基于卷积神经网络的图像单特征点识别

IF 0.3 4区 医学 Q4 ENGINEERING, BIOMEDICAL Journal of Hard Tissue Biology Pub Date : 2021-01-01 DOI:10.2485/JHTB.30.161
Miki Hori, Makoto Jincho, Tadasuke Hori, Hironao Sekine, A. Kato, Atsuko Ueno, T. Kawai
{"title":"基于卷积神经网络的图像单特征点识别","authors":"Miki Hori, Makoto Jincho, Tadasuke Hori, Hironao Sekine, A. Kato, Atsuko Ueno, T. Kawai","doi":"10.2485/JHTB.30.161","DOIUrl":null,"url":null,"abstract":": Most studies of artificial intelligence in the medical field involve classification problems, but few consider recog nition of one characteristic point in images or regression analysis such as data recognition. In this research, we constructed a fundamental convolutional neural network framework for regression analysis. Images of the handwritten digit “3” from the MNIST dataset were used as training data, with the protruding middle point as an image feature point. Input images and training data (x1, y1) were connected to 6 convolutional layers and then run through 2 affine layers to produce the output data (x2, y2). The loss function was the mean radial error (MRE) between the training and output data. After machine learn ing, the error converged to 0.75 pixels on average. We expect that this algorithm can be clinically applied to points having certain characteristics in images, such as locating hard tissue lesions or recognizing measurement points in cephalograms.","PeriodicalId":16040,"journal":{"name":"Journal of Hard Tissue Biology","volume":null,"pages":null},"PeriodicalIF":0.3000,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Recognition of Image One Feature Point Using Convolutional Neural Networks\",\"authors\":\"Miki Hori, Makoto Jincho, Tadasuke Hori, Hironao Sekine, A. Kato, Atsuko Ueno, T. Kawai\",\"doi\":\"10.2485/JHTB.30.161\",\"DOIUrl\":null,\"url\":null,\"abstract\":\": Most studies of artificial intelligence in the medical field involve classification problems, but few consider recog nition of one characteristic point in images or regression analysis such as data recognition. In this research, we constructed a fundamental convolutional neural network framework for regression analysis. Images of the handwritten digit “3” from the MNIST dataset were used as training data, with the protruding middle point as an image feature point. Input images and training data (x1, y1) were connected to 6 convolutional layers and then run through 2 affine layers to produce the output data (x2, y2). The loss function was the mean radial error (MRE) between the training and output data. After machine learn ing, the error converged to 0.75 pixels on average. We expect that this algorithm can be clinically applied to points having certain characteristics in images, such as locating hard tissue lesions or recognizing measurement points in cephalograms.\",\"PeriodicalId\":16040,\"journal\":{\"name\":\"Journal of Hard Tissue Biology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.3000,\"publicationDate\":\"2021-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Hard Tissue Biology\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.2485/JHTB.30.161\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"ENGINEERING, BIOMEDICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Hard Tissue Biology","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.2485/JHTB.30.161","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

摘要

医学领域的人工智能研究大多涉及分类问题,但很少考虑图像中一个特征点的识别或数据识别等回归分析。在这项研究中,我们构建了一个基本的卷积神经网络框架进行回归分析。以MNIST数据集中的手写体数字“3”作为训练数据,突出的中间点作为图像特征点。输入图像和训练数据(x1, y1)连接到6个卷积层,再通过2个仿射层生成输出数据(x2, y2)。损失函数是训练数据和输出数据之间的平均径向误差(MRE)。经过机器学习,误差平均收敛到0.75像素。我们期望该算法可以在临床上应用于图像中具有一定特征的点,如定位硬组织病变或识别脑电图中的测量点。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Recognition of Image One Feature Point Using Convolutional Neural Networks
: Most studies of artificial intelligence in the medical field involve classification problems, but few consider recog nition of one characteristic point in images or regression analysis such as data recognition. In this research, we constructed a fundamental convolutional neural network framework for regression analysis. Images of the handwritten digit “3” from the MNIST dataset were used as training data, with the protruding middle point as an image feature point. Input images and training data (x1, y1) were connected to 6 convolutional layers and then run through 2 affine layers to produce the output data (x2, y2). The loss function was the mean radial error (MRE) between the training and output data. After machine learn ing, the error converged to 0.75 pixels on average. We expect that this algorithm can be clinically applied to points having certain characteristics in images, such as locating hard tissue lesions or recognizing measurement points in cephalograms.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Journal of Hard Tissue Biology
Journal of Hard Tissue Biology ENGINEERING, BIOMEDICAL-
CiteScore
0.90
自引率
0.00%
发文量
28
审稿时长
6-12 weeks
期刊介绍: Information not localized
期刊最新文献
Evaluation of Maxillary First Molar Intrusion Mechanics with Mini-Implant Anchorages Using the Finite Element Method Study on the Effect of Soft-Start Light on Microleakage in Pit and Fissure Closure Upregulation of miR-101-3p Overcomes Ibrutinib Resistance by Targeting ABCC5 in Diffuse Large B-Cell Lymphoma (DLBCL) miR-141 Improve Osteoporosis by Promoting Osteoblast Differentiation through Targeting RICTOR A Study of Submandibular Gland Changes in Mice of a Murine Model of Sjögren’s Syndrome Administered Dental Pulp Stem Cell-Conditioned Medium
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1