000163185 001__ 163185
000163185 005__ 20251017161900.0
000163185 0247_ $$2doi$$a10.1007/978-3-031-73281-2_9
000163185 0248_ $$2sideral$$a145647
000163185 037__ $$aART-2024-145647
000163185 041__ $$aeng
000163185 100__ $$0(orcid)0009-0001-2112-0939$$aTomasini, Clara$$uUniversidad de Zaragoza
000163185 245__ $$aSim2Real in Endoscopy Segmentation with a Novel Structure Aware Image Translation
000163185 260__ $$c2024
000163185 5203_ $$aAutomatic segmentation of anatomical landmarks in endoscopic images can provide assistance to doctors and surgeons for diagnosis, treatments or medical training. However, obtaining the annotations required to train commonly used supervised learning methods is a tedious and difficult task, in particular for real images. While ground truth annotations are easier to obtain for synthetic data, models trained on such data often do not generalize well to real data. Generative approaches can add realistic texture to it, but face difficulties to maintain the structure of the original scene. The main contribution in this work is a novel image translation model that adds realistic texture to simulated endoscopic images while keeping the key scene layout information. Our approach produces realistic images in different endoscopy scenarios. We demonstrate these images can effectively be used to successfully train a model for a challenging end task without any real labeled data. In particular, we demonstrate our approach for the task of fold segmentation in colonoscopy images. Folds are key anatomical landmarks that can occlude parts of the colon mucosa and possible polyps. Our approach generates realistic images maintaining the shape and location of the original folds, after the image-style-translation, better than existing methods. We run experiments both on a novel simulated dataset for fold segmentation, and real data from the EndoMapper (EM) dataset [1]. All our new generated data and new EM metadata is being released to facilitate further research, as no public benchmark is currently available for the task of fold segmentation.
000163185 540__ $$9info:eu-repo/semantics/closedAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000163185 592__ $$a0.352$$b2024
000163185 593__ $$aComputer Science (miscellaneous)$$c2024$$dQ2
000163185 593__ $$aTheoretical Computer Science$$c2024$$dQ3
000163185 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/publishedVersion
000163185 700__ $$0(orcid)0000-0002-6722-5541$$aRiazuelo, Luis$$uUniversidad de Zaragoza
000163185 700__ $$0(orcid)0000-0002-7580-9037$$aMurillo, Ana C.$$uUniversidad de Zaragoza
000163185 7102_ $$15007$$2520$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Ingen.Sistemas y Automát.
000163185 773__ $$g(2024), 89-101$$pLect. notes comput. sci.$$tLecture Notes in Computer Science$$x0302-9743
000163185 8564_ $$s9913232$$uhttps://zaguan.unizar.es/record/163185/files/texto_completo.pdf$$yVersión publicada
000163185 8564_ $$s1582755$$uhttps://zaguan.unizar.es/record/163185/files/texto_completo.jpg?subformat=icon$$xicon$$yVersión publicada
000163185 909CO $$ooai:zaguan.unizar.es:163185$$particulos$$pdriver
000163185 951__ $$a2025-10-17-14:07:40
000163185 980__ $$aARTICLE