|International Journal of Computer Applications
|Foundation of Computer Science (FCS), NY, USA
|Volume 101 - Number 5
|Year of Publication: 2014
|Authors: S. P. Washimkar, S. D. Chede
S. P. Washimkar, S. D. Chede . AM FM based Prediction of Multiple Sclerosis in Brain MRI Images. International Journal of Computer Applications. 101, 5 ( September 2014), 13-17. DOI=10.5120/17682-8533
White matter is one of the two components of central nervous system and consists mostly of glialcells and myelinated axons that transmits signal from one region of the cerebrum to another and between cerebrum and lower brain centers. Multiple sclerosis (MS) is one of the most common diseases which affect white matter. Multiple sclerosis is a chronic idiopathic disease resulted in multiple areas of inflammatory demyelization in the Central nervous system. MS lesion formation often leads to unpredictable cognitive decline & Physical disability. Due to the sensitivity in detecting MS lesions, MRI has become an important tool for diagnosing MS & monitoring its progression. Radiological criteria for MS include the number of lesions (some scattered bright spot) on the MRI, their location and their size. Due to the complexity & variance of automated MRI segmentation of brain MS became a complex task. A structural texture analysis method on MS segmentation scheme gives emphasis on structural analysis of MS as well as on normal tissues. An important tool that has been developed and used in variety of research is the Image Modulation model, also termed the Amplitude-Modulation, Frequency-Modulation (AM-FM) image model, which models non-stationary image content using an AM-FM expansion. The AM-FM technique offers advantages for feature extraction at different frequency scales and orientations that can be used to detect different patterns, directions, or structures in an image. Thus high-frequency scale instantaneous amplitude can be used to differentiate between lesions associated with early and advanced disease stages and thus AM-FM technique can offer excellent results in classification of Multiple sclerosis from the white matter of the nervous system.