Data from OM1's PremiOM™ Multiple Sclerosis Dataset was fed into RWD and machine learning methods to derive an Estimated EDSS (eEDSS).
FREMONT, CA: "It is exciting to see a machine learning model that performs at this level based on real-world data," states Dr. Carl Marci, Chief Psychiatrist and Managing Director of Mental Health & Neuroscience at OM1 and one of the authors of the paper. Sage Publications has published OM1's validation of a machine learning approach to predict expanded disability status scale scores for multiple sclerosis in the Multiple Sclerosis Journal. OM1 is a leader in real-world data, results, and technology for chronic diseases.
"Importantly, it can be easily applied to a neurologist's clinical note - saving time for clinicians and adding valuable tracking information for patients," adds Dr. Carl Marci.
Multiple sclerosis (MS) is an autoimmune, inflammatory illness that assaults the central nervous system (CNS) and causes various neurological symptoms, such as muscle weakness, stiffness, fatigue, numbness, vision issues, disorientation, bowel and bladder dysfunction, depression, and others. MS affects about 1 million Americans. In the absence of a cure, a delay in diagnosis and treatment may lead to irreversible disability. Existing drugs try to alter the progression of the disease, treat relapses, and manage symptoms.
Once diagnosed, the Expanded Disability Status Score (EDSS) is used to evaluate and quantify MS patients' disability levels. Scores can assist in identifying the course of a disease, the level of care required, and treatment recommendations. The EDSS evaluates functional impairment in seven systems, including the pyramidal, cerebellar, brainstem, sensory, bladder, bowel, visual, and cognitive systems. Due to the time necessary for physicians to complete the scale and the intricacy of scoring, its usage in ordinary clinical practice is limited, despite its extensive use in clinical studies.
To address the absence of functional disability tests, estimated EDSS (eEDSS) scores were generated using RWD and machine learning techniques. For this study, OM1 extrapolated data from the OM1 PremiOM MS Dataset to amplify and enlarge existing EDSS ratings as determined by clinicians. Nearly 14,000 MS patients with clinical notes were screened for an EDSS score extracted from the notes using medical language processing (MLP) techniques, and a test set of 684 patients with 3,489 scores was further divided into a model training cohort (75 percent) and a model validation cohort (25 percent) (25 percent ).
The PremiOM MS Dataset is a continuously updated database of over 19,400 MS patients prospectively followed with extensive clinical, laboratory, and other data, including longitudinal outcomes and Expanded Disability Status Scale (EDSS) scores, relapses, subtypes, and therapy response. In addition, researchers get access to the data of an additional 485,000 MS patients via the OM1 Real-World Data Cloud™ for modeling, analytics, and other research purposes.