Deep learning in medical imaging: A comparative study of techniques

View at publisher Summarise with VIRAT
Authors: Shreesh S. Kolekar, Monali Gulhane, Pawar Atul Namdev, Nitin Rakesh, Tulsidas A. Patil, Akhil Gupta
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The way doctors identify and treat different health problems has changed a lot since deep learning methods were introduced to medical imaging. In this work, we compare and contrast a number of deep learning models that are used for medical imaging tasks like picture classification, segmentation, and detection. In particular, we look at how well generative adversarial networks (GANs), recurrent neural networks (RNNs), and convolutional neural networks (CNNs) work in medical picture analysis. CNNs are very good at finding patterns and groups in space, and they’ve done really well at jobs like finding tumors, separating organs, and classifying diseases. RNNs are mostly used for handling sequential data, but they are also being used more and more in medical imaging, especially to look at time-series data from moving imaging methods like MRI and ultrasound. GANs, on the other hand, are used to improve the quality of pictures and make fake medical images for training. We look at the pros and cons of these deep learning methods using various medical picture datasets, including CT scans, MRI scans, and X-ray photos, in this work. To make a full comparison, metrics like sensitivity, precision, accuracy, and processing speed are looked at. According to the data, CNN-based models are good at standard picture recognition tasks, while RNNs are better at dealing with time sequences. GANs have the potential to create high-quality fake data to add to training datasets. This can help models do better when there isn’t enough labeled data. This study compares different deep learning methods to see which ones work best for certain medical imaging tasks. It also suggests how these approaches could be used in the future in hospital settings.

Fields of Study & Subjects

Artificial Intelligence in Healthcare and Education Advanced Neural Network Applications Medical Image Segmentation Techniques

Publication details

Journal / source
AIP conference proceedings
Volume
3401
Pages
020112–020112
Publication date
1 January 2026
Language
EN