Abstract: Artificial neural networks (ANN) have evolved to process data from various Computer-aided detection systems in the medical field. ANNs automate the diagnosis and reporting models of the radiological images, which aids medical practitioners in effectively managing various abnormalities. This paper presents various techniques used in radiology to interpret the images generated by CAD systems. We surveyed current advancements in this field and analyzed the advantages and disadvantages of such systems, which we think would result in a more practical and challenging approach to detecting critical information from the available data. Specialized areas of ANNs such as convolutional neural networks, Group Method of Data Handling (GMDH) neural networks, Feed-Forward Neural Networks, and Feed-Forward Back Propagation Neural Networks are considered, and their efficiency in this application is analyzed. This paper provides a critical discussion of the role of ANNs in the medical imaging field as well as future directions and research recommendations.
Keywords: Deep Learning, Artificial Neural Networks, Convolutional Neural Networks, Pulmonary Abnormalities
|
DOI:
10.17148/IJARCCE.2023.12722
