A Study on Deep Learning in Neurodegenerative Diseases and Other Brain Disorders
Deep Learning (DL) is an emerging field that attracts researchers, especially in the field of engineering and medical sciences. DL gives us many solutions to date, that is why it is still an active field of interest and will for many years. In this paper,
- PDF / 230,037 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 75 Downloads / 157 Views
Abstract Deep Learning (DL) is an emerging field that attracts researchers, especially in the field of engineering and medical sciences. DL gives us many solutions to date, that is why it is still an active field of interest and will for many years. In this paper, we provide a short introduction of deep learning architectures and the applications and the role of deep learning in the prediction of neurodegenerative diseases like Alzheimer’s, Parkinson’s, Huntington’s disease, mild cognitive impairment, and other dementia. We also discuss other brain disorders and how deep learning is essential, nowadays, in the medical sector for providing better, accurate, and fast treatment to the subject. Keywords Deep learning · Alzheimer · Parkinson’s disease · Neurodegenerative disorder
1 Introduction In recent years, Machine Learning (ML) is the most popular research area and in machine learning, Deep Learning (DL) includes exciting trends in prediction and healthcare sectors. Deep learning is used in image classification, image segmentation, objection detection, image registration, image de-noising, and others. Nowadays, deep learning plays a vital role in Computer-Aided Diagnosis (CAD) for the detection of diseases in early stage and more accurate diagnosis. We have different deep learning architectures which are used in health sectors. – Deep Neural Network (DNN)—It is mostly used in drug design, RNA binding protein, DNA methylation, gene variants, tumor detection, air pollutant prediction, and hemorrhage detection.
M. Jyotiyana (B) · N. Kesswani Central University of Rajasthan, Ajmer, Rajasthan, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 V. S. Rathore et al. (eds.), Rising Threats in Expert Applications and Solutions, Advances in Intelligent Systems and Computing 1187, https://doi.org/10.1007/978-981-15-6014-9_95
791
792
M. Jyotiyana and N. Kesswani
– Convolutional Neural Network (CNN)—This architecture is frequently accessed in neural cell classification, organ segmentation, human activity recognition, human behavior monitoring, and infectious disease epidemics. – Deep Autoencoder (DA)—It is commonly used in cancer diagnosis, 3D brain reconstruction, cell clustering, prediction of diseases, and predicting demographic information. – Recurrent Neural Networks (RNN)—RNN architecture is used mostly in data mining, signal processing, character recognition, speech recognition, and NLPrelated tasks. – Deep Belief Networks (DBN)—This type of architecture is used mainly in gene classification or gene selection, compound–protein interaction, brain tissue classification, anomaly detection, prediction of diseases, and diagnosis of lifestyle diseases. There are some more deep learning architectures [1, 2] such as Deep Conventional Extreme Learning Machine (DC-ELM) and Deep Boltzmann Machine (DBM) which are used in other research areas. Deep learning has achieved a central position in recent years in ML and Pattern Recognition (PR). In this paper, we have outlined the various types of deep
Data Loading...
