Showing papers in "Medical Image Analysis in 2015"
••
TL;DR: A novel method for the automatic segmentation of vessel trees in retinal fundus images by summing up the responses of the two rotation-invariant B-COSFIRE filters followed by thresholding is introduced.
626 citations
••
TL;DR: Multi-atlas segmentation (MAS) is becoming one of the most widely used and successful image segmentation techniques in biomedical applications as mentioned in this paper, and it has been widely used in medical image classification.
587 citations
••
Utrecht University1, Case Western Reserve University2, National University of Colombia3, Technical University of Denmark4, Dalle Molle Institute for Artificial Intelligence Research5, Işık University6, Curie Institute7, French Institute of Health and Medical Research8, PSL Research University9, National Taiwan University of Science and Technology10, Konica Minolta11, Panasonic12, University of Central Lancashire13, University of Nice Sophia Antipolis14, University of Surrey15, University of Sheffield16, University of Warwick17, Qatar University18, Princess Alexandra Hospital NHS Trust19
TL;DR: The results from the Assessment of Mitosis Detection Algorithms 2013 (AMIDA13) challenge are described and the top performing method has an error rate that is comparable to the inter-observer agreement among pathologists.
405 citations
••
TL;DR: This paper compares its approach with a recently presented descriptor of pulmonary nodule morphology, namely Bag of Frequencies, and illustrates the advantages offered by the two strategies, achieving performance of AUC = 0.868, which is close to the one of human experts.
270 citations
••
TL;DR: It is concluded that most segmentation methods may benefit from GPU processing due to the methods' data parallel structure and high thread count, however, factors such as synchronization, branch divergence and memory usage can limit the speedup.
230 citations
••
TL;DR: Best results show that an average 80% Dice accuracy and a 1cm Hausdorff distance can be expected from semi-automated algorithms for this challenging task on the datasets, and that an automated algorithm can reach similar performance, at the expense of a high computational burden.
220 citations
••
TL;DR: Experimental results have shown that this novel methodology can uncover multiple functional networks that can be well characterized and interpreted in spatial, temporal and frequency domains based on current brain science knowledge.
175 citations
••
TL;DR: It is shown that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques, and is able to stratify TBI patients with favourable outcomes from non-favourable outcomes.
157 citations
••
TL;DR: An automated segmentation method is presented for multi-organ segmentation in abdominal CT images and achieves a promising segmentation performance with Dice overlap values of 94.9%, 93.6%, 71.1%, and 92.5% for liver, kidneys, pancreas, and spleen, respectively.
156 citations
••
TL;DR: This paper proposes to combine the intensity, gradient and contextual information into an augmented feature vector and incorporate it into multi-atlas segmentation, and explores the alternative to the K nearest neighbour (KNN) classifier in performing multi- atlas label fusion, by using the support vector machine (SVM) for label fusion instead.
151 citations
••
TL;DR: A novel method for MRI denoising that exploits both the sparseness and self-similarity properties of the MR images by automatically estimating the local noise level present in the image and using it as a guide image within a rotationally invariant non-local means filter.
••
TL;DR: A general framework of multi-organ segmentation which effectively incorporates interrelations among multiple organs and easily adapts to various imaging conditions without the need for supervised intensity information is proposed.
••
TL;DR: In this paper, a random forest classifier is combined with an anisotropic smoothing prior in a Conditional Random Field framework to generate multiple segmentation hypotheses per image, which are then combined into geometrically consistent 3D objects by segmentation fusion.
••
TL;DR: The results suggest that the proposed FA system improves upon the state-of-the-art, and the SA system offers a considerable boost over the FA system.
••
TL;DR: This work has shown how statistical parametric mapping (SPM) can be combined with a general linear model to study the impact of gender and age on regional myocardial wall thickness and investigated the influence of the population size on atlas construction and atlas-based analysis.
••
TL;DR: A novel automated breast cancer localization system for DCE-MRI that initially corrects for motion artifacts and segments the breast, and a malignancy score for each lesion candidate is obtained using region-based morphological and kinetic features computed on the segmented lesion candidates.
••
TL;DR: The first spatiotemporal (4D) high-definition cortical surface atlases for the dynamic developing infant cortical structures are constructed, and for the first time, they reveal the spatially-detailed, region-specific correlation patterns of the dynamic cortical developmental trajectories between different cortical regions during early brain development.
••
TL;DR: The SPArse Reconstruction Challenge (SPARC) was held along with the workshop on Computational Diffusion MRI to validate the performance of multiple reconstruction methods using data acquired from a physical phantom to provide appropriate guidelines to neuroscientists on making an informed decision while designing their acquisition protocols.
••
TL;DR: How a stack-of-stars MRI acquisition on integrated PET/MRI systems can be used to derive a high-resolution motion model, how many respiratory phases need to be differentiated, how much MRI scan time is required, and how the model is employed for motion-corrected PET reconstruction are described.
••
TL;DR: The cortical bone mapping technique is developed by exploiting local estimates of imaging blur to correct the global density estimate, thus providing a local density estimate as well as more accurate estimates of thickness.
••
TL;DR: It is shown that performance on multi-organ classification can be improved by accounting for exogenous information through Bayesian priors (so called context learning) and moves toward efficient segmentation of large-scale clinically acquired CT data for biomarker screening, surgical navigation, and data mining.
••
TL;DR: The results demonstrate that the participating methods were able to segment all tissue classes well, except myelinated white matter.
••
TL;DR: Experimental results over synthetic and real 3D MR data demonstrate that the proposed wiener-augmented HOSVD method outperforms current state-of-the-art denoising methods.
••
TL;DR: It is demonstrated that ultrasound parameters derived from the ultrasound backscattered power spectrum can potentially serve as non-invasive early measures of clinical tumor response to chemotherapy treatments.
••
TL;DR: It is shown that the learning-based approach outperforms different state-of-the-art methods and proves highly accurate and robust with regard to both vessel segmentation and centerline extraction in spite of the high level of label noise in the training data.
••
TL;DR: A robust segmentation method is developed to delineate cells accurately using Gaussian-based hierarchical voting and repulsive balloon model to enable cell-level analysis in a real-time fashion of histopathological images.
••
TL;DR: DXA-based FE simulation was able to explain 85% of the CT-predicted strength of the femur in stance loading, and the present method can be used to accurately reconstruct the 3D shape and internal density of the Femur from 2D DXA images.
••
TL;DR: The proposed approach effectively works with non-fluorescein fundus images and proves highly accurate and robust in complicated regions such as the central reflex, close vessels, and crossover points, despite a high level of illumination noise in the original data.
••
TL;DR: An image synthesis approach that first estimates the pulse sequence parameters of the subject image to yield the particular target pulse sequence within the atlas is presented and superior in both intensity standardization and synthesis to other established methods is shown.
••
TL;DR: This work proposes a new method that can accurately estimate the non-stationary parameters of noise from just a single magnitude image and shows the better performance and the lowest error variance of the proposed methodology when compared to the state-of-the-art methods.