Deep Learning Driven Image Segmentation in Medical Science - An Intense Learning

Authors

  • Dr.V.P.Gladis Pushparathi, Dr.S.Thanga ramya, Dr.P.Rangarajan

Abstract

Segmentation of medical images is a required mechanism for assessment, preparation, treatment, etc. to achieve and localize malignancy or clear structural anomalies in medical images. For a body part or lesion area, a wide variety of automated medical image segmentation techniques, such as vessel and skin cancer, have been proposed. Deep learning can learn motive in computer vision to anticipate groups of objects that form an image for robust segmentation. The primary in-depth learning architecture used in image processing is a Convolutionary Neural Network (CNN) or special CNN systems such as AlexNet, VGG, Inception and ResNet. Deep computer vision models are typically taught and conducted to shorten the time required to handle graphics units (GPUs).A huge effort is made in this paper to discuss various image segmentation approaches and explains how images are segmented by deep learning power.

Keywords-Artificial intelligence, Clinical practice, Healthcare, Machine learning, Medical science.

Published

2021-02-03

Issue

Section

Articles