Early Stage Detection of Multiple Sclerosis using FCNN
Pallavi Abhale, Amol Lashkare, Amol Deshpande · 2022 10th International Conference on Emerging Trends in Engineering and Technology - Signal and Information Processing (ICETET-SIP-22) · 2022
Multiple sclerosis (MS) is a disorder in which the immune system damages the protective covering of the nerves. In clinical trials for medicines and determining illness prognosis, quantitative examination of these lesions has proven to be quite beneficial. The accuracy with which the MS lesions are recognised and segmented in brain MRI determines the efficacy of these quantitative studies. This is mainly done by radiologists who use commonly accessible segmentation tools to label 3D MR images slice by slice. Manual methods, on the other hand, take a long time and are prone to mistakes. Several automatic segmentation strategies have been studied in recent years to sidestep this challenge. A new framework for autonomous brain lesion segmentation is detailed in the proposed method, which is built on a novel fully convolutional neural network (FCNN) architecture. The results suggest that integrating an FCNN increased the network's effectiveness in segmenting MS lesions.