176 results found
Serruys PW, Chichareon P, Modolo R, et al., 2019, The SYNTAX score on its way out or … towards artificial intelligence: part I., EuroIntervention
Serruys PW, Chichareon P, Modolo R, et al., 2019, The SYNTAX score on its way out or … towards artificial intelligence: part II., EuroIntervention
Castro DC, Tan J, Kainz B, et al., 2019, Morpho-MNIST: Quantitative Assessment and Diagnostics for Representation Learning, Journal of Machine Learning Research, ISSN: 1532-4435
Revealing latent structure in data is an active field of research, havingintroduced exciting technologies such as variational autoencoders andadversarial networks, and is essential to push machine learning towardsunsupervised knowledge discovery. However, a major challenge is the lack ofsuitable benchmarks for an objective and quantitative evaluation of learnedrepresentations. To address this issue we introduce Morpho-MNIST, a frameworkthat aims to answer: "to what extent has my model learned to represent specificfactors of variation in the data?" We extend the popular MNIST dataset byadding a morphometric analysis enabling quantitative comparison of trainedmodels, identification of the roles of latent variables, and characterisationof sample diversity. We further propose a set of quantifiable perturbations toassess the performance of unsupervised and supervised methods on challengingtasks such as outlier detection and domain adaptation. Data and code areavailable at https://github.com/dccastro/Morpho-MNIST.
Glocker B, Robinson R, Castro DC, et al., 2019, Machine Learning with Multi-Site Imaging Data: An Empirical Study on the Impact of Scanner Effects, Medical Imaging meets NeurIPS
This is an empirical study to investigate the impact of scanner effects whenusing machine learning on multi-site neuroimaging data. We utilize structuralT1-weighted brain MRI obtained from two different studies, Cam-CAN and UKBiobank. For the purpose of our investigation, we construct a datasetconsisting of brain scans from 592 age- and sex-matched individuals, 296subjects from each original study. Our results demonstrate that even aftercareful pre-processing with state-of-the-art neuroimaging pipelines aclassifier can easily distinguish between the origin of the data with very highaccuracy. Our analysis on the example application of sex classificationsuggests that current approaches to harmonize data are unable to removescanner-specific bias leading to overly optimistic performance estimates andpoor generalization. We conclude that multi-site data harmonization remains anopen challenge and particular care needs to be taken when using such data withadvanced machine learning methods for predictive modelling.
Steyerberg EW, Wiegers E, Sewalt C, et al., 2019, Case-mix, care pathways, and outcomes in patients with traumatic brain injury in CENTER-TBI: a European prospective, multicentre, longitudinal, cohort study., Lancet Neurol, Vol: 18, Pages: 923-934
BACKGROUND: The burden of traumatic brain injury (TBI) poses a large public health and societal problem, but the characteristics of patients and their care pathways in Europe are poorly understood. We aimed to characterise patient case-mix, care pathways, and outcomes of TBI. METHODS: CENTER-TBI is a Europe-based, observational cohort study, consisting of a core study and a registry. Inclusion criteria for the core study were a clinical diagnosis of TBI, presentation fewer than 24 h after injury, and an indication for CT. Patients were differentiated by care pathway and assigned to the emergency room (ER) stratum (patients who were discharged from an emergency room), admission stratum (patients who were admitted to a hospital ward), or intensive care unit (ICU) stratum (patients who were admitted to the ICU). Neuroimages and biospecimens were stored in repositories and outcome was assessed at 6 months after injury. We used the IMPACT core model for estimating the expected mortality and proportion with unfavourable Glasgow Outcome Scale Extended (GOSE) outcomes in patients with moderate or severe TBI (Glasgow Coma Scale [GCS] score ≤12). The core study was registered with ClinicalTrials.gov, number NCT02210221, and with Resource Identification Portal (RRID: SCR_015582). FINDINGS: Data from 4509 patients from 18 countries, collected between Dec 9, 2014, and Dec 17, 2017, were analysed in the core study and from 22 782 patients in the registry. In the core study, 848 (19%) patients were in the ER stratum, 1523 (34%) in the admission stratum, and 2138 (47%) in the ICU stratum. In the ICU stratum, 720 (36%) patients had mild TBI (GCS score 13-15). Compared with the core cohort, the registry had a higher proportion of patients in the ER (9839 [43%]) and admission (8571 [38%]) strata, with more than 95% of patients classified as having mild TBI. Patients in the core study were older than those in previous studies (median age 50 years [IQR 30-66], 1254 [28%] aged >65
Jimenez-Pastor A, Alberich-Bayarri A, Fos-Guarinos B, et al., 2019, Automated vertebrae localization and identification by decision forests and image-based refinement on real-world CT data., Radiol Med
PURPOSE: Development of a fully automatic algorithm for the automatic localization and identification of vertebral bodies in computed tomography (CT). MATERIALS AND METHODS: This algorithm was developed using a dataset based on real-world data of 232 thoraco-abdominopelvic CT scans retrospectively collected. In order to achieve an accurate solution, a two-stage automated method was developed: decision forests for a rough prediction of vertebral bodies position, and morphological image processing techniques to refine the previous detection by locating the position of the spinal canal. RESULTS: The mean distance error between the predicted vertebrae centroid position and truth was 13.7 mm. The identification rate was 79.6% on the thoracic region and of 74.8% on the lumbar segment. CONCLUSION: The algorithm provides a new method to detect and identify vertebral bodies from arbitrary field-of-view body CT scans.
McCouat J, Glocker B, 2019, Vertebrae detection and localization in CT with two-stage CNNs and dense annotations, Computational Methods and Clinical Applications in Musculoskeletal Imaging (MSKI), Publisher: Springer Verlag, ISSN: 0302-9743
We propose a new, two-stage approach to the vertebrae cen-troid detection and localization problem. The first stage detects wherethe vertebrae appear in the scan using 3D samples, the second identifiesthe specific vertebrae within that region-of-interest using 2D slices. Oursolution utilizes new techniques to improve the accuracy of the algorithmsuch as a revised approach to dense labelling from sparse centroid anno-tations and usage of large anisotropic kernels in the base level of a U-netarchitecture to maximize the receptive field. Our method improves thestate-of-the-art’s mean localization accuracy by 0.87mm on a publiclyavailable spine CT benchmark.
Dou Q, Coelho De Castro D, Kamnitsas K, et al., 2019, Domain Generalization via Model-Agnostic Learning of Semantic Features, Neural Information Processing Systems (NeurIPS), Publisher: Massachusetts Institute of Technology Press, ISSN: 1049-5258
Monteiro M, Kamnitsas K, Ferrante E, et al., 2019, TBI Lesion Segmentation in Head CT: Impact of Preprocessing and Data Augmentation, MICCAI Brain Lesion Workshop
Pawlowski N, Bhooshan S, Ballas N, et al., 2019, Needles in haystacks: On classifying tiny objects in large images, Publisher: arXiv
In some computer vision domains, such as medical or hyperspectral imaging, wecare about the classification of tiny objects in large images. However, mostConvolutional Neural Networks (CNNs) for image classification were developedand analyzed using biased datasets that contain large objects, most often, incentral image positions. To assess whether classical CNN architectures workwell for tiny object classification we build a comprehensive testbed containingtwo datasets: one derived from MNIST digits and other from histopathologyimages. This testbed allows us to perform controlled experiments to stress-testCNN architectures using a broad spectrum of signal-to-noise ratios. Ourobservations suggest that: (1) There exists a limit to signal-to-noise belowwhich CNNs fail to generalize and that this limit is affected by dataset size -more data leading to better performances; however, the amount of training datarequired for the model to generalize scales rapidly with the inverse of theobject-to-image ratio (2) in general, higher capacity models exhibit bettergeneralization; (3) when knowing the approximate object sizes, adaptingreceptive field is beneficial; and (4) for very small signal-to-noise ratio thechoice of global pooling operation affects optimization, whereas for relativelylarge signal-to-noise values, all tested global pooling operations exhibitsimilar performance.
Sokooti H, Saygili G, Glocker B, et al., 2019, Quantitative error prediction of medical image registration using regression forests, Medical Image Analysis, Vol: 56, Pages: 110-121, ISSN: 1361-8415
Predicting registration error can be useful for evaluation of registration procedures, which is important for the adoption of registration techniques in the clinic. In addition, quantitative error prediction can be helpful in improving the registration quality. The task of predicting registration error is demanding due to the lack of a ground truth in medical images. This paper proposes a new automatic method to predict the registration error in a quantitative manner, and is applied to chest CT scans. A random regression forest is utilized to predict the registration error locally. The forest is built with features related to the transformation model and features related to the dissimilarity after registration. The forest is trained and tested using manually annotated corresponding points between pairs of chest CT scans in two experiments: SPREAD (trained and tested on SPREAD) and inter-database (including three databases SPREAD, DIR-Lab-4DCT and DIR-Lab-COPDgene). The results show that the mean absolute errors of regression are 1.07 ± 1.86 and 1.76 ± 2.59 mm for the SPREAD and inter-database experiment, respectively. The overall accuracy of classification in three classes (correct, poor and wrong registration) is 90.7% and 75.4%, for SPREAD and inter-database respectively. The good performance of the proposed method enables important applications such as automatic quality control in large-scale image analysis.
Pawlowski N, Glocker B, 2019, Is texture predictive for age and sex in brain MRI?, Publisher: arXiv
Deep learning builds the foundation for many medical image analysis taskswhere neuralnetworks are often designed to have a large receptive field toincorporate long spatialdependencies. Recent work has shown that largereceptive fields are not always necessaryfor computer vision tasks on naturalimages. We explore whether this translates to certainmedical imaging tasks suchas age and sex prediction from a T1-weighted brain MRI scans.
Dou Q, Ouyang C, Chen C, et al., 2019, PnP-AdaNet: plug-and-play adversarial domain adaptation network at unpaired cross-modality cardiac segmentation, IEEE Access, ISSN: 2169-3536
Deep convolutional networks have demonstrated the state-of-the-art performance on variouschallenging medical image processing tasks. Leveraging images from different modalities for the sameanalysis task holds large clinical benefits. However, the generalization capability of deep networks ontest data sampled from different distribution remains as a major challenge. In this paper, we propose aPnP-AdaNet(plug-and-play adversarial domain adaptation network) for adapting segmentation networksbetween different modalities of medical images, e.g., MRI and CT. We tackle the significant domain shift byaligning the feature spaces of source and target domains at multiple scales in an unsupervised manner. Withadversarial loss, we learn a domain adaptation module which flexibly replaces the early encoder layers of thesource network, and the higher layers are shared between two domains. We validate our domain adaptationmethod on cardiac segmentation in unpaired MRI and CT, with four different anatomical structures. Theaverage Dice achieved 63.9%, which is a significant recover from the complete failure (Dice score of13.2%) if we directly test a MRI segmentation network on CT data. In addition, our proposedPnP-AdaNetoutperforms many state-of-the-art unsupervised domain adaptation approaches on the same dataset. Theexperimental results with comprehensive ablation studies have demonstrated the excellent efficacy of ourproposed method for unsupervised cross-modality domain adaptation. Our code is publically available at:https://github.com/carrenD/Medical-Cross-Modality-Domain-Adaptation
Lee M, Oktay O, Schuh A, et al., 2019, Image-and-spatial transformer networks for structure-guided image registration, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: Springer Verlag, ISSN: 0302-9743
mage registration with deep neural networks has become anactive field of research and exciting avenue for a long standing problem inmedical imaging. The goal is to learn a complex function that maps theappearance of input image pairs to parameters of a spatial transforma-tion in order to align corresponding anatomical structures. We argue andshow that the current direct, non-iterative approaches are sub-optimal,in particular if we seek accurate alignment of Structures-of-Interest (SoI).Information about SoI is often available at training time, for example,in form of segmentations or landmarks. We introduce a novel, genericframework, Image-and-Spatial Transformer Networks (ISTNs), to lever-age SoI information allowing us to learn new image representations thatare optimised for the downstream registration task. Thanks to these rep-resentations we can employ a test-specific, iterative refinement over thetransformation parameters which yields highly accurate registration evenwith very limited training data. Performance is demonstrated on pairwise3D brain registration and illustrative synthetic data.
Li Z, Kamnitsas K, Glocker B, 2019, Overfitting of neural nets under class imbalance: Analysis and improvements for segmentation, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: Springer Verlag, ISSN: 0302-9743
Overfitting in deep learning has been the focus of a num-ber of recent works, yet its exact impact on the behaviour of neuralnetworks is not well understood. This study analyzes overfitting by ex-amining how the distribution of logits alters in relation to how muchthe model overfits. Specifically, we find that when training with few datasamples, the distribution of logit activations when processing unseen testsamples of an under-represented class tends to shift towards and evenacross the decision boundary, while the over-represented class seems un-affected. In image segmentation, foreground samples are often heavilyunder-represented. We observe that sensitivity of the model drops asa result of overfitting, while precision remains mostly stable. Based onour analysis, we derive asymmetric modifications of existing loss func-tions and regularizers including a large margin loss, focal loss, adver-sarial training and mixup, which specifically aim at reducing the shiftobserved when embedding unseen samples of the under-represented class.We study the case of binary segmentation of brain tumor core and showthat our proposed simple modifications lead to significantly improvedsegmentation performance over the symmetric variants.
Zlocha M, Dou Q, Glocker B, 2019, Improving retinanet for CT lesion detection with dense masks from weak recist labels., International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: arXiv, ISSN: 0302-9743
Accurate, automated lesion detection in Computed Tomography (CT) is animportant yet challenging task due to the large variation of lesion types,sizes, locations and appearances. Recent work on CT lesion detection employstwo-stage region proposal based methods trained with centroid or bounding-boxannotations. We propose a highly accurate and efficient one-stage lesiondetector, by re-designing a RetinaNet to meet the particular challenges inmedical imaging. Specifically, we optimize the anchor configurations using adifferential evolution search algorithm. For training, we leverage the responseevaluation criteria in solid tumors (RECIST) annotation which are measured inclinical routine. We incorporate dense masks from weak RECIST labels, obtainedautomatically using GrabCut, into the training objective, which in combinationwith other advancements yields new state-of-the-art performance. We evaluateour method on the public DeepLesion benchmark, consisting of 32,735 lesionsacross the body. Our one-stage detector achieves a sensitivity of 90.77% at 4false positives per image, significantly outperforming the best reportedmethods by over 5%.
Winzeck S, Mocking SJT, Bezerra R, et al., 2019, Ensemble of Convolutional Neural Networks Improves Automated Segmentation of Acute Ischemic Lesions Using Multiparametric Diffusion-Weighted MRI, AMERICAN JOURNAL OF NEURORADIOLOGY, Vol: 40, Pages: 938-945, ISSN: 0195-6108
Walker I, Glocker B, 2019, Graph convolutional Gaussian processes, International Conference on Machine Learning (ICML), Publisher: PMLR, Pages: 6495-6504, ISSN: 2640-3498
We propose a novel Bayesian nonparametricmethod to learn translation-invariant relationshipson non-Euclidean domains. The resulting graphconvolutional Gaussian processes can be appliedto problems in machine learning for which theinput observations are functions with domains ongeneral graphs. The structure of these models al-lows for high dimensional inputs while retainingexpressibility, as is the case with convolutionalneural networks. We present applications of graphconvolutional Gaussian processes to images andtriangular meshes, demonstrating their versatilityand effectiveness, comparing favorably to existingmethods, despite being relatively simple models.
Folgoc LL, Castro DC, Tan J, et al., 2019, Controlling meshes via curvature: spin transformations for pose-invariant shape processing, International Conference on Information Processing in Medical Imaging (IPMI 2019), Publisher: Springer Verlag, Pages: 221-234, ISSN: 0302-9743
We investigate discrete spin transformations, a geometric framework tomanipulate surface meshes by controlling mean curvature. Applications includesurface fairing -- flowing a mesh onto say, a reference sphere -- and meshextrusion -- e.g., rebuilding a complex shape from a reference sphere andcurvature specification. Because they operate in curvature space, theseoperations can be conducted very stably across large deformations with no needfor remeshing. Spin transformations add to the algorithmic toolbox forpose-invariant shape analysis. Mathematically speaking, mean curvature is ashape invariant and in general fully characterizes closed shapes (together withthe metric). Computationally speaking, spin transformations make thatrelationship explicit. Our work expands on a discrete formulation of spintransformations. Like their smooth counterpart, discrete spin transformationsare naturally close to conformal (angle-preserving). This quasi-conformalitycan nevertheless be relaxed to satisfy the desired trade-off between areadistortion and angle preservation. We derive such constraints and propose aformulation in which they can be efficiently incorporated. The approach isshowcased on subcortical structures.
Lavdas I, Glocker B, Rueckert D, et al., 2019, Machine learning in whole-body MRI: experiences and challenges from an applied study using multicentre data, CLINICAL RADIOLOGY, Vol: 74, Pages: 346-356, ISSN: 0009-9260
Tarroni G, Oktay O, Bai W, et al., 2019, Learning-based quality control for cardiac MR images, IEEE Transactions on Medical Imaging, Vol: 38, Pages: 1127-1138, ISSN: 0278-0062
The effectiveness of a cardiovascular magnetic resonance (CMR) scan depends on the ability of the operator to correctly tune the acquisition parameters to the subject being scanned and on the potential occurrence of imaging artefacts such as cardiac and respiratory motion. In the clinical practice, a quality control step is performed by visual assessment of the acquired images: however, this procedure is strongly operatordependent, cumbersome and sometimes incompatible with the time constraints in clinical settings and large-scale studies. We propose a fast, fully-automated, learning-based quality control pipeline for CMR images, specifically for short-axis image stacks. Our pipeline performs three important quality checks: 1) heart coverage estimation, 2) inter-slice motion detection, 3) image contrast estimation in the cardiac region. The pipeline uses a hybrid decision forest method - integrating both regression and structured classification models - to extract landmarks as well as probabilistic segmentation maps from both long- and short-axis images as a basis to perform the quality checks. The technique was tested on up to 3000 cases from the UK Biobank as well as on 100 cases from the UK Digital Heart Project, and validated against manual annotations and visual inspections performed by expert interpreters. The results show the capability of the proposed pipeline to correctly detect incomplete or corrupted scans (e.g. on UK Biobank, sensitivity and specificity respectively 88% and 99% for heart coverage estimation, 85% and 95% for motion detection), allowing their exclusion from the analysed dataset or the triggering of a new acquisition.
Alansary A, Oktay O, Li Y, et al., 2019, Evaluating reinforcement learning agents for anatomical landmark detection, Medical Image Analysis, Vol: 53, Pages: 156-164, ISSN: 1361-8415
Automatic detection of anatomical landmarks is an important step for a wide range of applications in medical image analysis. Manual annotation of landmarks is a tedious task and prone to observer errors. In this paper, we evaluate novel deep reinforcement learning (RL) strategies to train agents that can precisely and robustly localize target landmarks in medical scans. An artificial RL agent learns to identify the optimal path to the landmark by interacting with an environment, in our case 3D images. Furthermore, we investigate the use of fixed- and multi-scale search strategies with novel hierarchical action steps in a coarse-to-fine manner. Several deep Q-network (DQN) architectures are evaluated for detecting multiple landmarks using three different medical imaging datasets: fetal head ultrasound (US), adult brain and cardiac magnetic resonance imaging (MRI). The performance of our agents surpasses state-of-the-art supervised and RL methods. Our experiments also show that multi-scale search strategies perform significantly better than fixed-scale agents in images with large field of view and noisy background such as in cardiac MRI. Moreover, the novel hierarchical steps can significantly speed up the searching process by a factor of 4-5 times.
Schlemper J, Oktay O, Schaap M, et al., 2019, Attention gated networks: Learning to leverage salient regions in medical images., Med Image Anal, Vol: 53, Pages: 197-207
We propose a novel attention gate (AG) model for medical image analysis that automatically learns to focus on target structures of varying shapes and sizes. Models trained with AGs implicitly learn to suppress irrelevant regions in an input image while highlighting salient features useful for a specific task. This enables us to eliminate the necessity of using explicit external tissue/organ localisation modules when using convolutional neural networks (CNNs). AGs can be easily integrated into standard CNN models such as VGG or U-Net architectures with minimal computational overhead while increasing the model sensitivity and prediction accuracy. The proposed AG models are evaluated on a variety of tasks, including medical image classification and segmentation. For classification, we demonstrate the use case of AGs in scan plane detection for fetal ultrasound screening. We show that the proposed attention mechanism can provide efficient object localisation while improving the overall prediction performance by reducing false positives. For segmentation, the proposed architecture is evaluated on two large 3D CT abdominal datasets with manual annotations for multiple organs. Experimental results show that AG models consistently improve the prediction performance of the base architectures across different datasets and training sizes while preserving computational efficiency. Moreover, AGs guide the model activations to be focused around salient regions, which provides better insights into how model predictions are made. The source code for the proposed AG models is publicly available.
Lee M, Petersen K, Pawlowski N, et al., 2019, TeTrIS: template transformer networks for image segmentation with shape priors, IEEE Transactions on Medical Imaging, ISSN: 0278-0062
In this paper we introduce and compare different approaches for incorporating shape prior information into neural network based image segmentation. Specifically, we introduce the concept of template transformer networks where a shape template is deformed to match the underlying structure of interest through an end-to-end trained spatial transformer network. This has the advantage of explicitly enforcing shape priors and is free of discretisation artefacts by providing a soft partial volume segmentation. We also introduce a simple yet effective way of incorporating priors in state-of-the-art pixel-wise binary classification methods such as fully convolutional networks and U-net. Here, the template shape is given as an additional input channel, incorporating this information significantly reduces false positives. We report results on synthetic data and sub-voxel segmentation of coronary lumen structures in cardiac computed tomography showing the benefit of incorporating priors in neural network based image segmentation.
Bakas S, Reyes M, Jakab A, et al., 2019, Identifying the best machine learning algorithms for brain tumorsegmentation, progression assessment, and overall survival prediction in the BRATS challenge
Gliomas are the most common primary brain malignancies, with differentdegrees of aggressiveness, variable prognosis and various heterogeneoushistologic sub-regions, i.e., peritumoral edematous/invaded tissue, necroticcore, active and non-enhancing core. This intrinsic heterogeneity is alsoportrayed in their radio-phenotype, as their sub-regions are depicted byvarying intensity profiles disseminated across multi-parametric magneticresonance imaging (mpMRI) scans, reflecting varying biological properties.Their heterogeneous shape, extent, and location are some of the factors thatmake these tumors difficult to resect, and in some cases inoperable. The amountof resected tumor is a factor also considered in longitudinal scans, whenevaluating the apparent tumor for potential diagnosis of progression.Furthermore, there is mounting evidence that accurate segmentation of thevarious tumor sub-regions can offer the basis for quantitative image analysistowards prediction of patient overall survival. This study assesses thestate-of-the-art machine learning (ML) methods used for brain tumor imageanalysis in mpMRI scans, during the last seven instances of the InternationalBrain Tumor Segmentation (BraTS) challenge, i.e., 2012-2018. Specifically, wefocus on i) evaluating segmentations of the various glioma sub-regions inpre-operative mpMRI scans, ii) assessing potential tumor progression by virtueof longitudinal growth of tumor sub-regions, beyond use of the RECIST/RANOcriteria, and iii) predicting the overall survival from pre-operative mpMRIscans of patients that underwent gross total resection. Finally, we investigatethe challenge of identifying the best ML algorithms for each of these tasks,considering that apart from being diverse on each instance of the challenge,the multi-institutional mpMRI BraTS dataset has also been a continuouslyevolving/growing dataset.
Robinson R, Valindria VV, Bai W, et al., 2019, Automated quality control in image segmentation: application to the UK Biobank cardiac MR imaging study, Journal of Cardiovascular Magnetic Resonance, Vol: 21, ISSN: 1097-6647
Background: The trend towards large-scale studies including population imaging poses new challenges in terms of quality control (QC). This is a particular issue when automatic processing tools, e.g. image segmentation methods, are employed to derive quantitative measures or biomarkers for later analyses. Manual inspection and visual QC of each segmentation isn't feasible at large scale. However, it's important to be able to automatically detect when a segmentation method fails so as to avoid inclusion of wrong measurements into subsequent analyses which could lead to incorrect conclusions. Methods: To overcome this challenge, we explore an approach for predicting segmentation quality based on Reverse Classification Accuracy, which enables us to discriminate between successful and failed segmentations on a per-cases basis. We validate this approach on a new, large-scale manually-annotated set of 4,800 cardiac magnetic resonance scans. We then apply our method to a large cohort of 7,250 cardiac MRI on which we have performed manual QC. Results: We report results used for predicting segmentation quality metrics including Dice Similarity Coefficient (DSC) and surface-distance measures. As initial validation, we present data for 400 scans demonstrating 99% accuracy for classifying low and high quality segmentations using predicted DSC scores. As further validation we show high correlation between real and predicted scores and 95% classification accuracy on 4,800 scans for which manual segmentations were available. We mimic real-world application of the method on 7,250 cardiac MRI where we show good agreement between predicted quality metrics and manual visual QC scores. Conclusions: We show that RCA has the potential for accurate and fully automatic segmentation QC on a per-case basis in the context of large-scale population imaging as in the UK Biobank Imaging Study.
Baweja C, Glocker B, Kamnitsas K, 2018, Towards continual learning in medical imaging, Medical imaging meets NIPS
This work investigates continual learning of two segmentation tasks in brain MRIwith neural networks. To explore in this context the capabilities of current methodsfor countering catastrophic forgetting of the first task when a new one is learned,we investigateelastic weight consolidation, a recently proposed method basedon Fisher information, originally evaluated on reinforcement learning of Atarigames. We use it to sequentially learn segmentation of normal brain structures andthen segmentation of white matter lesions. Our findings show this recent methodreduces catastrophic forgetting, while large room for improvement exists in thesechallenging settings for continual learning.
Mitchell J, Kamnitsas K, Singleton K, et al., 2018, DEEP LEARNING FOR ACCURATE, RAPID, FULLY AUTOMATIC MEASUREMENT OF BRAIN TUMOR-ASSOCIATED ABNORMALITY SEEN ON MRI, 23rd Annual Scientific Meeting and Education Day of the Society-for-Neuro-Oncology (SNO) / 3rd CNS Anticancer Drug Discovery and Development Conference, Publisher: OXFORD UNIV PRESS INC, Pages: 180-181, ISSN: 1522-8517
Singleton K, Mitchell J, Ranjbar S, et al., 2018, DEEP LEARNING DETECTS DIFFERENCES IN THE MRIs OF MALE AND FEMALE GLIOMAS, 23rd Annual Scientific Meeting and Education Day of the Society-for-Neuro-Oncology (SNO) / 3rd CNS Anticancer Drug Discovery and Development Conference, Publisher: OXFORD UNIV PRESS INC, Pages: 177-177, ISSN: 1522-8517
Hou B, Miolane N, Khanal B, et al., 2018, Computing CNN loss and gradients for pose estimation with Riemannian geometry, International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Publisher: Springer Verlag, ISSN: 0302-9743
Pose estimation, i.e. predicting a 3D rigid transformation with respect to a fixed co-ordinate frame in, SE(3), is an omnipresent problem in medical image analysis. Deep learning methods often parameterise poses with a representation that separates rotation and translation.As commonly available frameworks do not provide means to calculate loss on a manifold, regression is usually performed using the L2-norm independently on the rotation’s and the translation’s parameterisations. This is a metric for linear spaces that does not take into account the Lie group structure of SE(3). In this paper, we propose a general Riemannian formulation of the pose estimation problem, and train CNNs directly on SE(3) equipped with a left-invariant Riemannian metric. The loss between the ground truth and predicted pose (elements of the manifold) is calculated as the Riemannian geodesic distance, which couples together the translation and rotation components. Network weights are updated by back-propagating the gradient with respect to the predicted pose on the tangent space of the manifold SE(3). We thoroughly evaluate the effectiveness of our loss function by comparing its performance with popular and most commonly used existing methods, on tasks such as image-based localisation and intensity-based 2D/3D registration. We also show that hyper-parameters, used in our loss function to weight the contribution between rotations andtranslations, can be intrinsically calculated from the dataset to achievegreater performance margins.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.