Search

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., Comment: 14 pages, 7 figures, Journal of Cardiovascular Magnetic Resonance

The variations in multi-center data in medical imaging studies have brought the necessity of domain adaptation. Despite the advancement of machine learning in automatic segmentation, performance often degrades when algorithms are applied on new data acquired from different scanners or sequences than the training data. Manual annotation is costly and time consuming if it has to be carried out for every new target domain. In this work, we investigate automatic selection of suitable subjects to be annotated for supervised domain adaptation using the concept of reverse classification accuracy (RCA). RCA predicts the performance of a trained model on data from the new domain and different strategies of selecting subjects to be included in the adaptation via transfer learning are evaluated. We perform experiments on a two-center MR database for the task of organ segmentation. We show that subject selection via RCA can reduce the burden of annotation of new data for the target domain., Comment: Accepted at the International Conference on Medical Imaging with Deep Learning (MIDL) 2018

Advances in machine learning techniques have been shown to bring benefit for analysing medical images. Whole-body scans contain multiple organs, which makes the manual annotation harder, hence the amount of data obtained is still limited. Meanwhile, machine learning needs a large amount of annotated data in order to perform well. Image segmentation is a crucial task in medical image analysis and beneficial for clinical diagnosis. One of the segmentation challenges is to infer the automatic segmentation quality between algorithms. Unfortunately, the manual annotations required to compute the segmentation accuracy are sometimes unavailable. In order to tackle this, we propose a new method to predict the segmentation evaluation on a per-case basis without annotated data. Segmenting scans acquired from different imaging centers than the one used for algorithm training, can lead to degraded segmentation results. This problem has brought the necessity of domain adaptation. We studied how to select useful samples from unlabeled target domain for supervised domain adaptation. This framework is shown to be less time-consuming with promising results to be useful in clinical practice. A novel dual-stream network is also introduced to learn useful features on unpaired images from different modalities. This network demonstrates its power in multi-modal learning to segment multiple abdominal organs on both modalities. Additionally, strategies to segment challenging small organs on whole-body scans are investigated. We show that using two-stage networks with weighting schemes is useful to solve class imbalance and multi-scale contextual information. Machine learning is not only helping the work of radiologists with faster automatic segmentation tools, but also leveraging the useful features for alleviating real clinical problems. We believe that our proposed methods could offer an important contribution to the field of medical image analysis; advancing our understanding on how machine-learning approaches could be applied in several problems in whole-body image analysis. Open Access

When integrating computational tools such as automatic segmentation into clinical practice, it is of utmost importance to be able to assess the level of accuracy on new data, and in particular, to detect when an automatic method fails. However, this is difficult to achieve due to absence of ground truth. Segmentation accuracy on clinical data might be different from what is found through cross-validation because validation data is often used during incremental method development, which can lead to overfitting and unrealistic performance expectations. Before deployment, performance is quantified using different metrics, for which the predicted segmentation is compared to a reference segmentation, often obtained manually by an expert. But little is known about the real performance after deployment when a reference is unavailable. In this paper, we introduce the concept of reverse classification accuracy (RCA) as a framework for predicting the performance of a segmentation method on new data. In RCA we take the predicted segmentation from a new image to train a reverse classifier which is evaluated on a set of reference images with available ground truth. The hypothesis is that if the predicted segmentation is of good quality, then the reverse classifier will perform well on at least some of the reference images. We validate our approach on multi-organ segmentation with different classifiers and segmentation methods. Our results indicate that it is indeed possible to predict the quality of individual segmentations, in the absence of ground truth. Thus, RCA is ideal for integration into automatic processing pipelines in clinical routine and as part of large-scale image analysis studies., Comment: Accepted article to appear in IEEE Transactions on Medical Imaging 2017

International audience; Delay Enhancement - Magnetic Resonance Imaging (DE-MRI) can be considered as the gold standard for the assessment of myocardial viability after myocardial infarction (MI). Around the infarcted areas that appear with hyper-enhanced signal, there is a peri-infarct border zone that may be an important arrhythmogenic substrate. The extent of this area is an independent predictor of post-MI mortality (1). However, it is difficult to separate peri-infarct border zones from infarct core or normal areas, because of their intermediate signal intensity. We propose a new automatic approach to detect this area, based on a modified Gaussian Mixture Model (GMM) and spatial-weighted fuzzy clustering.