Technology ID

OASIS: Automated brain lesion detection using cross-sectional multimodal magnetic resonance imaging

Lead Inventor
Reich, Daniel (National Institute of Neurological Disorders and Stroke)
Crainiceanu, Ciprian (Johns Hopkins University)
Sweeney, Elizabeth (Johns Hopkins University)
Shinohara, Russell (Johns Hopkins University)
Goldsmith, Arthur (Columbia University)
Shiee, Navid (Henry M Jackson Foundation)
Pham, Dzung (Henry M Jackson Foundation)
Software / Apps
Research Materials
Non-Medical Devices
Medical Devices
Therapeutic Areas
Lead IC
This invention is a novel statistical method for automatically detecting lesions in cross-sectional brain magnetic resonance imaging (MRI) studies. OASIS uses multimodal MRI from one image acquisition session and produces voxel-level probability maps of the brain that quantifies the likelihood that each voxel is part of a lesion. Binary lesion segmentations are created from these probability maps using a validated population-level threshold. In this application, fluid attenuated inversion recovery (FLAIR), proton density (PD), T2-weighted, and Tl-weighted volumes were used. The OASIS lesion segmentations are robust to changes in imaging centers and scanning parameters.
Commercial Applications
Traditionally trained observers do lesion identification and quantification manually on each MRI image. But this manual quantification is slow and prone to human error. There are many available methods for automated lesion segmentation. These methods are often tuned to one dataset and do not generalize to new datasets with different scanning parameters. The OASIS statistical method provides a fully automated, sensitive, and specific method for lesion segmentation that easily generalizes to new datasets.

Competitive Advantages
  • Use of a robust intensity normalization procedure that allows it to be employed in large population samples from different imaging centers
  • The built-in spatial normalization procedure permits the method to be more successful in segmenting certain areas of the brain
  • Training data from one imaging center was used and validated on data from a separate center to illustrate the robustness of the procedure to different imaging centers and MRI scanning parameters
  • Actual lesion detection is relatively fast
Licensing Contact: