A novel game-theoretic framework for landmark-based image segmentation is presented. Landmark detection is formulated as a game, in which landmarks are players, landmark candidate points are strategies, and likelihoods that candidate points represent landmarks are payoffs, determined according to the similarity of image intensities and spatial relationships between the candidate points in the target image and their corresponding landmarks in images from the training set. The solution of the formulated game-theoretic problem is the equilibrium of candidate points that represent landmarks in the target image and is obtained by a novel iterative scheme that solves the segmentation problem in polynomial time. The performance of the proposed framework was evaluated for segmentation of lung fields from chest radiographs and heart ventricles from cardiac magnetic resonance cross-sections. The comparison to other landmark-based segmentation techniques shows that the results obtained by the proposed game-theoretic framework are highly accurate and precise.
COBISS.SI-ID: 9231956
Endovascular image-guided interventions (EIGIs) involve navigation of a catheter through the vasculature followed by application of treatment at the site of anomaly using live 2D projection images for guidance. 3D images acquired prior to EIGI are used to quantify the vascular anomaly and plan the intervention. If fused with the information of live 2D images they can also facilitate navigation and treatment. For this purpose 3D2D image registration is required. Although several 3D2D registration methods for EIGI achieve registration accuracy below 1 mm, their clinical application is still limited by insufficient robustness or reliability. In this paper, a 3D2D registration method based on matching a 3D vasculature model to intensity gradients of live 2D images is proposed. To objectively validate 3D2D registration methods, we acquired a clinical image database of 10 patients undergoing cerebral EIGI and established "gold standard" registrations by aligning fiducial markers in 3D and 2D images. The proposed method had mean registration accuracy below 0.65 mm, which was comparable to tested state-of-the-art methods, and execution time below 1s. With the highest rate of successful registrations and the highest capture range the proposed method was the most robust and thus a good candidate for being incorporated in EIGI.
COBISS.SI-ID: 9795156
The aim of our study was to explore whether earlier hip arthroplasty for idiopathic osteoarthritis (OA) might be explained by enlarged contact stress in the hip joint, and to what extent can that be attributed to obesity and biomechanical constitution of the pelvis. Fifty subjects were selected from a list of consecutive recipients of hip endoprosthesis due to idiopathic OA; standard pelvic radiographs made years prior to surgery were the main selection criteria. For 65 hips resultant hip force and peak contact hip stress normalized to the body weight (R/Wb and p(max)/Wb) were determined from the radiographs with the Hipstress method. Body weight and body mass index (BMI) were obtained with an interview. Regression analysis was used to correlate parameters of obesity (body weight,BMI), biomechanical constitution of the pelvis (R/Wb, p(max)/Wb) and mechanical loading within the hip joint (R, p(max)) with age at hip arthroplasty. We found that younger age at hip arthroplasty was associated with higher body weight (P=0.009), higher peak contact hip stress normalized to the body weight - p(max)/Wb (P=0.019), higher resultant hip force -R (P=0.027) and larger peak contact hip stress - p(max) (P(0.001), but not with BMI (P=0.121) or R/Wb (P=0.614). The results suggest that enlarged contact stress plays an important role in rapid progression of hip OA with both obesity and unfavorable biomechanical constitution of the pelvis (greater p(max)/Wb) contributing.
COBISS.SI-ID: 25556185
Quantification of the time-varying nature of cardiorespiratory interactions is important in study of normal subjects and patients with various cardiovascular diseases or diseases that affect the cardiovascular system. A new method is introduced for analysis of interactions between time-dependent coupled oscillators, based on the signal they generate. The method is capable to distinguishe unsynchronized dynamics from noise-induced phaseslips and enables the evolution of the coupling functions and other parameters to be followed. It is based on phase dynamics, with Bayesian inference of the time-evolving parameters achieved by shaping the prior densities to incorporate knowledge of previous samples. The method is tested numerically and applied to reveal and quantify the timevarying nature of cardiorespiratory interactions.
COBISS.SI-ID: 9358420
We consider the Kuramoto model of an ensemble of interacting oscillators allowing for an arbitrary distribution of frequencies and coupling strengths. We define a family of traveling wave states as stationary in a rotating frame, and derive general equations for their parameters. We suggest empirical stability conditions which, for the case of incoherence, become exact. In addition to making new theoretical predictions, we show that many earlier results follow naturally from our general framework. The results are applicable in scientific contexts ranging from physics to biology.
COBISS.SI-ID: 9778772