Dr Arman Eshaghi on the power of AI to revolutionise multiple sclerosis care

As a teenager, Dr Arman Eshaghi became interested in computer programming. What started as a fun after-school activity quickly became a passion. Little did he know then that this hobby would one day inspire him to use technology with the potential to revolutionise the treatment and care of people with multiple sclerosis (MS).

“In the mid-2000s, I was a studying medicine, and I realised my coding skills could help analyse medical images—especially MRI scans, which are central to diagnosing multiple sclerosis,” Dr Eshaghi recalls, “Back then, using computers in healthcare was unusual, but as medical practice became more digital, I combined my medical knowledge with tech expertise.”

By 2015, Dr Eshaghi was spearheading work at UCL, focusing on MS imagining and AI.

“The chance to use technology to improve patient care and understand complex brain diseases like MS continues to inspire my work. MS is a highly complex disease. AI has the potential to help scientists ask fundamental questions about the disease and get answers. For example, about how the disease progresses and how treatments may be able to stop MS one day,” he says.

Multiple sclerosis is a condition that affects the brain and spinal cord. Although it affects every individual differently, many patients experience problems with their movement, vision and thinking.

Dr Eshaghi explains: “MS affects 3 million people worldwide and often strikes young adults in their 30s, disrupting careers, family life, and daily activities. There’s no cure yet, but treatments have grown from none to over 20 in about two decades. This transformation and complexity fascinate me. Understanding MS and helping shape better therapies to improve people’s lives keeps me engaged in this field.”

In 2022, Dr Eshaghi was awarded an Advanced Fellowship from the National Institute for Health and Care Research (NIHR) for research that develops digital tools to personalise treatments for MS.

“With so many MS treatments now available, it’s hard to choose the best one for each individual,” Dr Eshaghi says.

Due to the unpredictable nature of MS, one in three people with the condition in the UK need to change their treatment within two years because their initial treatment does not provide sufficient benefits or experience adverse events.

As a solution to this problem, Dr Eshaghi is working on a suit of AI tools called MS-PINPOINT to predict how symptoms may change, when MS may worsen, and which drugs are likely to help.

“My MS-PINPOINT project uses real-world hospital data and advanced AI to group patients by similar characteristics, based on their routine scans and records,” Dr Eshaghi explains, “By doing this, we can predict which treatments might work best for each person, making care more personalised and effective. This approach aims to reduce guesswork, shorten the time to find the right therapy, and improve everyone’s quality of life.”

Midway through the five-year research project, Dr Eshaghi has found that everyday hospital data, while convenient to use, is not always ideal for research. To overcome this, the team created an AI model called MindGlide that works with lower quality MRI images to measure brain changes, a key factor in MS.

“This innovation helps us unlock valuable insights from regular healthcare data without needing expensive, high-quality research images. It’s a step forward in using existing hospital records for better MS understanding and care,” Dr Eshaghi says.

Working with colleagues from seven NHS Trusts and two international clinical centres, it is hoped that MS-PINPOINT will enable healthcare teams to measure and predict whether a treatment works at an individual level. AI will also help refine these predictions further so that they are even more personalised.

To protect patient privacy all identifying details—names, birthdays, addresses—are removed from the records and faces are removed from MRI images.

Dr Eshaghi says: “We never share raw data between hospitals or with external researchers. Instead, we use ‘federated learning’, where AI is trained locally, and only the learned patterns are shared. Strict NHS guidelines, cybersecurity, audits, and transparency measures ensure privacy remains central. This careful approach lets us benefit from AI while safeguarding personal information.”

So how does Dr Eshaghi see AI being used in the patient pathway in future?

“I imagine AI as a support tool that assists doctors, rather than replacing them. It could quickly analyse MRI scans, help predict how MS might progress and recommend suitable treatments. It can also help generate reports or answer common patient questions. By handling repetitive tasks, AI frees clinicians to focus on patient care. Ultimately, integrating AI into clinics could improve efficiency, help patients understand their condition, and enhance overall MS management,” Dr Eshaghi says.

As part of the research, the team will also calculate how these AI tools may save costs in future healthcare by recommending appropriate treatments earlier and delaying disability. This has the potential to be life changing for people living with MS, improving their quality of life and reducing their pain.

Dr Eshaghi explains: “If AI tools guide patients and doctors toward the right treatments sooner, patients might experience fewer relapses and delayed disability. This leads to better quality of life and can reduce costs by avoiding trial-and-error prescriptions. In turn, savings could be reinvested into the NHS or brought back to the public purse.

“Over time, improved understanding of MS and better therapies also help everyone, boosting both health outcomes and the wealth of the nation. These benefits can be quantified using health economics tools, which will be a mainstay of my research in the next few years.”

At the halfway mark of this incredible study, it is wonderful to see such promising preliminary results. Looking back on his journey, what would Dr Eshaghi say to his 13-year-old self about his computer hobby?

“I would ask my 13-year-old self to be more humble and give more credit to my parents, who once told me that the world would drastically change in a few decades, and those who learned to code computers will change the world.”

Biography

Dr Arman Eshaghi is a National Insitute for Health and Care Research (NIHR) Advanced Fellow at UCL, developing advanced artificial intelligence models using brain imaging and electronic health records. He obtained his medical doctorate degree in 2013 from Tehran University of Medical Sciences and was awarded a PhD in Neuroscience from UCL in 2018. He is a member of the Progression of Neurodegenerative Disorders (POND) Team at the Centre for Medical Image Computing at the Department of Computer Science at UCL, working closely with Professor Daniel Alexander.