AI Model Analyzes Patient Data to Diagnose Multiple Sclerosis With 90% Accuracy

By MedImaging International staff writers
Posted on 01 May 2025

Image: The new AI model improves MS diagnostics (Photo courtesy of 123RF)

Multiple sclerosis (MS) is a chronic inflammatory condition affecting the central nervous system. Most patients initially experience the relapsing-remitting form (RRMS), characterized by periods of symptom flare-ups followed by stability. Over time, many individuals transition to secondary progressive MS (SPMS), where symptoms gradually worsen without noticeable breaks. Identifying this transition is critical, as the two forms of MS require different treatment approaches. Currently, the diagnosis is typically made an average of three years after the transition begins, which can result in patients receiving treatments that are no longer effective. Now, a new artificial intelligence (AI) model can predict with 90 percent certainty which form of MS a patient has. This model enhances the likelihood of starting the correct treatment promptly, helping to slow disease progression.

The AI model, developed by researchers at Uppsala University (Uppsala, Sweden), synthesizes clinical data from over 22,000 patients in the Swedish MS Registry. The model is based on data routinely collected during regular healthcare visits, including neurological tests, magnetic resonance imaging (MRI) scans, and ongoing treatments. In a study published in Digital Medicine, the model was able to identify the transition to secondary progressive MS correctly or earlier than recorded in the patient's medical history in nearly 87 percent of cases, achieving an overall accuracy of around 90 percent. For patients, this means an earlier diagnosis, allowing for timely adjustments in treatment to slow the disease's progression. This also reduces the likelihood of patients receiving medications that are no longer effective. In the future, the model could be used to identify appropriate candidates for clinical trials, potentially leading to more effective and personalized treatment strategies.

“By recognizing patterns from previous patients, the model can determine whether a patient has the relapsing-remitting form or whether the disease has transitioned to secondary progressive MS,” said Kim Kultima, researcher at the Department of Medical Sciences, Uppsala University, who led the study. “What is unique about the model is that it also indicates how confident it is in each individual assessment. This means that the doctor will know how reliable the conclusion is and how confident the AI is in its assessment.”

Related Links:
Uppsala University