Pacal Ishak, Karaboga Dervis, Basturk Alper, Akay Bahriye, Nalbantoglu Ufuk
Computer Engineering Department, Engineering Faculty, Igdir University, Igdir, Turkey.
Computer Engineering Department, Engineering Faculty, Erciyes University, Kayseri, Turkey; Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah, Saudi Arabia.
Comput Biol Med. 2020 Nov;126:104003. doi: 10.1016/j.compbiomed.2020.104003. Epub 2020 Sep 17.
Deep learning has emerged as a leading machine learning tool in object detection and has attracted attention with its achievements in progressing medical image analysis. Convolutional Neural Networks (CNNs) are the most preferred method of deep learning algorithms for this purpose and they have an essential role in the detection and potential early diagnosis of colon cancer. In this article, we hope to bring a perspective to progress in this area by reviewing deep learning practices for colon cancer analysis. This study first presents an overview of popular deep learning architectures used in colon cancer analysis. After that, all studies related to colon cancer analysis are collected under the field of colon cancer and deep learning, then they are divided into five categories that are detection, classification, segmentation, survival prediction, and inflammatory bowel diseases. Then, the studies collected under each category are summarized in detail and listed. We conclude our work with a summary of recent deep learning practices for colon cancer analysis, a critical discussion of the challenges faced, and suggestions for future research. This study differs from other studies by including 135 recent academic papers, separating colon cancer into five different classes, and providing a comprehensive structure. We hope that this study is beneficial to researchers interested in using deep learning techniques for the diagnosis of colon cancer.
深度学习已成为目标检测领域领先的机器学习工具，并因其在推进医学图像分析方面的成就而备受关注。卷积神经网络（CNN）是用于此目的的深度学习算法中最受欢迎的方法，它们在结肠癌的检测和潜在早期诊断中起着至关重要的作用。在本文中，我们希望通过回顾用于结肠癌分析的深度学习实践，为该领域的进展带来一个视角。本研究首先概述了用于结肠癌分析的流行深度学习架构。之后，将所有与结肠癌分析相关的研究收集到结肠癌与深度学习领域下，然后将它们分为检测、分类、分割、生存预测和炎症性肠病五类。然后，对每类下收集的研究进行详细总结并列出。我们以对近期结肠癌分析深度学习实践的总结、对所面临挑战的批判性讨论以及对未来研究的建议来结束我们的工作。本研究与其他研究的不同之处在于纳入了135篇近期学术论文，将结肠癌分为五个不同类别，并提供了一个全面的结构。我们希望这项研究对有兴趣使用深度学习技术进行结肠癌诊断的研究人员有所帮助。
