Application of deep learning in detecting neurological disorders from magnetic resonance images: a survey on the detecti
- PDF / 1,875,161 Bytes
- 21 Pages / 595.276 x 790.866 pts Page_size
- 93 Downloads / 172 Views
Brain Informatics
Open Access
REVIEW
Application of deep learning in detecting neurological disorders from magnetic resonance images: a survey on the detection of Alzheimer’s disease, Parkinson’s disease and schizophrenia Manan Binth Taj Noor1, Nusrat Zerin Zenia1, M Shamim Kaiser1* , Shamim Al Mamun1 and Mufti Mahmud2*
Abstract Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders—focusing on Alzheimer’s disease, Parkinson’s disease and schizophrenia—from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided. Keywords: Machine learning, Alzheimer’s disease, Parkinson’s disease, Schizophrenia, Neuroimaging 1 Introduction Alzheimer’s disease (AD), Parkinson’s disease (PD) and schizophrenia (SZ) are three most common neurological disorders (NLD) which are characterized by the disruption of regular operations of brain functions [1–3]. A patient with either of these three NLD puts a heavy burden on the family as well as the health system. It is therefore imperative to detect these disorders at the earliest stage possible so that their progression can be slowed *Correspondence: [email protected]; [email protected]; [email protected] 1 Institute of Information Technology, Jahangirnagar University, Savar, 1342 Dhaka, Bangladesh 2 Department of Computing & Technology, Nottingham Trent University, NG11 8NS Nottingham, UK
down, if not fully stopped [4, 5]. Towards this aim, a number of different neuroimaging techniques (such as magnetic resonance imaging (MRI), computed tomography (CT) and positron emission tomography (PET)) and deep learning (DL)-based analysis methods have been developed to classify these disorders for early detection [3, 6–8], and to devise appropriate treatment strategies [9–11]. Over the last decade machine learning (ML) has been successfully applied to biological data mining [12, 13], image analysis [14], financial forecasting [15], anomaly detection [16, 17], disease detection [18, 19], natural language processing [20, 21] and strategic game playing [22]. In particular, the success of DL algorithms in computer vision
Data Loading...
