Pub Date : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864532
Guijie Li, Qiang Wang, Xin Zhao
Microbubble ultrasound contrast agent which is highly echo-genic has many unique properties. Microbubbles can improve the sensitivity of the conventional ultrasound imaging; tissue permeability can be moderately increased by high-frequency ultrasonic cavitation, and hardly any damage is caused even at the high acoustic pressure. Microbubbles can carry drugs and release them under ultrasound mediated microbubble destruction. Meanwhile, Microbubbles increased vascular permeability while enhancing drugs deposition in tissue. Scientists have focused on the research of targeted microbubble contrast agent, e.g. carrying genetic drugs to the target tissue, mediating tumor cell apoptosis and blocking tumor cells metastasis. This article reviews the microbubble-specific imaging, for its principles and applications; ultrasound-aided drug delivery, and targeted imaging are also reviewed. All the applications though promising require further improvements for future clinical use.
{"title":"Introduction of ultrasound microbubble technology aided tumor imaging and treatment","authors":"Guijie Li, Qiang Wang, Xin Zhao","doi":"10.1109/ICMIPE.2013.6864532","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864532","url":null,"abstract":"Microbubble ultrasound contrast agent which is highly echo-genic has many unique properties. Microbubbles can improve the sensitivity of the conventional ultrasound imaging; tissue permeability can be moderately increased by high-frequency ultrasonic cavitation, and hardly any damage is caused even at the high acoustic pressure. Microbubbles can carry drugs and release them under ultrasound mediated microbubble destruction. Meanwhile, Microbubbles increased vascular permeability while enhancing drugs deposition in tissue. Scientists have focused on the research of targeted microbubble contrast agent, e.g. carrying genetic drugs to the target tissue, mediating tumor cell apoptosis and blocking tumor cells metastasis. This article reviews the microbubble-specific imaging, for its principles and applications; ultrasound-aided drug delivery, and targeted imaging are also reviewed. All the applications though promising require further improvements for future clinical use.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123159654","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864492
Yuchao Tang, Yong Cai, Xiang Wang, Jigen Peng
Total variation (TV) minimization problems are widely used for solving incomplete data problems in computed tomography (CT) image reconstruction. The present paper investigates a primal dual proximal point method of Chambolle-Pock algorithm to solve the CT image reconstruction problem which consisting the sum of l2 data fidelity term and TV regularization term. We tested these methods on computer simulated data, and they exhibited good performance when used to few-view and limited-angle CT image reconstruction.
{"title":"A primal dual proximal point method of Chambolle-Pock algorithm for total variation image reconstruction","authors":"Yuchao Tang, Yong Cai, Xiang Wang, Jigen Peng","doi":"10.1109/ICMIPE.2013.6864492","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864492","url":null,"abstract":"Total variation (TV) minimization problems are widely used for solving incomplete data problems in computed tomography (CT) image reconstruction. The present paper investigates a primal dual proximal point method of Chambolle-Pock algorithm to solve the CT image reconstruction problem which consisting the sum of l2 data fidelity term and TV regularization term. We tested these methods on computer simulated data, and they exhibited good performance when used to few-view and limited-angle CT image reconstruction.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123629161","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864502
Li Yu, Xu Hui-jun, Zhang Su-jing
With spine auxiliary positioning during IGRT treatment, utilize the simulated human phantom and dynamic lung phantom to detect the accuracy of fiducial tracking, and evaluate the value of it. Utilize CT to scan the phantoms which contain films, and design plans for phantoms and spine auxiliary positioning which use 1, 2, 3, 4, 5 and 6 fiducials for tracking, respectively. Each phantom plan shall be repeatedly implemented 3 times. The E2E software is used to analyze the accuracy of irradiation and then the average value is acquired. The results of detecting the simulated human phantom: the accuracy of irradiation with 1 and 2 fiducials for tracking (combining with spine auxiliary positioning) was 1.11 mm and 1.05 mm; with 3, 4, 5 and 6 fiducials, the accuracy was 1.07 mm, 0.92 mm, 0.97 mm and 1.15 mm, respectively. The results of detecting dynamic lung phantom: with 1 and 2 fiducials used for tracking (combining with spine auxiliary positioning), the accuracy of irradiation was 1.03 mm and 0.70 mm; with 3, 4, 5 and 6 fiducials, the accuracy was 0.58 mm, 1.02 mm, 0.65 mm and 0.96 mm, respectively. For the relatively static tumor or moving tumor, if the rotation direction of tumor is consistent with spine, the accuracy of using 1 or 2 fiducials for tracking was the same as using 3 to 6 fiducials. When the number of fiducials is not enough, fiducial tracking along with spine auxiliary positioning can well solve this problem.
{"title":"Detection and evaluation on the accuracy of fiducial tracking with spine auxiliary positioning during IGRT treatment","authors":"Li Yu, Xu Hui-jun, Zhang Su-jing","doi":"10.1109/ICMIPE.2013.6864502","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864502","url":null,"abstract":"With spine auxiliary positioning during IGRT treatment, utilize the simulated human phantom and dynamic lung phantom to detect the accuracy of fiducial tracking, and evaluate the value of it. Utilize CT to scan the phantoms which contain films, and design plans for phantoms and spine auxiliary positioning which use 1, 2, 3, 4, 5 and 6 fiducials for tracking, respectively. Each phantom plan shall be repeatedly implemented 3 times. The E2E software is used to analyze the accuracy of irradiation and then the average value is acquired. The results of detecting the simulated human phantom: the accuracy of irradiation with 1 and 2 fiducials for tracking (combining with spine auxiliary positioning) was 1.11 mm and 1.05 mm; with 3, 4, 5 and 6 fiducials, the accuracy was 1.07 mm, 0.92 mm, 0.97 mm and 1.15 mm, respectively. The results of detecting dynamic lung phantom: with 1 and 2 fiducials used for tracking (combining with spine auxiliary positioning), the accuracy of irradiation was 1.03 mm and 0.70 mm; with 3, 4, 5 and 6 fiducials, the accuracy was 0.58 mm, 1.02 mm, 0.65 mm and 0.96 mm, respectively. For the relatively static tumor or moving tumor, if the rotation direction of tumor is consistent with spine, the accuracy of using 1 or 2 fiducials for tracking was the same as using 3 to 6 fiducials. When the number of fiducials is not enough, fiducial tracking along with spine auxiliary positioning can well solve this problem.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126428862","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864521
Hongzhe Yang, Lihui Zhao, Songyuan Tang, Yongtian Wang
The segmentation of brain tumor using Magnetic Resonance Image (MRI) plays an important role in the medical image process. This paper presents a comprehensive survey on brain tumor methods and technology using MRI images. Generally, brain tumor segmentation methods can divided into two main categories, spatial continuous and spatial discrete methods. Several methods, techniques, related advantage and weakness will be described and discussed. The evaluation measures are mentioned and the qualities of different method focus on the methods that were applied on the standard data sets. The efficient and stably brain tumor segmentation is still a challenging task for the unpredictable appearance and shape of the brain tumor.
{"title":"Survey on brain tumor segmentation methods","authors":"Hongzhe Yang, Lihui Zhao, Songyuan Tang, Yongtian Wang","doi":"10.1109/ICMIPE.2013.6864521","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864521","url":null,"abstract":"The segmentation of brain tumor using Magnetic Resonance Image (MRI) plays an important role in the medical image process. This paper presents a comprehensive survey on brain tumor methods and technology using MRI images. Generally, brain tumor segmentation methods can divided into two main categories, spatial continuous and spatial discrete methods. Several methods, techniques, related advantage and weakness will be described and discussed. The evaluation measures are mentioned and the qualities of different method focus on the methods that were applied on the standard data sets. The efficient and stably brain tumor segmentation is still a challenging task for the unpredictable appearance and shape of the brain tumor.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126932932","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864545
C. Zhao, Zhipeng Liu, T. Yin, Yi-Mei Chen
Object: For realizing the Finite Element (FE) analysis of real human head electromagnetic distribution induced by Trans-cranial Magnetic Stimulation (TMS), a complex human head FE model containing seven organizational structures, i.e. scalp, skull, cerebrospinal fluid (CSF), grey matter, white matter, cerebellum and eyeballs, was established, with the distribution of conductivities, and hence the corresponding tetrahedral FE model. Method: 1) Structural volume models establishment, from which seven organizational structures were distinguished. 2) Conductivity distribution definition, to generate the physical simulation model. and 3) meshed FE model generation. Result: A real human head FE model reflecting features of seven craniocerebral organs and distribution of conductivities. Conclusion: On the basis of five-shell craniocerebral structures, cerebellum and eyeballs are taken into account, providing more materials for the study of magnetic stimulating effects on human neural systems.
{"title":"Establishment and simulation of real human head conductivity Finite Element model","authors":"C. Zhao, Zhipeng Liu, T. Yin, Yi-Mei Chen","doi":"10.1109/ICMIPE.2013.6864545","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864545","url":null,"abstract":"Object: For realizing the Finite Element (FE) analysis of real human head electromagnetic distribution induced by Trans-cranial Magnetic Stimulation (TMS), a complex human head FE model containing seven organizational structures, i.e. scalp, skull, cerebrospinal fluid (CSF), grey matter, white matter, cerebellum and eyeballs, was established, with the distribution of conductivities, and hence the corresponding tetrahedral FE model. Method: 1) Structural volume models establishment, from which seven organizational structures were distinguished. 2) Conductivity distribution definition, to generate the physical simulation model. and 3) meshed FE model generation. Result: A real human head FE model reflecting features of seven craniocerebral organs and distribution of conductivities. Conclusion: On the basis of five-shell craniocerebral structures, cerebellum and eyeballs are taken into account, providing more materials for the study of magnetic stimulating effects on human neural systems.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127221079","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864507
Li Tong, Ying Zeng, S. Bao, Bin Yan, Linyuan Wang
Despite recent progress in X-ray computed tomography (CT), metal artifact remains to be a major problem in CT image processing and significantly limits numerous important applications. To overcome this problem, a CT image segmentation algorithm using an image multi-scale decomposition model based on the total variation/L1 (TV/L1) model is proposed in this paper. Different metal areas are extracted using the multi-scale decomposition properties of the TV/L1 model, which allows the separation of metal areas from artifacts areas. Segmentation experiments are performed on real CT images of a printed circuit board with a distinct metal artifact. Results show that the proposed method is as effective as the interactive segmentation method, and it does not require interactions.
{"title":"Multi-scale decomposition model-based segmentation algorithm for CT image with metal artifacts","authors":"Li Tong, Ying Zeng, S. Bao, Bin Yan, Linyuan Wang","doi":"10.1109/ICMIPE.2013.6864507","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864507","url":null,"abstract":"Despite recent progress in X-ray computed tomography (CT), metal artifact remains to be a major problem in CT image processing and significantly limits numerous important applications. To overcome this problem, a CT image segmentation algorithm using an image multi-scale decomposition model based on the total variation/L1 (TV/L1) model is proposed in this paper. Different metal areas are extracted using the multi-scale decomposition properties of the TV/L1 model, which allows the separation of metal areas from artifacts areas. Segmentation experiments are performed on real CT images of a printed circuit board with a distinct metal artifact. Results show that the proposed method is as effective as the interactive segmentation method, and it does not require interactions.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124872495","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864509
Feng Lijun, Yan Yangfei, He Junmei, Yi Shiyuan, Zhou Mengdi, Zeng Bixin
In order to emphasize the significance of CT simulation and to help students to understand CT working principle, especially the graphic reconstruction principle, a reflection type ultrasonic CT experimental device based on the Stm32 is designed for the teaching of CT. The device consists of a transceiver ultrasonic module, a MCU module, a serial communication module, a 3D mechanical platform, a stepper motor drive module and a Matlab graphical reconstruction module. The ultrasonic signals are launched by the Stm32 MCU controlling transceiver ultrasonic module, the movements of the object under test and the rotation and translation of ultrasonic transducer are controlled by the VB interface. The signals reflected by the object under test will be collected and processed by the Stm32 MCU, and then transferred to a computer through a serial port communication. A 3-D contour graph reconstruction of the object under test is implemented by the Matlab software and is displayed in the VB interface imaging window. There are many advantages of this device, such as small volume, simple structure, widely-used components, easy maintenance and repair, and thus, broad market prospects.
{"title":"Research on reflective ultrasonic CT experimental device","authors":"Feng Lijun, Yan Yangfei, He Junmei, Yi Shiyuan, Zhou Mengdi, Zeng Bixin","doi":"10.1109/ICMIPE.2013.6864509","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864509","url":null,"abstract":"In order to emphasize the significance of CT simulation and to help students to understand CT working principle, especially the graphic reconstruction principle, a reflection type ultrasonic CT experimental device based on the Stm32 is designed for the teaching of CT. The device consists of a transceiver ultrasonic module, a MCU module, a serial communication module, a 3D mechanical platform, a stepper motor drive module and a Matlab graphical reconstruction module. The ultrasonic signals are launched by the Stm32 MCU controlling transceiver ultrasonic module, the movements of the object under test and the rotation and translation of ultrasonic transducer are controlled by the VB interface. The signals reflected by the object under test will be collected and processed by the Stm32 MCU, and then transferred to a computer through a serial port communication. A 3-D contour graph reconstruction of the object under test is implemented by the Matlab software and is displayed in the VB interface imaging window. There are many advantages of this device, such as small volume, simple structure, widely-used components, easy maintenance and repair, and thus, broad market prospects.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131064850","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864525
Yaonan Zhang, Sai Li, Xian Li, Hongliang Li, Hairong Zheng
Since the optical flow method can't estimate the large displacement, the two-dimensional compression expansion method is proposed in this article to compensate for the large-scale movements in the image before the optical flow estimation. As a result, the related effects of decorrelation caused by the lateral displacement of the longitudinal compression can be effectively eliminated. Experimental results show that the two-dimensional compression extension methods can enhance the accuracy and robustness of the optical flow estimation. The resulting axial displacement and axial strain are basically consistent with the finite element simulation results, which proves the correctness of the new method.
{"title":"A method of ultrasonic strain estimation under large tissue compression","authors":"Yaonan Zhang, Sai Li, Xian Li, Hongliang Li, Hairong Zheng","doi":"10.1109/ICMIPE.2013.6864525","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864525","url":null,"abstract":"Since the optical flow method can't estimate the large displacement, the two-dimensional compression expansion method is proposed in this article to compensate for the large-scale movements in the image before the optical flow estimation. As a result, the related effects of decorrelation caused by the lateral displacement of the longitudinal compression can be effectively eliminated. Experimental results show that the two-dimensional compression extension methods can enhance the accuracy and robustness of the optical flow estimation. The resulting axial displacement and axial strain are basically consistent with the finite element simulation results, which proves the correctness of the new method.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129333102","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864551
Chen Jian, Yan Bin, Jiang Hua, Zengrong Lei, Tong Li
In medical image processing, interactive image segmentation is an important part, because it can obtain accurate segment results with less human effort compared with manual scribing. We proposed an improved algorithm of maximal similarity based region merging, compared with the algorithm proposed in [3], our algorithm use SLIC superpixels segmentation to obtain presegmented regions, using SLIC superpixles, it is easy to control the number of presegmentation regions. We also introduce the texture features differences while rigion merging, so we can obtain the accuracy of similarity measurement. Experimental results show that our algorithm can obtain comparable results.
{"title":"Interactive image segmentation by improved maximal similarity based region merging","authors":"Chen Jian, Yan Bin, Jiang Hua, Zengrong Lei, Tong Li","doi":"10.1109/ICMIPE.2013.6864551","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864551","url":null,"abstract":"In medical image processing, interactive image segmentation is an important part, because it can obtain accurate segment results with less human effort compared with manual scribing. We proposed an improved algorithm of maximal similarity based region merging, compared with the algorithm proposed in [3], our algorithm use SLIC superpixels segmentation to obtain presegmented regions, using SLIC superpixles, it is easy to control the number of presegmentation regions. We also introduce the texture features differences while rigion merging, so we can obtain the accuracy of similarity measurement. Experimental results show that our algorithm can obtain comparable results.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134119222","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 : 2013-10-01DOI: 10.1109/ICMIPE.2013.6864540
Wen Lu, Ruhai Dou, Guangyu Zhang
In mammography computer-aided diagnosis, the automatic extraction of interesting region is one of the most difficult problems. This paper presents a method based on two-dimensional principal component analysis (2DPCA) to extract the region of interest (ROI) automatically. First, preprocess the mammograms, then, extract mammography features by 2DPCA method and edge-detection algorithm. Finally, extract ROI by neural network classifier. 60 cases were analyzed and 100 images which from Shandong medical imaging research institute were used in this investigation. The results show that a better positive detection ratio is obtained with this method. This approach can obtain better extraction accuracy by integrating 2DPCA, edge-detection algorithm and neural networks.
{"title":"A new method for extracting region of interest in mammograms","authors":"Wen Lu, Ruhai Dou, Guangyu Zhang","doi":"10.1109/ICMIPE.2013.6864540","DOIUrl":"https://doi.org/10.1109/ICMIPE.2013.6864540","url":null,"abstract":"In mammography computer-aided diagnosis, the automatic extraction of interesting region is one of the most difficult problems. This paper presents a method based on two-dimensional principal component analysis (2DPCA) to extract the region of interest (ROI) automatically. First, preprocess the mammograms, then, extract mammography features by 2DPCA method and edge-detection algorithm. Finally, extract ROI by neural network classifier. 60 cases were analyzed and 100 images which from Shandong medical imaging research institute were used in this investigation. The results show that a better positive detection ratio is obtained with this method. This approach can obtain better extraction accuracy by integrating 2DPCA, edge-detection algorithm and neural networks.","PeriodicalId":135461,"journal":{"name":"2013 IEEE International Conference on Medical Imaging Physics and Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128780923","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：