Enhanced study on Deep learning model for kidney segmentation using DCE-MRI

• Published in: 2023 International Conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering (RMKMATE) pp. 1 - 5
• Main Authors: S, Nagarajan, M, Ramprasath
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2023
• Subjects: Automatic Detection algorithms; Computer-Assisted diagnostic system; Deep learning; Detection algorithms; Image recognition; Image segmentation; Kidney tumor segmentation; Nonhomogeneous media; Planning; Solid modeling; Vision applications
• DOI: 10.1109/RMKMATE59243.2023.10368681

Medical imaging is capturing pictures of bodily components for diagnostic or research reasons. Because of advancements in image-handling techniques, which include picture recognition, examination, and upgrading, clinical imaging is expanding swiftly. Picture division is a vital subject in picture handling and PC vision with applications, for example, scene grasping, clinical picture examination, mechanical discernment, video observation, expanded reality, and picture pressure, among numerous others. Numerous picture segmentation methods have been created and are available in the literature. There has been a lot of work done to develop techniques for segmenting pictures using deep-learning models since they have proved effective in various vision applications. Deep learning has turned into a functioning exploration point in the field of clinical picture examination. Compared to the laborious and lengthy conventional technique, deep learning (DL) automatic detection algorithms can speed up diagnostics, increase test accuracy, lower expenses, and lessen the workload on the radiologist. Kidney segmentation is an integral part of any non-intrusive computer-assisted diagnostic (CAD) system for rapidly recognising severe renal repudiation because of the intensity of inhomogeneity brought on by errors in the image capture process, efficient kidney segmentation is challenging in evaluating and therapeutic planning of kidney-related disorders. This article addresses recent developments in DL-based kidney tumor segmentation systems. We highlight their fundamental components and numerous approaches as we discuss the main medical image kinds, segmentation methods, and evaluation standards for segmentation outcomes in kidney tumor++ segmentation.