About Multiple Sclerosis
MS is an unpredictable, autoimmune disease that affects the central nervous system. 2.8 million people living with MS worldwide (atlas of MS). This is often characterized by relapsing episodes of neurologic impairment followed by remissions. Symptoms could be different for each patient, therefore personalized medicine is crucial for patient-specific treatment planning.
The diagnosis of MS is based upon a clinical assessment; despite significant effort, no single biomarker has been found to independently confirm the diagnosis. In an attempt to assure the highest sensitivity and specificity, a set of guidelines, referred to as the McDonald criteria (Digre et al. 2002), utilizes MRI to provide supportive data to facilitate the diagnosis of MS.
The utilization of MRI has simplified the diagnosis and the decision-making as to when DMT should be initiated. MRI is now the gold standard for MS diagnosis (Rovira et al. 2015, Charil et al.2006, Wattjes et al. 2021) and a fundamental part of the clinical routine.
How AI can empower MRI-based clinical routine workflows?
Although MRI also has shortcomings, AI can address these shortcomings to improve current MRI-based clinical routine workflows.
The shortcoming in MRI that AI can address:
- High variability in the quality of images and interpretation
- A significant amount of manual intervention
- Limitation of standardized radiological reporting and interpretation
- Subtle progression of the disease and difficulty in disease progression detection even for experienced HCPs.
ECTRIMS 2022:
- ePoster 1008: “Machine Learning Parcellation of Multiple Sclerosis Lesions into Texturally Consistent Super-Voxels for Lesion Classification” Spinat Q, Caba B Gafson A, Ioannidou D, Jiang X, Cafaro A, Bradley DP, Perea R, Fisher E, Arnold DL, Elliott C, Paragios N, Belachew S
- Poster 227: “A Novel Deep Learning Algorithm for Multi-Modal Multiple Sclerosis Lesion Segmentation” Santos Garcia C, Caba B, Gafson A, Ioannidou D, Jiang X, Cafaro A, Bradley DP, Perea R, Fisher E, Arnold DL, Elliott C, Paragios N, Belachew S
- Unsupervised Clustering of Acute Multiple Sclerosis Lesions across Spatial, Geometric and Textural Domains, Caba, A. Gafson, D. Ioannidou, X. Jiang, A. Cafaro, D.P. Bradley, R. Perea, E. Fisher, D.L. Arnold, C. Elliott, N. Paragios, S. Belachew https://www.charcot-ms.org/files/Annual-Meetings/30/Abstract/20_ECF2022_Abstract_BS_Caba_B.pdf
AAN 2023:
- Deep Learning Reconstruction of Lesion-Free Representations of the Brain in Multiple Sclerosis Patients (P11-3.006)
https://www.aan.com/MSA/Public/Events/AbstractDetails/53315