinic. The team includes Professor Jens Rittscher (Institute of Biomedical Engineering and Big Data Institute) and Professor Martin Booth from the Department of Engineering Science. The work will also be supported by the Oxford University spinout company Ground Truth Labs.

Importantly, this work will include input from patient representatives from the Oxford Blood Group. They will give feedback on the visualisation tools designed to help patients better understand what’s happening in their bone marrow and the progress of their disease.

"Better diagnostics and management of MPN disease are key priorities for our patients. Receiving this funding from Cancer Research UK and Blood Cancer UK will allow us to make significant progress towards our aim of applying our AI-based tool for more accurately diagnosing MPN type in the clinic so that patients can benefit", says research lead Professor Daniel Royston.

Read the original article on the Oxford Centre for Early Cancer Detection website