Clinically Deployed Computational Assessment of Multiple Sclerosis Lesions.

Area of Science:

• Neurology
• Medical Imaging
• Computational Medicine

Background:

• Multiple Sclerosis (MS) is a central nervous system demyelinating disease affecting nearly 1 million US adults.
• Magnetic Resonance Imaging (MRI), particularly T2-FLAIR brain scans, is crucial for MS diagnosis and monitoring.
• White matter lesions on MRI are key indicators for disease activity and treatment decisions.

Purpose of the Study:

• To present a computational approach for detecting new MS disease activity.
• To determine the necessity of Gadolinium Based Contrast Agent (GBCA) injections in MS patients.
• To integrate this computer-aided detection (CAD) software into routine clinical workflows.

Main Methods:

• A computer-aided detection (CAD) software was developed and implemented for clinical use.
• The software analyzes brain MRI T2-FLAIR images to identify white matter lesions.
• It performs two primary tasks: detecting new disease activity and identifying patients needing GBCA injection.

Main Results:

• The CAD software has been used for over 10 years on >19,000 patients for disease activity detection.
• For GBCA injection necessity, the software achieved >85% sensitivity over 3 years.
• The approach provides reproducible and accurate clinical assessment of MS lesions.

Conclusions:

• The CAD software offers a reliable method for monitoring MS progression and treatment response.
• It helps reduce GBCA adverse effects and deposition by identifying patients who truly benefit from injection.
• This technology decreases healthcare costs, patient discomfort, and caregiver burden.