Deep Learning for Medical Image Segmentation

Deep Learning for Medical Image Segmentation

Kanchan Sarkar, Bohang Li
Copyright: © 2021 |Pages: 38
DOI: 10.4018/978-1-7998-5071-7.ch002
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Pixel accurate 2-D, 3-D medical image segmentation to identify abnormalities for further analysis is on high demand for computer-aided medical imaging applications. Various segmentation algorithms have been studied and applied in medical imaging for many years, but the problem remains challenging due to growing a large number of variety of applications starting from lung disease diagnosis based on x-ray images, nucleus detection, and segmentation based on microscopic pictures to kidney tumour segmentation. The recent innovation in deep learning brought revolutionary advances in computer vision. Image segmentation is one such area where deep learning shows its capacity and improves the performance by a larger margin than its successor. This chapter overviews the most popular deep learning-based image segmentation techniques and discusses their capabilities and basic advantages and limitations in the domain of medical imaging.
Chapter Preview
Top

Image Segmentation

Image segmentation is the process in computer vision to assign a label to each pixel in an image. Image segmentation splits the input image into different connected components based on some user-defined high-level semantics. Image segmentation and object detection have numerous applications in medical imaging. Here are a few use cases where image segmentation shows its capabilities in medical imaging which is not limited to lung disease diagnosis based on X-Ray images, nucleus detection and segmentation based on microscopic pictures, kidney tumour segmentation and many more. This section explains the different early traditional methods for image segmentation.

Thresholding

Thresholding is an early approach of segmentation where the image is segmented based on a threshold for each pixel intensity value. If the pixel intensity is greater than a pre-defined threshold it is considered as foreground pixel else belongs to background pixel. This method sometimes called ROI (Region of Interest) segmentation. Figure 1 is a typical grey level histogram of an image. At the pixel value “a” there is a deep valley on the histogram. The image can be segmented in background and foreground object based on the threshold value “a”. Thresholding is an early and novice approach of segmentation that can not be applied to sophisticated segmentation i.e instance segmentation, semantic segmentation and, images with multiple objects.

Figure 1.

ROI Segmentation

978-1-7998-5071-7.ch002.f01

Complete Chapter List

Search this Book:
Reset