Browsing by Subject "Regions of interest"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item Open Access A clustering-based method to estimate saliency in 3D animated meshes(Elsevier Ltd, 2014) Bulbul, A.; Arpa, S.; Capin, T.We present a model to determine the perceptually significant elements in animated 3D scenes using a motion-saliency method. Our model clusters vertices with similar motion-related behaviors. To find these similarities, for each frame of an animated mesh sequence, vertices' motion properties are analyzed and clustered using a Gestalt approach. Each cluster is analyzed as a single unit and representative vertices of each cluster are used to extract the motion-saliency values of each group. We evaluate our method by performing an eye-tracker-based user study in which we analyze observers' reactions to vertices with high and low saliencies. The experiment results verify that our proposed model correctly detects the regions of interest in each frame of an animated mesh. © 2014 Elsevier Ltd.Item Open Access Influence of cigarette smoking on white matter in patients with clinically isolated syndrome as detected by diffusion tensor imaging(Turkish Society of Radiology, 2016) Durhan, G.; Diker, S.; Has, A. C.; Karakaya, J.; Kurne, A. T.; Oguz, K. K.PURPOSE Cigarette smoking has been associated with increased occurrence of multiple sclerosis (MS), as well as clinical disability and disease progression in MS. We aimed to assess the effects of smoking on the white matter (WM) in patients with clinically isolated syndrome (CIS) using diffusion tensor imaging. METHODS Smoker patients with CIS (n=16), smoker healthy controls (n=13), nonsmoker patients with CIS (n=17) and nonsmoker healthy controls (n=14) were included. Thirteen regions-of-interest including nonenhancing T1 hypointense lesion and perilesional WM, and 11 normal-appearing white matter (NAWM) regions were drawn on color-coded fractional anisotropy (FA) maps. Lesion load was determined in terms of number and volume of WM hyperintensities. RESULTS A tendency towards greater lesion load was found in smoker patients. T1 hypointense lesions and perilesional WM had reduced FA and increased mean diffusivity to a similar degree in smoker and nonsmoker CIS patients. Compared with healthy smokers, smoker CIS patients had more extensive NAWM changes shown by increased mean diffusivity. There was no relationship between diffusion metrics and clinical disability scores, duration of the disease and degree of smoking exposure. CONCLUSION Smoker patients showed a tendency towards having greater number of WM lesions and displayed significantly more extensive NAWM abnormalities. © Turkish Society of Radiology 2016.Item Open Access Localization of diagnostically relevant regions of interest in whole slide images(IEEE, 2014-08) Mercan, E.; Aksoy, Selim; Shapiro, L. G.; Weaver, D. L.; Brunye, T.; Elmore, J. G.Whole slide imaging technology enables pathologists to screen biopsy images and make a diagnosis in a digital form. This creates an opportunity to understand the screening patterns of expert pathologists and extract the patterns that lead to accurate and efficient diagnoses. For this purpose, we are taking the first step to interpret the recorded actions of world-class expert pathologists on a set of digitized breast biopsy images. We propose an algorithm to extract regions of interest from the logs of image screenings using zoom levels, time and the magnitude of panning motion. Using diagnostically relevant regions marked by experts, we use the visual bag-of-words model with texture and color features to describe these regions and train probabilistic classifiers to predict similar regions of interest in new whole slide images. The proposed algorithm gives promising results for detecting diagnostically relevant regions. We hope this attempt to predict the regions that attract pathologists' attention will provide the first step in a more comprehensive study to understand the diagnostic patterns in histopathology.Item Open Access Nitelik tabanlı sınıflandırıcılar ve koşullu rastgele alan ile dikkat çeken görsel bölge tespiti(IEEE, 2016-05) Demirel, B.; Cinbiş, Ramazan Gökberk; İkizler-Cinbiş, N.Dikkat çeken görsel bölge tahmini, resimlerde ya da sahnelerde insan gözünün öncelikli olarak odaklandıgı bölgeleri bulmayı amaçlayan bir bilgisayarlı görü problemidir. Pekçok bilgisayarlı görü problemi bir sahnedeki arkaplan ögelerini yoksaymayı gerektirdigi için, bu tür problemlerde dikkat çeken görsel bölge tahmini bir önişlem adımı olarak kullanılabilir. Bu çalışmada yukarıdan aşagıya dikkat çeken bölge tahmini probleminin çözümüne yönelik olarak nitelik tabanlı sınıflandırıçılar ve Koşullu Rastgele Alan (KRA) yöntemlerinin bir arada kullanıldıgı bir yöntem sunulmaktadır. Deneysel sonuçlar nitelik tabanlı sınıflandırıcı sonuçlarının görsel bilgiyi alt seviye özelliklere göre daha iyi kodlayabildigini göstermiştir ve geliştirilen yöntemin, Graz-02 veri kümesi üzerinde en iyi yöntemlerle karşılaştırıldıgında umut verici sonuçlar ürettiği gözlenmiştir.