基于Zernike矩的结构MRI阿尔茨海默病分类方法

2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA) Pub Date : 2017-04-01 DOI:10.1109/PRIA.2017.7983061

Aref Shams-Baboli, M. Ezoji

{"title":"基于Zernike矩的结构MRI阿尔茨海默病分类方法","authors":"Aref Shams-Baboli, M. Ezoji","doi":"10.1109/PRIA.2017.7983061","DOIUrl":null,"url":null,"abstract":"This paper proposed a method based on Zernike moments to classify the various stages of Alzheimer's Disease(AD) from structural MRIs. The proposed method is benefited from all three orthogonal directions of MRIs i.e. Axial, Sagittal and Coronal images. Three back-propagation algorithms had been used to train the neural network with seven neurons in hidden layer to reach the best accuracy. We experimented this method with 232 MRIs from OASIS database. 70 percent of the subjects had been used for training and the other 30 percent was used to evaluate the trained network. We achieved accuracy of 86.46 percent in multiclass mode and 96.67 percent of accuracy in two class mode between HC and AD.","PeriodicalId":336066,"journal":{"name":"2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":"{\"title\":\"A Zernike moment based method for classification of Alzheimer's disease from structural MRI\",\"authors\":\"Aref Shams-Baboli, M. Ezoji\",\"doi\":\"10.1109/PRIA.2017.7983061\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper proposed a method based on Zernike moments to classify the various stages of Alzheimer's Disease(AD) from structural MRIs. The proposed method is benefited from all three orthogonal directions of MRIs i.e. Axial, Sagittal and Coronal images. Three back-propagation algorithms had been used to train the neural network with seven neurons in hidden layer to reach the best accuracy. We experimented this method with 232 MRIs from OASIS database. 70 percent of the subjects had been used for training and the other 30 percent was used to evaluate the trained network. We achieved accuracy of 86.46 percent in multiclass mode and 96.67 percent of accuracy in two class mode between HC and AD.\",\"PeriodicalId\":336066,\"journal\":{\"name\":\"2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA)\",\"volume\":\"85 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"9\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/PRIA.2017.7983061\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PRIA.2017.7983061","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 9

摘要

提出了一种基于Zernike矩的结构mri对阿尔茨海默病(AD)各阶段进行分类的方法。该方法利用了核磁共振成像的轴向、矢状面和冠状面三个正交方向图像。采用三种反向传播算法对隐层7个神经元的神经网络进行训练，以达到最佳精度。我们用OASIS数据库中的232张mri进行了实验。70%的受试者被用于训练，另外30%用于评估训练后的网络。在HC和AD之间的多类模式下，准确率为86.46%，两类模式下准确率为96.67%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A Zernike moment based method for classification of Alzheimer's disease from structural MRI

This paper proposed a method based on Zernike moments to classify the various stages of Alzheimer's Disease(AD) from structural MRIs. The proposed method is benefited from all three orthogonal directions of MRIs i.e. Axial, Sagittal and Coronal images. Three back-propagation algorithms had been used to train the neural network with seven neurons in hidden layer to reach the best accuracy. We experimented this method with 232 MRIs from OASIS database. 70 percent of the subjects had been used for training and the other 30 percent was used to evaluate the trained network. We achieved accuracy of 86.46 percent in multiclass mode and 96.67 percent of accuracy in two class mode between HC and AD.

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

2017 3rd International Conference on Pattern Recognition and Image Analysis (IPRIA)

自引率

0.00%

发文量