Show simple item record

dc.contributor.advisorÜnal, Gözde
dc.contributor.authorİşgüder, Gözde Gül
dc.date.accessioned2020-12-10T07:35:45Z
dc.date.available2020-12-10T07:35:45Z
dc.date.submitted2011
dc.date.issued2018-08-06
dc.identifier.urihttps://acikbilim.yok.gov.tr/handle/20.500.12812/217085
dc.description.abstractAnahtar Kelimeler: manifold ögrenmesi, sınıflandrma, IVUS, intravasküler ultrason,görüntü tabanlı gecitleme, ECG gecitlemesi, validasyonIntravasküler Ultrason(IVUS) ic koroner damarlarn kros-kesitlerinin görüntülenmesine olanaksaglayan bir görüntüleme teknigidir. IVUS cerceveleri, sabit hzl bir motor yardmyla kateterindamardan geriye dogru cekilmesiyle elde edilmektedir. Ancak, bu gericekilme esnasnda kalbinatmasndan ötürü görüntü üzerinde istenmeyen bozulmalar meydana gelmektedir. Bu bozulmalardamar ve ic damar hacim ölcümlerini yanltmakta ve islemlerin devamllgn snrlandrmaktadr.Kardiyo sisteminin bulundugu faza ait olan zaman aralgn tespit edebilen ECG( elektrokardiyogram) sinyaliyle bu bozulmalarn önlenmesi mümkündür. Ancak, ECG sinyalinin kullanlabilmesiicin özel bir gecitleme ünitesine ihtiyac vardr ve bu ünitenin kullanm damarla ilgili önemli bilgilerinkaybolmasna yol acabilecegi gibi, her klinikte mevcut olmayabilir. Bu problemi cözmeküzere manifold ögrenmesine ve yeni bir agrlkl ultrason benzerlik ölcütüne dayal bir görüntütabanl gecitleme teknigi önerilmistir. Görüntü tabanl gecitleme teknigimizin parametreleri,verilen uzunluk ve frekansla salnabilen IVUS cekimlerini gerceklestirebilen özel bir enstrümanlaalnan 25 farkl veri kümesi üzerinde deneyler yaplarak belirlenmistir.11 degisik hasta, 22 degisik IVUS cekimi ve 88 degisik damar kesiti üzerinde nicel testler uygulanmstr. Metodun gecerliligini test etmek amacyla metodun sonuclar, ECG gecitlemesininsonuclaryla karslastrlmstr. Buna ek olarak, literaturde mevcut olan diger goruntu tabanlgecitleme yöntemleriyle karslastrmalar yaplarak, önerilen metodun verimliligi gösterilmistir.
dc.description.abstractKeywords: Manifold Learning, Classication, IVUS, Intravascular Ultrasound, Image-basedgating, ECG gating, ValidationIntravascular Ultrasound(IVUS) is an imaging technology which provides cross-sectional imagesof internal coronary vessel structures. The IVUS frames are acquired by pulling the catheterback with a motor running at a constant speed. However,üring the pullback, some artifactsoccurüe to the beating heart. These artifacts cause inaccurate measurements for total vesseland lumen volume and limitation for further processing. Elimination of these artifacts are possiblewith an ECG (electrocardiogram) signal, which determines the time interval correspondingto a particular phase of the cardiac cycle. However, using ECG signal requires a special gatingunit, which causes loss of important information about the vessel, and furthermore, ECG gatingfunction may not be available in all clinical systems. To address this problem, we propose animage-based gating technique based on manifold learning and a novel weighted ultrasound similaritymeasure. The parameters for our image-based gating technique were chosen based on theexperiments performed on 25 different in-vitro IVUS pullback sequences, which were acquiredwith the help of a special mechanical instrument that oscillates with given length and frequency.Quantitative tests are performed on 12 different patients, 25 different pullbacks and 100 differentlongitudinal vessel cuts. In order to validate our method, the results of our method are comparedto those of ECG-Gating method. In addition, comparison studies against the results obtainedfrom the state of the art methods available in the literature were carried out to demonstrate theeffectiveness of the proposed method.en_US
dc.languageEnglish
dc.language.isoen
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rightsAttribution 4.0 United Statestr_TR
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectElektrik ve Elektronik Mühendisliğitr_TR
dc.subjectElectrical and Electronics Engineeringen_US
dc.titleManifold learning for image-based gating of intravascular ultrasound (IVUS) pullback sequences
dc.title.alternativeİntravasküler ultrası-on (IVUS) çekim dizilerinin manifold öğrenmesi kullanılarak görüntü tabanlı geçitlenmesi
dc.typemasterThesis
dc.date.updated2018-08-06
dc.contributor.departmentDiğer
dc.identifier.yokid460970
dc.publisher.instituteMühendislik ve Fen Bilimleri Enstitüsü
dc.publisher.universitySABANCI ÜNİVERSİTESİ
dc.identifier.thesisid322883
dc.description.pages79
dc.publisher.disciplineDiğer


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

info:eu-repo/semantics/openAccess
Except where otherwise noted, this item's license is described as info:eu-repo/semantics/openAccess