Vis enkel innførsel

dc.contributor.advisorLindseth, Franknb_NO
dc.contributor.authorØygard, Tordnb_NO
dc.date.accessioned2014-12-19T13:41:16Z
dc.date.available2014-12-19T13:41:16Z
dc.date.created2014-05-31nb_NO
dc.date.issued2014nb_NO
dc.identifier720588nb_NO
dc.identifierntnudaim:10324nb_NO
dc.identifier.urihttp://hdl.handle.net/11250/253675
dc.description.abstractUltrasound is a flexible medical imaging modality with many uses, one ofthem being intra-operative imaging for use in navigation. In order to obtainthe highest possible spatial resolution and avoiding big, clunky 3D ultra-sound probes, reconstruction of many 2D ultrasound images obtained by aconventional 2D ultrasound probe with a tracking system attached has beenemployed.Earlier work in this field has yielded fast Graphical Processing Unit(GPU)-based implementations of voxel-based reconstruction algorithms such as VoxelNearest Neighbor(VNN), Pixel Nearest Neighbor(PNN), VNN2 and Dis-tance Weighted(DW) reconstruction. However, it is desirable to improveupon the reconstruction quality of the methods mentioned above. To doso, we propose in this thesis an adaptive algorithm called VGDW, whichtries to intelligently smooth away speckles and noise, yet retains detail inhigh-frequency regions, while being not being much slower than the abovementioned algorithms. It also has a tunable weight function enabling valuecollisions to be handled gracefully.Using our novel adaptive algorithm, we are able to produce very high-qualityreconstructions, which are unanimously preferred over the output of theabove mentioned algorithms by both a group of medical personnel and agroup of technologists working with ultrasound, while having comparablecomputation time to VNN2 and DW, i.e. 16%, 10% and 5% difference fromDW when computing a volume with 128 millions of voxels from a small,medium-sized and very large input data set using an AMD Radeon 6470MGPU. The algorithm also performs especially well with complex scanningpatterns with overlapping data when using a customized weight function. Asfor future work, there are some aspects of the weight function that can benefitfrom improvement. Also, turning the problem upside down and looking atit from a pixel-based perspective could potentially give huge benefits bothin terms of probe movement robustness and performance.nb_NO
dc.languageengnb_NO
dc.publisherInstitutt for datateknikk og informasjonsvitenskapnb_NO
dc.titleImproved Distance Weighted GPU-based 3D Ultrasound Reconstruction Methodsnb_NO
dc.typeMaster thesisnb_NO
dc.source.pagenumber156nb_NO
dc.contributor.departmentNorges teknisk-naturvitenskapelige universitet, Fakultet for informasjonsteknologi, matematikk og elektroteknikk, Institutt for datateknikk og informasjonsvitenskapnb_NO


Tilhørende fil(er)

Thumbnail
Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel