Disease Classification using Improved Synchronous Neural Interaction
The improved synchronous neural interaction software algorithm does not eliminate potential predictors but utilizes the entire set of predictors in a computationally efficient manner to classify subjects into either a healthy or a disease classification. This technology has been used to successfully classify PTSD patients with 90% overall accuracy using a simple, 60 second fixation magnetoencephalography (MEG) scan. Similarly, it has been used to categorize the subclassification of multiple sclerosis (MS), which is critical from a treatment perspective. This technology is expected to find utility with other neurological diseases and may be applicable to other modalities, such as electroencephalography (EEG) and magnetic resonance imaging (MRI).
|MN-IP Try and Buy|
This technology is offered as a portfolio with z09185.
Uses More Predictors from MEG Scans
Many neurological and psychological diseases do not use biomarkers for objectively diagnosing the disease. Post-traumatic stress disorder (PTSD), for example, is currently diagnosed by a health professional through a review of the patient’s symptoms, typically including, history of potentially traumatic events, determination of whether the patient meets the DSM-IV criteria for PTSD, the frequency and severity of symptoms and associated disability, and whether there are comorbid psychiatric and medical conditions. The current procedure can be lengthy and is subjective. Emerging methods utilize synchronous neural interaction to analyze massive quantities of predictors, which enables classification of a subject into a disease group. However, with such a large predictor space, the data is unable to be analyzed efficiently and it is difficult to determine what data is most relevant, what is not, and how the information can be used to classify a subject into a specific disease classification. Current methods utilizing MEG scans, therefore, use only a relatively small percentage of the predictors for classification, ignoring a significant amount of the available data that could improve classification.
BENEFITS OF DISEASE CLASSIFICATION USING IMPROVED SYNCHRONOUS NEURAL INTERACTION:
- Potential to replace subjective diagnostic techniques with an objective functional biomarker
- Technology is more computationally efficient relative to current synchronous neural interaction techniques
- Ability to handle massive data sets with large numbers of predictors without a large memory requirement
- Better specificity in classifying and diagnosing neurological diseases