Automate analysis of MRI brain scan images using multi-modal features derived from brain magnetic resonance imaging (MRI) scans.
For more information, contact us
Today, many mental diseases do not have a well-established, non-invasive diagnosis Bio-marker. As the symptoms overlap with other mental illnesses (like bipolar disorder), the diagnosis is subjective by process of elimination. The multimodal information and select features available from MRI brain scans may help enhance diagnosis of a few anomalies to help improve diagnosis.
The data set consists of two sets of information collected by different imaging modalities: Functional Network Connectivity (FNC) and Source-Based Morphometry (SBM) loadings. The FNC are derived from functional magnetic resonance imaging (fMRI) scans, and can be seen as a functional modality feature describing the subject’s overall level of ‘synchronicity’ between brain areas. SBM loadings are derived from structural MRI scans, and they indicate the concentration of grey matter in different regions of the subject’s brain. Using both FNC and SBM loadings, data predictive models can be developed with machine learning to enhance the diagnosis of a few anomalies.