The biggest advantage of Artificial Intelligence (AI) is its ability to analyse large amounts of data and categorize it in a way that is not immediately obvious to people. For example, now AI identified three new subtypes of multiple sclerosis (MS) in a recent UCL research. Scientists believe this will help categorize patients more accurately.
AI analysed different MRI scans and identified three previously unknown subtypes of MS. Image credit: James Heilman, MD via Wikimedia (CC BY-SA 4.0)
At the moment MS patients are categorized into progressive and relapsing groups. In other words, these categories are not based on the symptoms and are very broad. MS is a disabling condition and it can have many faces. It affects several millions of people in the world and not everyone is experiencing the same symptoms.
Researchers used an AI tool called SuStaIn to look through MRI brain scans of 6,322 MS patients. SuStaIn worked unsupervised and in some time managed to identify three previously unknown patterns in those brain scans. Scientists named them as three subtypes of MS according to the earliest abnormalities seen on the MRI scans – cortex-led, normal-appearing white matter-led and lesion-led. Scientists confirmed these findings in a separate round of SuStaIn research when it analysed a separate independent cohort of 3,068 patients. AI system was able to categorize them according to the newly defined subtypes of MS.
Alan Thompson, one of the lead authors of the study, said: “Now with the help of AI and large datasets, we have made the first step towards a better understanding of the underlying disease mechanisms which may inform our current clinical classification. This is a fantastic achievement and has the potential to be a real game-changer, informing both disease evolution and selection of patients for clinical trials”.
This is a huge achievement in MS research, which could have implications on the treatment. In fact, scientists believe that these subtypes could help predict MS disability progression and response to treatment. New treatments could be developed more accurately according to these subtypes. Finally, these findings could aid future MS research, as scientists are trying to understand what drives progression in MS. AI could be used as a diagnostics tool in that case, requiring only the same MRI scans for accurate understanding or what would be the best way to help every particular patient.
Accurate and defined classification of MS subtypes will lead to more accurate treatments. MS is a neurological condition, one of the most common causes of disability in young people. There are treatment options, but they do not work very well in every case. It helps if the disease is diagnosed early, so watch for the early signs – tingling, numbness, a loss of balance and problems with vision.
Source: UCL