Home » News » Researchers devise AI-based method of detecting response to MS treatment

Researchers devise AI-based method of detecting response to MS treatment

UCL and KCL researchers supported by the UCLH BRC have devised a new artificial intelligence-based method for detecting the brain’s response to treatment in multiple sclerosis (MS) that is substantially better than what a human expert is able to do using conventional techniques, representing potentially ‘superhuman’ performance in the task.

The researchers – led by Dr Parashkev Nachev and Prof Olga Ciccarelli, both of the UCL Institute of Neurology – hope in future this method will be used to predict an individual’s response to a drug before they start treatment, and which drug a patient should be given.

One way of assessing MS treatment response is by analysing patients’ MRI scans. At present, radiologists assess scans by counting the number of lesions and measuring lesion volumes, comparing these observations with those made on scans done before treatment started.

But the researchers’ new AI-based method of analysing scans means that regions of the brain can be analysed in much greater detail, and in a way which more closely reflects the complexity of the brain.

In the study, published in npj Digital Medicine, researchers looked at patients at the UCLH National Hospital for Neurology and Neurosurgery with relapsing-remitting MS being treated with the disease-modifying drug natalizumab.

‘Machine vision’ was used to extract from each scan an ‘imaging fingerprint’ of the state of the brain, capturing detailed changes in white and grey matter and capturing a rich set of data on how regions of the brain change over time with treatment.

Compared with conventional analysis of the traditional measures of total lesion and grey matter volume a radiologist can extract, AI-assisted modelling of the complex imaging fingerprints was able to discriminate between pre- and post-treatment changes with much higher fidelity.

Researchers said in future the approach could guide therapy in individual patients, detect treatment success or failure faster, and could be used to conduct trials of new drugs more effectively and with smaller patient cohorts.

Neurologist Dr Parashkev Nachev, who led the study, said, "Rather than attempting to copy what radiologists do perfectly well already, complex computational modelling in neurology is best deployed on tasks human experts cannot do at all: to synthesise a rich multiplicity of clinical and imaging features into a coherent, quantified description of the individual patient as a whole. This allows us to combine the flexibility and finesse of a clinician with the rigour and objectivity of a machine.”

Professor Olga Ciccarelli, NIHR Research Professor, who is the senior author of the study, said, "The method is currently focused on imaging changes only; we are extending the approach to predicting the clinical response to disease modifying treatment, in terms of cognitive and motor outcomes. I hope that this exciting field of research will lead to an individual prediction of treatment response in multiple sclerosis using AI.”

This work was funded by the NIHR UCLH Biomedical Research Centre, and the Wellcome Trust. It falls within a broader programme of research also supported by the NIHR, the MS Society, and Novartis. 

The work is one of many ongoing collaborations with the School of Biomedical Engineering and Imaging Sciences at King’s College London, where Professor Seb Ourselin and his team, formerly at UCL, are helping to extend the scale and power of AI-assisted studies conducted in collaboration with London-based hospitals.

The research programme is aligned with UCLH's Research Hospital Initiative, which seeks to embed advanced modelling techniques into real world clinical practice.