A Comprehensive Review on the Application of 3D Convolutional Neural Networks in Medical Imaging

Abstract

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Reservoir Computing Based on Oxygen-Vacancy-Mediated X-ray Optical Synaptic Device for Medical CT Bone Diagnosis

“…4 Particularly, in the medical field, large amounts of data are generated, including high-resolution medical imaging such as computed tomography (CT), and the limits of human analysis of such data have been exceeded, so neuromorphic computing is urgently needed. 5 CT, a method of scanning and diagnosing human organs using accurately collimated X-ray beams, has the advantages of rapid diagnosis, simplicity and safety, noninvasiveness, etc. It has high application value in the diagnosis of various organ diseases, especially in the diagnosis of tumors.…”

“…6−9 At present, the convolutional neural network (CNN) has been widely used in CT visual neuromorphic computing systems to analysis and process CT images in medical treatment. 5 However, due to the complex network structure, CNN also has the shortcomings of large energy consumption and a large calculation amount. Reservoir computing (RC), as a lightweight deep learning model, consists of only two parts: a dynamic reservoir layer and readout layer, so it is easy for hardware implementation.…”

“…By simulating the function of the neural network in the human brain, the neuromorphic computing architecture has been put forward, which has the advantages of integrating storage and computation with low power consumption and has been developed vigorously recently. − Neuromorphic computing has a wide range of applications in the fields of visual and auditory recognition . Particularly, in the medical field, large amounts of data are generated, including high-resolution medical imaging such as computed tomography (CT), and the limits of human analysis of such data have been exceeded, so neuromorphic computing is urgently needed . CT, a method of scanning and diagnosing human organs using accurately collimated X-ray beams, has the advantages of rapid diagnosis, simplicity and safety, noninvasiveness, etc.…”

“…At present, the convolutional neural network (CNN) has been widely used in CT visual neuromorphic computing systems to analysis and process CT images in medical treatment . However, due to the complex network structure, CNN also has the shortcomings of large energy consumption and a large calculation amount.…”