Richard McKinley

ISLES 2015 - A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI

&NA; Ischemic stroke is the most common cerebrovascular disease, and its diagnosis, treatment, and study relies on non‐invasive imaging. Algorithms for stroke lesion segmentation from magnetic resonance imaging (MRI) volumes are intensely researched, but the reported results are largely incomparable due to different datasets and evaluation schemes. We approached this urgent problem of comparability with the Ischemic Stroke Lesion Segmentation (ISLES) challenge organized in conjunction with the MICCAI 2015 conference. In this paper we propose a common evaluation framework, describe the publicly available datasets, and present the results of the two sub‐challenges: Sub‐Acute Stroke Lesion Segmentation (SISS) and Stroke Perfusion Estimation (SPES). A total of 16 research groups participated with a wide range of state‐of‐the‐art automatic segmentation algorithms. A thorough analysis of the obtained data enables a critical evaluation of the current state‐of‐the‐art, recommendations for further developments, and the identification of remaining challenges. The segmentation of acute perfusion lesions addressed in SPES was found to be feasible. However, algorithms applied to sub‐acute lesion segmentation in SISS still lack accuracy. Overall, no algorithmic characteristic of any method was found to perform superior to the others. Instead, the characteristics of stroke lesion appearances, their evolution, and the observed challenges should be studied in detail. The annotated ISLES image datasets continue to be publicly available through an online evaluation system to serve as an ongoing benchmarking resource (www.isles‐challenge.org). HighlightsEvaluation framework for automatic stroke lesion segmentation from MRIPublic multi‐center, multi‐vendor, multi‐protocol databases releasedOngoing fair and automated benchmark with expert created ground truth setsComparison of 14+7 groups who responded to an open challenge in MICCAISegmentation feasible in acute and unsolved in sub‐acute cases Graphical abstract Figure. No caption available.

Proof Nets for Herbrand’s Theorem

This article explores Herbrand’s theorem as the source of a natural notion of abstract proof object for classical logic, embodying the “essence” of a sequent calculus proof. We see how to view a calculus of abstract Herbrand proofs (Herbrand nets) as an analytic proof system with syntactic cut-elimination. Herbrand nets can also be seen as a natural generalization of Miller’s expansion tree proofs to a setting including cut. We demonstrate sequentialization of Herbrand nets into a sequent calculus LKH; each net corresponds to an equivalence class of LKH proofs under natural proof transformations. A surprising property of our cut-reduction algorithm is that it is non-confluent despite not supporting the usual examples of non-confluent reduction in classical logic.

An Algorithmic Interpretation of a Deep Inference System

We set out to find something that corresponds to deep inference in the same way that the lambda-calculus corresponds to natural deduction. Starting from natural deduction for the conjunction-implication fragment of intuitionistic logic we design a corresponding deep inference system together with reduction rules on proofs that allow a fine-grained simulation of beta-reduction.

Fully automated stroke tissue estimation using random forest classifiers (FASTER)

Several clinical trials have recently proven the efficacy of mechanical thrombectomy for treating ischemic stroke, within a six-hour window for therapy. To move beyond treatment windows and toward personalized risk assessment, it is essential to accurately identify the extent of tissue-at-risk (“penumbra”). We introduce a fully automated method to estimate the penumbra volume using multimodal MRI (diffusion-weighted imaging, a T2w- and T1w contrast-enhanced sequence, and dynamic susceptibility contrast perfusion MRI). The method estimates tissue-at-risk by predicting tissue damage in the case of both persistent occlusion and of complete recanalization. When applied to 19 test cases with a thrombolysis in cerebral infarction grading of 1–2a, mean overestimation of final lesion volume was 30 ml, compared with 121 ml for manually corrected thresholding. Predicted tissue-at-risk volume was positively correlated with final lesion volume (p < 0.05). We conclude that prediction of tissue damage in the event of either persistent occlusion or immediate and complete recanalization, from spatial features derived from MRI, provides a substantial improvement beyond predefined thresholds. It may serve as an alternative method for identifying tissue-at-risk that may aid in treatment selection in ischemic stroke.

Automatic quality control in clinical (1) H MRSI of brain cancer.

MRSI grids frequently show spectra with poor quality, mainly because of the high sensitivity of MRS to field inhomogeneities. These poor quality spectra are prone to quantification and/or interpretation errors that can have a significant impact on the clinical use of spectroscopic data. Therefore, quality control of the spectra should always precede their clinical use. When performed manually, quality assessment of MRSI spectra is not only a tedious and time‐consuming task, but is also affected by human subjectivity. Consequently, automatic, fast and reliable methods for spectral quality assessment are of utmost interest. In this article, we present a new random forest‐based method for automatic quality assessment of 1H MRSI brain spectra, which uses a new set of MRS signal features. The random forest classifier was trained on spectra from 40 MRSI grids that were classified as acceptable or non‐acceptable by two expert spectroscopists. To account for the effects of intra‐rater reliability, each spectrum was rated for quality three times by each rater. The automatic method classified these spectra with an area under the curve (AUC) of 0.976. Furthermore, in the subset of spectra containing only the cases that were classified every time in the same way by the spectroscopists, an AUC of 0.998 was obtained. Feature importance for the classification was also evaluated. Frequency domain skewness and kurtosis, as well as time domain signal‐to‐noise ratios (SNRs) in the ranges 50–75 ms and 75–100 ms, were the most important features. Given that the method is able to assess a whole MRSI grid faster than a spectroscopist (approximately 3 s versus approximately 3 min), and without loss of accuracy (agreement between classifier trained with just one session and any of the other labelling sessions, 89.88%; agreement between any two labelling sessions, 89.03%), the authors suggest its implementation in the clinical routine. The method presented in this article was implemented in jMRUIs SpectrIm plugin. Copyright

Enhancing interpretability of automatically extracted machine learning features: application to a RBM-Random Forest system on brain lesion segmentation.

HighlightsWe propose methodologies to enhance the interpretability of a machine learning system.The approach can yield two levels of interpretability (global and local), allowing us to assess how the system learned task‐specific relations and its individual predictions.Validation on brain tumor segmentation and penumbra estimation in acute stroke.Based on the evaluated clinical scenarios, the proposed approach allows us to confirm that the machine learning system learns relations coherent with expert knowledge and annotation protocols. Graphical abstract Figure. No Caption available. Abstract Machine learning systems are achieving better performances at the cost of becoming increasingly complex. However, because of that, they become less interpretable, which may cause some distrust by the end‐user of the system. This is especially important as these systems are pervasively being introduced to critical domains, such as the medical field. Representation Learning techniques are general methods for automatic feature computation. Nevertheless, these techniques are regarded as uninterpretable “black boxes”. In this paper, we propose a methodology to enhance the interpretability of automatically extracted machine learning features. The proposed system is composed of a Restricted Boltzmann Machine for unsupervised feature learning, and a Random Forest classifier, which are combined to jointly consider existing correlations between imaging data, features, and target variables. We define two levels of interpretation: global and local. The former is devoted to understanding if the system learned the relevant relations in the data correctly, while the later is focused on predictions performed on a voxel‐ and patient‐level. In addition, we propose a novel feature importance strategy that considers both imaging data and target variables, and we demonstrate the ability of the approach to leverage the interpretability of the obtained representation for the task at hand. We evaluated the proposed methodology in brain tumor segmentation and penumbra estimation in ischemic stroke lesions. We show the ability of the proposed methodology to unveil information regarding relationships between imaging modalities and extracted features and their usefulness for the task at hand. In both clinical scenarios, we demonstrate that the proposed methodology enhances the interpretability of automatically learned features, highlighting specific learning patterns that resemble how an expert extracts relevant data from medical images.

Improving labeling efficiency in automatic quality control of MRSI data

To improve the efficiency of the labeling task in automatic quality control of MR spectroscopy imaging data.

Nabla-net: A Deep Dag-Like Convolutional Architecture for Biomedical Image Segmentation

Biomedical image segmentation requires both voxel-level information and global context. We report on a deep convolutional architecture which combines a fully-convolutional network for local features and an encoder-decoder network in which convolutional layers and maxpooling compute high-level features, which are then upsampled to the resolution of the initial image using further convolutional layers and tied unpooling. We apply the method to segmenting multiple sclerosis lesions and gliomas.

Objective Evaluation of Multiple Sclerosis Lesion Segmentation using a Data Management and Processing Infrastructure

We present a study of multiple sclerosis segmentation algorithms conducted at the international MICCAI 2016 challenge. This challenge was operated using a new open-science computing infrastructure. This allowed for the automatic and independent evaluation of a large range of algorithms in a fair and completely automatic manner. This computing infrastructure was used to evaluate thirteen methods of MS lesions segmentation, exploring a broad range of state-of-theart algorithms, against a high-quality database of 53 MS cases coming from four centers following a common definition of the acquisition protocol. Each case was annotated manually by an unprecedented number of seven different experts. Results of the challenge highlighted that automatic algorithms, including the recent machine learning methods (random forests, deep learning, …), are still trailing human expertise on both detection and delineation criteria. In addition, we demonstrate that computing a statistically robust consensus of the algorithms performs closer to human expertise on one score (segmentation) although still trailing on detection scores.

Relating Acute Lesion Loads to Chronic Outcome in Ischemic Stroke – An Exploratory Comparison of Mismatch Patterns and Predictive Modelling

Objectives: To investigate the relationship between imaging features derived from lesion loads and 3 month clinical assessments in ischemic stroke patients. To support clinically implementable predictive modeling with information from lesion-load features. Methods: A retrospective cohort of ischemic stroke patients was studied. The dataset was dichotomized based on revascularization treatment outcome (TICI score). Three lesion delineations were derived from magnetic resonance imaging in each group: two clinically implementable (threshold based and fully automatic prediction) and 90-day follow-up as final groundtruth. Lesion load imaging features were created through overlay of the lesion delineations on a histological brain atlas, and were correlated with the clinical assessment (NIHSS). Significance of the correlations was assessed by constructing confidence intervals using bootstrap sampling. Results: Overall, high correlations between lesion loads and clinical score were observed (up to 0.859). Delineations derived from acute imaging yielded on average somewhat lower correlations than delineations derived from 90-day follow-up imaging. Correlations suggest that both total lesion volume and corticospinal tract lesion load are associated with functional outcome, and in addition highlight other potential areas associated with poor clinical outcome, including the primary somatosensory cortex BA3a. Fully automatic prediction was comparable to ADC threshold-based delineation on the successfully treated cohort and superior to the Tmax threshold-based delineation in the unsuccessfully treated cohort. Conclusions: The confirmation of established predictors for stroke outcome (e.g., corticospinal tract integrity and total lesion volume) gives support to the proposed methodology—relating acute lesion loads to 3 month outcome assessments by way of correlation. Furthermore, the preliminary results indicate an association of further brain regions and structures with three month NIHSS outcome assessments. Hence, prediction models might observe an increased accuracy when incorporating regional (instead of global) lesion loads. Also, the results lend support to the clinical utilization of the automatically predicted volumes from FASTER, rather than the simpler DWI and PWI lesion delineations.