An Analysis Of Deep Learning In CXR Medical Image Processing

Abstract

When it comes to the process of extracting information from chest X-ray data, Deep Learning models have shown exceptional
performance thanks to more powerful processing capabilities and improved training techniques. These models also provide a number
of other benefits. As one of the most common imaging techniques, its expanding popularity is reflected in the increasing amount of
work that radiologists are required to do. Since of this, computer-aided diagnostic tools would be beneficial in the healthcare business
because they would enable medical professionals to prioritize certain tests and more accurately detect certain ailments. To the best of
the author's knowledge, there is no publication in the existing body of research that particularly reviews relevant work on anomaly
detection and multi-label thoracic pathology categorization. For the purpose of facilitating comparison, the most effective chest X-raybased deep learning algorithms have been chosen for this study. Recent developments in deep learning technology have led to
improvements in the performance of a variety of professions in the field of medical image analysis. Because chest radiographs are the
kind of radiological examination that is performed the most often and because they have the most varied range of applications that
have been investigated, this modality is considered to be of utmost significance. In recent years, as a consequence of the public release
of multiple big chest X-ray datasets, there has been a rise in the level of interest shown by academics. Explanations of every dataset
that is available to the public are provided in extensive detail, and descriptions of currently available commercial solutions are also
provided.
Within the realms of medical image analysis and computer-aided radiology diagnosis, a contentious issue that is being discussed at
length is the classification of X-ray pictures taken of the chest. The primary goal of this project is to improve the efficiency as well as
the quality of the work performed by radiologists by designing and putting into action an automated approach for recognizing and
categorizing different diseases." This study focuses on chest X-ray image classification approaches that make use of machine learning
methods. The purpose of this work is to improve existing surveys, and it does so by using strategies for chest X-ray image classification
that are based on machine learning methods." At the start, a basic comprehension of chest radiography and medical image processing
is presented, coupled with a short introduction to data mining.