Medical Image Analysis Using Machine Learning and Deep Learning Techniques

EasyChair Preprint 15183

Abstract

In the medical area, computer-aided detection through machine learning (ML) and deep learning (DL) is rapidly expanding. Medical pictures are thought to be the real source of the relevant data needed for disease diagnosis. One of the most crucial things to reduce the death rate from cancer and tumours is early disease detection utilising a variety of modalities. Radiologists and medical professionals can better understand the internal anatomy of a discovered disease by using modalities to extract the necessary features. Large data sets limit ML's ability to use current modalities, however DL is capable of handling any volume of data with ease. As a result, DL is seen as an improved method of ML, in which ML makes use of learning strategies and DL gathers information about how machines need to behave in human environments. To obtain more details about the datasets that are used, DL makes use of a multi-layered neural network. The purpose of this study is to provide a comprehensive assessment of the literature on the use of ML and DL for the identification and categorization of various diseases. An extensive analysis was previously conducted, and it offers a summary of several methods based on ML and DL for the identification and categorization of various illnesses, medical imaging modalities, instruments and procedures for assessment, and dataset descriptions. By enabling medical professionals and researchers to select an optimal diagnosis technique for a certain condition with high accuracy and minimal time, this work will benefit the healthcare community.

Keyphrases: Healthcare, deep learning, machine learning, medical image processing

@booklet{EasyChair:15183,
  author    = {Aaliya Nashat and Sudipta Hazra and Lucky Adhikary and Siddhartha Chatterjee and Senjuti Banerjee},
  title     = {Medical Image Analysis Using Machine Learning and Deep Learning Techniques},
  howpublished = {EasyChair Preprint 15183},
  year      = {EasyChair, 2024}}