Pub Date : 2018-08-20DOI: 10.3969/J.ISSN.0258-8021.2018.04.004
Xiaohan Kong, T. Tan, L. Bao, Guangzhi Wang
The automatic classification of breast tumor in ultrasound images is of great significance to improve doctors' efficiency and reduce the rate of misdiagnosis. The novel 3D breast ultrasound data contains more information for diagnosis, but images from different directions have their distinct performance as a result of this ultrasound imaging mechanism. For this breast ultrasound data, this paper designed three kinds of convolutional neural network model using its flexibility and characteristic of learning automatically, and the three models were able to accept transverse plane images, transverse plane and coronal plane images, images and annotations information. The effects of different information fusion on the accuracy of breast tumor classification were investigated. A dataset contains 880 images (i. e., 401 benign images, 479 malign images) and their annotations were employed, and we performed 5-fold cross validation to calculate the accuracy and AUC of each model. The experimental results indicated that the models designed in this paper can deal with the images and annotations simultaneously. Compared with the single-input model, the multi-information fusion model improved the accuracy of classification by 2.91%, and achieved the accuracy of 75.11% and AUC of 0.8294. The proposed models provided a reference for the classification application of convolutional neural networks with multi-information fusion.
{"title":"Classification of breast mass in 3D ultrasound images with annotations based on convolutional neural networks","authors":"Xiaohan Kong, T. Tan, L. Bao, Guangzhi Wang","doi":"10.3969/J.ISSN.0258-8021.2018.04.004","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2018.04.004","url":null,"abstract":"The automatic classification of breast tumor in ultrasound images is of great significance to improve doctors' efficiency and reduce the rate of misdiagnosis. The novel 3D breast ultrasound data contains more information for diagnosis, but images from different directions have their distinct performance as a result of this ultrasound imaging mechanism. For this breast ultrasound data, this paper designed three kinds of convolutional neural network model using its flexibility and characteristic of learning automatically, and the three models were able to accept transverse plane images, transverse plane and coronal plane images, images and annotations information. The effects of different information fusion on the accuracy of breast tumor classification were investigated. A dataset contains 880 images (i. e., 401 benign images, 479 malign images) and their annotations were employed, and we performed 5-fold cross validation to calculate the accuracy and AUC of each model. The experimental results indicated that the models designed in this paper can deal with the images and annotations simultaneously. Compared with the single-input model, the multi-information fusion model improved the accuracy of classification by 2.91%, and achieved the accuracy of 75.11% and AUC of 0.8294. The proposed models provided a reference for the classification application of convolutional neural networks with multi-information fusion.","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"61 1","pages":"414-422"},"PeriodicalIF":0.0,"publicationDate":"2018-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73556422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-01-01DOI: 10.1007/978-81-322-1695-7_64
Junyu Long, Hong Yu, Aiming Yu
{"title":"Research on Medical Image Fusion Algorithms Based on Nonsubsampled Contourlet","authors":"Junyu Long, Hong Yu, Aiming Yu","doi":"10.1007/978-81-322-1695-7_64","DOIUrl":"https://doi.org/10.1007/978-81-322-1695-7_64","url":null,"abstract":"","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"4 1","pages":"543-551"},"PeriodicalIF":0.0,"publicationDate":"2014-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75585192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-10-20DOI: 10.3969/J.ISSN.0258-8021.2013.05.07
Jianjun Meng, X. Sheng, Lin Yao, Xiangyang Zhu
{"title":"Common spatial spectral pattern for motor imagery tasks in small channel configuration","authors":"Jianjun Meng, X. Sheng, Lin Yao, Xiangyang Zhu","doi":"10.3969/J.ISSN.0258-8021.2013.05.07","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2013.05.07","url":null,"abstract":"","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"1 1","pages":"553-561"},"PeriodicalIF":0.0,"publicationDate":"2013-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88257007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2012-12-20DOI: 10.3969/J.ISSN.0258-8021.2012.06.005
Wei Wu, H. Duan, Yan Li, C. Jiang, Bing Qin, David Huang
{"title":"3D anterior chamber angle measurements with high resolution Fourier-domain optical coherence tomography","authors":"Wei Wu, H. Duan, Yan Li, C. Jiang, Bing Qin, David Huang","doi":"10.3969/J.ISSN.0258-8021.2012.06.005","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2012.06.005","url":null,"abstract":"","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"187 5-6","pages":"831-838"},"PeriodicalIF":0.0,"publicationDate":"2012-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72470107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2012-10-20DOI: 10.3969/J.ISSN.0258-8021.2012.05.006
L. Zhang, M. Jiang, An De Bao
{"title":"Advection-diffusion-reaction equations based tumor cells growth modeling","authors":"L. Zhang, M. Jiang, An De Bao","doi":"10.3969/J.ISSN.0258-8021.2012.05.006","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2012.05.006","url":null,"abstract":"","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"28 1","pages":"679-689"},"PeriodicalIF":0.0,"publicationDate":"2012-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83617183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2012-02-20DOI: 10.3969/J.ISSN.0258-8021.2012.01.008
Meng Ma, Yang Wang, Ying Ru, Zefeng Wang
The sequence classification methods have broad application in various bioinformatics areas such as the identification of regulatory elements of transcription and the prediction of protein structure.Here we presented a new classification method to analyze short sequences based on their sequential features,and used this method to study RNA splicing regulatory elements.This method extracted the sequential features from the known spicing regulatory elements,and developed a scoring system to evaluate how possible a given short sequence can regulate RNA splicing.This method was compared with some other methods through applying to a set of exonic splicing enhancer(ESE) and silencer(ESS) octamers.The average prediction accuracy of this sequential feature-based method for three kinds of computation validation experiments reached about 93% and the transparent predictive structure of the method helps to interpret the biological mechanism.This paper shows a new method for biology series' data analysis and presents a new way for the study of regulatory sequences that control gene expression.
{"title":"A Classification Method for RNA Splicing Regulatory Elements","authors":"Meng Ma, Yang Wang, Ying Ru, Zefeng Wang","doi":"10.3969/J.ISSN.0258-8021.2012.01.008","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2012.01.008","url":null,"abstract":"The sequence classification methods have broad application in various bioinformatics areas such as the identification of regulatory elements of transcription and the prediction of protein structure.Here we presented a new classification method to analyze short sequences based on their sequential features,and used this method to study RNA splicing regulatory elements.This method extracted the sequential features from the known spicing regulatory elements,and developed a scoring system to evaluate how possible a given short sequence can regulate RNA splicing.This method was compared with some other methods through applying to a set of exonic splicing enhancer(ESE) and silencer(ESS) octamers.The average prediction accuracy of this sequential feature-based method for three kinds of computation validation experiments reached about 93% and the transparent predictive structure of the method helps to interpret the biological mechanism.This paper shows a new method for biology series' data analysis and presents a new way for the study of regulatory sequences that control gene expression.","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"14 1","pages":"45-52"},"PeriodicalIF":0.0,"publicationDate":"2012-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89032271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2010-10-20DOI: 10.3969/J.ISSN.0258-8021.2010.05.023
Zhuo Weng, G. Xie, Xin Liu, Cheng Xiong, Hairong Zheng, B. Qiu
Magnetic resonance imaging(MRI) is one of the most important non-invasive diagnostic tools in routine clinical examination.However,the temporal resolution is still low due to the limitation of Nyquist sampling theorem in k-space signal acquisition.Under the conditions of certain magnetic and gradient field,it takes a long time for signal acquisition to obtain a high resolution image with clinical value.In addition to enhancing the strength of main magnetic field and gradient as well speeding gradient field switch,some mathematical methods have been used to reduce the amount of k-space signal acquisition to shorten MR imaging time.Although under sparse sampling,the final reconstructed image data could be satisfied with Nyquist sampling theorem through these mathematical methods.Furthermore,many fast MRI methods based on data sharing and undersampling of k-space were proposed,such as half-Fourier imaging,keole imaging,parallel imaging,partially separable functions(PSF) and so on.In this review,several typical fast imaging methods were summarized and discussed based on k-space sampling techniques.
{"title":"Development of fast magnetic resonance imaging techniques based on k-space accelerated collection","authors":"Zhuo Weng, G. Xie, Xin Liu, Cheng Xiong, Hairong Zheng, B. Qiu","doi":"10.3969/J.ISSN.0258-8021.2010.05.023","DOIUrl":"https://doi.org/10.3969/J.ISSN.0258-8021.2010.05.023","url":null,"abstract":"Magnetic resonance imaging(MRI) is one of the most important non-invasive diagnostic tools in routine clinical examination.However,the temporal resolution is still low due to the limitation of Nyquist sampling theorem in k-space signal acquisition.Under the conditions of certain magnetic and gradient field,it takes a long time for signal acquisition to obtain a high resolution image with clinical value.In addition to enhancing the strength of main magnetic field and gradient as well speeding gradient field switch,some mathematical methods have been used to reduce the amount of k-space signal acquisition to shorten MR imaging time.Although under sparse sampling,the final reconstructed image data could be satisfied with Nyquist sampling theorem through these mathematical methods.Furthermore,many fast MRI methods based on data sharing and undersampling of k-space were proposed,such as half-Fourier imaging,keole imaging,parallel imaging,partially separable functions(PSF) and so on.In this review,several typical fast imaging methods were summarized and discussed based on k-space sampling techniques.","PeriodicalId":35998,"journal":{"name":"中国生物医学工程学报","volume":"32 1","pages":"785-792"},"PeriodicalIF":0.0,"publicationDate":"2010-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76791883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：