Deep Learning Models for Semantic Multi-Modal Medical Image Segmentation

Deep Learning Models for Semantic Multi-Modal Medical Image Segmentation

V. R. S. Mani (National Engineering College, Kovilpatti, India)
DOI: 10.4018/978-1-7998-3591-2.ch002
OnDemand PDF Download:
No Current Special Offers


In this chapter, the author paints a comprehensive picture of different deep learning models used in different multi-modal image segmentation tasks. This chapter is an introduction for those new to the field, an overview for those working in the field, and a reference for those searching for literature on a specific application. Methods are classified according to the different types of multi-modal images and the corresponding types of convolution neural networks used in the segmentation task. The chapter starts with an introduction to CNN topology and describes various models like Hyper Dense Net, Organ Attention Net, UNet, VNet, Dilated Fully Convolutional Network, Transfer Learning, etc.
Chapter Preview

Multi Modal Brain Mri Image Segmentation:

Brain MR image segmentation is a key function in radio therapy and image guided surgery. Quantitative analysis of brain abnormality requires an accurate segmentation of MR images of brain which is a complex and hard challenge. The heterogeneous nature of the lesions which include its large scale variability in phrases of length, shape and area make segmentation assignment extremely difficult. Manual segmentation the use of a human expert is the high-quality approach that is time eating, tedious, high priced and impractical in larger research and introduces inter observer variability. A couple of image sequences with varying contrast want to be taken into consideration for identifying whether a selected location is a lesion or not. And additionally the extent of expert knowledge and enjoy impacts the segmentation accuracy. In mind imaging studies special modality pix are mixed to enhance the short comings of character imaging techniques.T1 weighted photos produce correct evaluation among gray remember (GM) and white matter (WM) tissues. T2 weighted and Proton Density (PD) photographs enables in visualizing the lesions and different abnormalities Fluid Attenuated Inversion recovery (aptitude) photographs improve the photograph assessment of white remember lesions on account of a couple of sclerosis [2]. To enhance the accuracy in brain photo segmentation, fusing distinct modality photograph is important. Fusing multi modality photograph is important within the case of toddler brains which has poor assessment.

Complete Chapter List

Search this Book: