摘要
卷积神经网络(Convolutional Neural Network,CNN)在计算机图像分析中应用广泛,取得了很多突破性进展。将CNN应用于医学影像处理任务,不仅可以将医护人员从繁杂的手工劳动中解放,而且可以在很大程度上代替专业医生对某些常见病症做出高准确率和高效率的诊断分析。文章旨在总结基于卷积神经网络的医学影像分析方法和应用,先概括CNN的由来、发展和主要机制,再介绍CNN在医学图像分割、重建以及生成等方面的应用,并列举相关领域的国内外研究成果,重点阐述基于卷积神经网络的技术应用于医学影像分析的整体优越性和高效性。
Convolutional Neural Network(CNN) is widely used in computer image analysis and has made numerous breakthrough achievements. Applying CNN to medical image processing tasks can not only liberate medical staffs from a lot of complicated manual labor,but also replace professional doctors, to a certain extent, to make the efficient disease diagnosis with high accuracy and high efficiency.This paper aims to summarize the methods and applications of medical image analysis based on convolutional neural network. It first sums up the origin, development and main mechanism of CNN, and then introduces the applications of CNN in medical image segmentation,reconstruction and generation, and lists the main achievements of domestic and foreign researches in related fields. The overall superiority and high efficiency of CNN-based technologies in medical image analysis are emphasized.
作者
朱春燕
姚健
钱鹏江
ZHU Chunyan;YAO Jian;QIAN Pengjiang(School of Information Technology,Suzhou Top Institute of Information Technology,Suzhou Jiangsu 215311,China;School of Artificial Intelligence and Computer Science,Jiangnan University,Wuxi Jiangsu 214122,China)
出处
《信息与电脑》
2022年第14期56-60,共5页
Information & Computer
基金
江苏省高职院校教师专业带头人高端研修项目(项目编号:2021GRFX059)。
关键词
卷积神经网络(CNN)
深度学习
医学影像
疾病诊断
智能医疗
Convolutional Neural Network(CNN)
deep learning
medical images
disease diagnosis
intelligent medicine
