Author
Benjamin Irving
Other affiliations: University of Cape Town, University College London
Bio: Benjamin Irving is an academic researcher from University of Oxford. The author has contributed to research in topics: Airway & Liver disease. The author has an hindex of 11, co-authored 32 publications receiving 513 citations. Previous affiliations of Benjamin Irving include University of Cape Town & University College London.
Papers
More filters
••
University of Copenhagen1, Radboud University Nijmegen Medical Centre2, University of Iowa3, Utrecht University4, University College London5, Telecom SudParis6, University of Antwerp7, Technische Universität München8, University of Łódź9, Graz University of Technology10, University of Seville11, Philips12, Cornell University13, Leipzig University14, University of Mainz15, Nagoya University16, Siemens17, New York University18, Erasmus University Rotterdam19, Copenhagen University Hospital20
TL;DR: A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.
Abstract: This paper describes a framework for establishing a reference airway tree segmentation, which was used to quantitatively evaluate 15 different airway tree extraction algorithms in a standardized manner. Because of the sheer difficulty involved in manually constructing a complete reference standard from scratch, we propose to construct the reference using results from all algorithms that are to be evaluated. We start by subdividing each segmented airway tree into its individual branch segments. Each branch segment is then visually scored by trained observers to determine whether or not it is a correctly segmented part of the airway tree. Finally, the reference airway trees are constructed by taking the union of all correctly extracted branch segments. Fifteen airway tree extraction algorithms from different research groups are evaluated on a diverse set of 20 chest computed tomography (CT) scans of subjects ranging from healthy volunteers to patients with severe pathologies, scanned at different sites, with different CT scanner brands, models, and scanning protocols. Three performance measures covering different aspects of segmentation quality were computed for all participating algorithms. Results from the evaluation showed that no single algorithm could extract more than an average of 74% of the total length of all branches in the reference standard, indicating substantial differences between the algorithms. A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.
241 citations
••
TL;DR: Doses delivered by the linear slit scanning system were significantly lower than those from conventional X-ray equipment, and effective doses were between 9 and 75% of the United Nations Scientific Committee Report on the Effects of Ionising Radiation doses for standard examinations.
Abstract: Doses for a range of examinations and views using digital X-ray equipment with full-body linear slit scanning capabilities (Statscan) have been compared with those from other published studies. Entrance doses (free-in-air) were measured using a dosimeter, and effective doses were generated using a Monte Carlo simulator. Doses delivered by the linear slit scanning system were significantly lower than those from conventional X-ray equipment. Effective doses were between 9 and 75% of the United Nations Scientific Committee Report on the Effects of Ionising Radiation doses for standard examinations. This dose reduction can be explained by the properties of linear slit scanning technology, including low scatter, beam geometry, the use of a digital detector and the use of higher than usual tube voltages.
39 citations
01 Jan 2009
TL;DR: An automatic method for segmentation of the airway tree is outlined, which includes algorithms to detect the trachea, segment thetrachea and main bronchi by thresholding and region growing, and segment the remaining Bronchi by morphological filtering and reconstruction.
Abstract: Segmentation of the airways is useful for the analysis of airway compression and obstruction caused by pathology. This paper outlines an automatic method for segmentation of the airway tree. This method includes algorithms to detect the trachea, segment the trachea and main bronchi by thresholding and region growing, and segment the remaining bronchi by morphological filtering and reconstruction. Morphological filtering and reconstruction are applied to all slices in the axial, sagittal and coronal planes and are used to extract the smaller airways. Bounded space dilation with a leak removal restriction is applied as a region growing method. This method was evaluated on 20 cases as part of the MICCAI pulmonary image analysis workshop. The mean number of branches detected as a percentage of possible branches was 43.5%, the mean tree length detected as a percentage of the entire tree length was 36.4% and the false positive rate – that is, the percentage of the total volume that was incorrectly segmented – was 1.27%
34 citations
••
14 Sep 2014TL;DR: There is a need for a consistent approach to colorectal tumour segmentation in DCE-MRI and a novel method based on detection of the tumour from signal enhancement characteristics of homogeneous tumour subregions and their neighbourhoods is proposed.
Abstract: Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a powerful protocol for assessing tumour progression from changes in tissue contrast enhancement. Manual colorectal tumour delineation is a challenging and time consuming task due to the complex enhancement patterns in the 4D sequence. There is a need for a consistent approach to colorectal tumour segmentation in DCE-MRI and we propose a novel method based on detection of the tumour from signal enhancement characteristics of homogeneous tumour subregions and their neighbourhoods. Our method successfully detected 20 of 23 cases with a mean Dice score of 0.68 ±0.15 compared to expert annotations, which is not significantly different from expert inter-rater variability of 0.73 ±0.13 and 0.77 ±0.10. In comparison, a standard DCE-MRI tumour segmentation technique, fuzzy c-means, obtained a Dice score of 0.28 ±0.17.
33 citations
••
TL;DR: An automatic segmentation method is proposed for dynamic contrast enhanced MRI that achieves promising results for the underexplored area of automatic rectal tumour segmentation from DCE-MRI scans.
28 citations
Cited by
More filters
••
Technische Universität München1, ETH Zurich2, University of Bern3, Harvard University4, National Institutes of Health5, University of Debrecen6, University Hospital Heidelberg7, McGill University8, University of Pennsylvania9, French Institute for Research in Computer Science and Automation10, University at Buffalo11, Microsoft12, University of Cambridge13, Stanford University14, University of Virginia15, Imperial College London16, Massachusetts Institute of Technology17, Columbia University18, Sabancı University19, Old Dominion University20, RMIT University21, Purdue University22, General Electric23
TL;DR: The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) as mentioned in this paper was organized in conjunction with the MICCAI 2012 and 2013 conferences, and twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low and high grade glioma patients.
Abstract: In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients—manually annotated by up to four raters—and to 65 comparable scans generated using tumor image simulation software Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%–85%), illustrating the difficulty of this task We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource
3,699 citations
••
TL;DR: This paper introduces a robust, learning-based brain extraction system (ROBEX), which combines a discriminative and a generative model to achieve the final result and shows that ROBEX provides significantly improved performance measures for almost every method/dataset combination.
Abstract: Automatic whole-brain extraction from magnetic resonance images (MRI), also known as skull stripping, is a key component in most neuroimage pipelines. As the first element in the chain, its robustness is critical for the overall performance of the system. Many skull stripping methods have been proposed, but the problem is not considered to be completely solved yet. Many systems in the literature have good performance on certain datasets (mostly the datasets they were trained/tuned on), but fail to produce satisfactory results when the acquisition conditions or study populations are different. In this paper we introduce a robust, learning-based brain extraction system (ROBEX). The method combines a discriminative and a generative model to achieve the final result. The discriminative model is a Random Forest classifier trained to detect the brain boundary; the generative model is a point distribution model that ensures that the result is plausible. When a new image is presented to the system, the generative model is explored to find the contour with highest likelihood according to the discriminative model. Because the target shape is in general not perfectly represented by the generative model, the contour is refined using graph cuts to obtain the final segmentation. Both models were trained using 92 scans from a proprietary dataset but they achieve a high degree of robustness on a variety of other datasets. ROBEX was compared with six other popular, publicly available methods (BET, BSE, FreeSurfer, AFNI, BridgeBurner, and GCUT) on three publicly available datasets (IBSR, LPBA40, and OASIS, 137 scans in total) that include a wide range of acquisition hardware and a highly variable population (different age groups, healthy/diseased). The results show that ROBEX provides significantly improved performance measures for almost every method/dataset combination.
539 citations
••
University of Colorado Denver1, Columbia University2, University of British Columbia3, Pierre-and-Marie-Curie University4, Heidelberg University5, Harvard University6, Mayo Clinic7, American Institute for Economic Research8, Université libre de Bruxelles9, University of Iowa10, University of Florence11
TL;DR: The classification system proposed and illustrated in this article provides a structured approach to visual and quantitative assessment of COPD and helps to contribute to a personalized approach to the treatment of patients with COPD.
Abstract: Integration of visual characterization of emphysema and airway abnormalities with physiologic and quantitative CT assessment permits categorization of chronic obstructive pulmonary disease into distinct structurally and functionally defined subtypes.
389 citations
••
University of Copenhagen1, Radboud University Nijmegen Medical Centre2, University of Iowa3, Utrecht University4, University College London5, Telecom SudParis6, University of Antwerp7, Technische Universität München8, University of Łódź9, Graz University of Technology10, University of Seville11, Philips12, Cornell University13, Leipzig University14, University of Mainz15, Nagoya University16, Siemens17, New York University18, Erasmus University Rotterdam19, Copenhagen University Hospital20
TL;DR: A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.
Abstract: This paper describes a framework for establishing a reference airway tree segmentation, which was used to quantitatively evaluate 15 different airway tree extraction algorithms in a standardized manner. Because of the sheer difficulty involved in manually constructing a complete reference standard from scratch, we propose to construct the reference using results from all algorithms that are to be evaluated. We start by subdividing each segmented airway tree into its individual branch segments. Each branch segment is then visually scored by trained observers to determine whether or not it is a correctly segmented part of the airway tree. Finally, the reference airway trees are constructed by taking the union of all correctly extracted branch segments. Fifteen airway tree extraction algorithms from different research groups are evaluated on a diverse set of 20 chest computed tomography (CT) scans of subjects ranging from healthy volunteers to patients with severe pathologies, scanned at different sites, with different CT scanner brands, models, and scanning protocols. Three performance measures covering different aspects of segmentation quality were computed for all participating algorithms. Results from the evaluation showed that no single algorithm could extract more than an average of 74% of the total length of all branches in the reference standard, indicating substantial differences between the algorithms. A fusion scheme that obtained superior results is presented, demonstrating that there is complementary information provided by the different algorithms and there is still room for further improvements in airway segmentation algorithms.
241 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