3D tracker based 3D reconstruction of cardiac ultrasound images

碩士 === 國立臺灣大學 === 資訊工程研究所 === 84 === This thesis proposes new techniques for visualization and 3D reconstruction of cardiac ultrasound images. Ultrasonic imaging techniques have been broadly used in clinical diagnosis, because of lower cost and less harm,...

Full description

Bibliographic Details
Main Authors: Chen,Jiann-Hwa, 陳建華
Other Authors: Ouhyoung,Ming
Format: Others
Language:zh-TW
Published: 1996
Online Access:http://ndltd.ncl.edu.tw/handle/47617760655828923517
id ndltd-TW-084NTU00392045
record_format oai_dc
spelling ndltd-TW-084NTU003920452016-07-13T04:10:50Z http://ndltd.ncl.edu.tw/handle/47617760655828923517 3D tracker based 3D reconstruction of cardiac ultrasound images 以空間定位器做為心臟超音波影像定位之三維立體重建 Chen,Jiann-Hwa 陳建華 碩士 國立臺灣大學 資訊工程研究所 84 This thesis proposes new techniques for visualization and 3D reconstruction of cardiac ultrasound images. Ultrasonic imaging techniques have been broadly used in clinical diagnosis, because of lower cost and less harm, as compared to x-ray based imaging. However, we have encountered two problems in ultrasound imaging which are images of poor signal to noise ratio and the positioning problem in a sequence of unaligned images due to manual control. To get precise ultrasound images of cutting planes of heart, we attached a 3D tracker which can report the position and orientation of the transducer used in 2D echocardiography. Then, we can determine the position and orientation of the transducer with the help of the 3D tracker. In order to improve the quality of ultrasound images, several image processing methods have been applied. We used three kinds of image processing techniques to enhance the quality of image, such as background subtraction, image segmentation, and image filter. Besides, human heart is a moving organ with a semi- constant period. The movement of human heart in its 3D position is not only caused by heart beat, but also due to a little displacement following inhaling and exhaling. Therefore, when we try to reconstruct 3D image for a human heart, we should take heart beat and breathing problems into consideration. Grouping these ultrasound images according to heart beat and breathing, we can then fill the gaps of 3D heart model using a distance weight algorithm among multiple image planes. We have adopted several rendering techniques to visualize the 3D ultrasound data, such as volume rendering by ray casting, fast shear-warp volume rendering, and depth-cueing rendering, in order to show the movement of heart during a cardiac cycle dynamically. Ouhyoung,Ming 歐陽明 1996 學位論文 ; thesis 92 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立臺灣大學 === 資訊工程研究所 === 84 === This thesis proposes new techniques for visualization and 3D reconstruction of cardiac ultrasound images. Ultrasonic imaging techniques have been broadly used in clinical diagnosis, because of lower cost and less harm, as compared to x-ray based imaging. However, we have encountered two problems in ultrasound imaging which are images of poor signal to noise ratio and the positioning problem in a sequence of unaligned images due to manual control. To get precise ultrasound images of cutting planes of heart, we attached a 3D tracker which can report the position and orientation of the transducer used in 2D echocardiography. Then, we can determine the position and orientation of the transducer with the help of the 3D tracker. In order to improve the quality of ultrasound images, several image processing methods have been applied. We used three kinds of image processing techniques to enhance the quality of image, such as background subtraction, image segmentation, and image filter. Besides, human heart is a moving organ with a semi- constant period. The movement of human heart in its 3D position is not only caused by heart beat, but also due to a little displacement following inhaling and exhaling. Therefore, when we try to reconstruct 3D image for a human heart, we should take heart beat and breathing problems into consideration. Grouping these ultrasound images according to heart beat and breathing, we can then fill the gaps of 3D heart model using a distance weight algorithm among multiple image planes. We have adopted several rendering techniques to visualize the 3D ultrasound data, such as volume rendering by ray casting, fast shear-warp volume rendering, and depth-cueing rendering, in order to show the movement of heart during a cardiac cycle dynamically.
author2 Ouhyoung,Ming
author_facet Ouhyoung,Ming
Chen,Jiann-Hwa
陳建華
author Chen,Jiann-Hwa
陳建華
spellingShingle Chen,Jiann-Hwa
陳建華
3D tracker based 3D reconstruction of cardiac ultrasound images
author_sort Chen,Jiann-Hwa
title 3D tracker based 3D reconstruction of cardiac ultrasound images
title_short 3D tracker based 3D reconstruction of cardiac ultrasound images
title_full 3D tracker based 3D reconstruction of cardiac ultrasound images
title_fullStr 3D tracker based 3D reconstruction of cardiac ultrasound images
title_full_unstemmed 3D tracker based 3D reconstruction of cardiac ultrasound images
title_sort 3d tracker based 3d reconstruction of cardiac ultrasound images
publishDate 1996
url http://ndltd.ncl.edu.tw/handle/47617760655828923517
work_keys_str_mv AT chenjiannhwa 3dtrackerbased3dreconstructionofcardiacultrasoundimages
AT chénjiànhuá 3dtrackerbased3dreconstructionofcardiacultrasoundimages
AT chenjiannhwa yǐkōngjiāndìngwèiqìzuòwèixīnzàngchāoyīnbōyǐngxiàngdìngwèizhīsānwéilìtǐzhòngjiàn
AT chénjiànhuá yǐkōngjiāndìngwèiqìzuòwèixīnzàngchāoyīnbōyǐngxiàngdìngwèizhīsānwéilìtǐzhòngjiàn
_version_ 1718346082359967744