C11947: Automated System for Analysis of MRI Data for Neurological AbnormalitiesNovelty:
OASIS (Automated Statistical Inference for Segmentation) is a statistically principled, fast, and accurate tool for the analysis of multi-sequence MRI data that provides a segmentation identifying how much lesion load a subject has, and where in the brain these lesions exist, from a single cross-sectional imaging study. This analysis is effective in determining etiologies and clinical management of many neurological diseases, including but not limited to, multiple sclerosis (MS).
Value Proposition:
While the use of MRI is standard in the detection of brain lesions associated with Multiple Sclerosis, manual assessment of these images is cumbersome. There are automated methods aimed at addressing this issue, but they often require a large amount of "teaching" data, or images of the same patient at different times. This technology provides a robust means of automatically detecting lesions at the voxel level. Other advantages include:
• Utilizes multiple modalities to improve accuracy.
• Works with a small set of “teaching” data.
• Can be applied to other diseases.
Technical Details:
Johns Hopkins researchers have developed a method to detect Multiple Sclerosis using advanced image processing and mathematical modeling techniques. By providing analysis on the voxel-level, a quantitative analysis can be done on the extent or progression of the disease over time. While originally developed for MS, this technique can easily be applied to other imaging methods or to detect other diseases.
Looking for Partners:
To develop and commercialize the technology as a new tool in the Computer Aided Detection Market.
Stage of Development:
Prototype
Data Availability:
Performance analysis of images for 98 patients
Publications/Associated Cases:
Not at this time
