Quality assurance of late gadolinium enhancement cardiac magnetic resonance images: a deep learning classifier for confidence in the presence or absence of abnormality with potential to prompt real-time image optimization.

BACKGROUND: Late gadolinium enhancement (LGE) of the myocardium has significant diagnostic and prognostic implications, with even small areas of enhancement being important. Distinguishing between definitely normal and definitely abnormal LGE images is usually straightforward, but diagnostic uncertainty arises when reporters are not sure whether the observed LGE is genuine or not. This uncertainty might be resolved by repetition (to remove artifact) or further acquisition of intersecting images, but this must take place before the scan finishes. Real-time quality assurance by humans is a complex task requiring training and experience, so being able to identify which images have an intermediate likelihood of LGE while the scan is ongoing, without the presence of an expert is of high value. This decision-support could prompt immediate image optimization or acquisition of supplementary images to confirm or refute the presence of genuine LGE. This could reduce ambiguity in reports. METHODS: Short-axis, phase-sensitive inversion recovery late gadolinium images were extracted from our clinical cardiac magnetic resonance (CMR) database and shuffled. Two, independent, blinded experts scored each individual slice for "LGE likelihood" on a visual analog scale, from 0 (absolute certainty of no LGE) to 100 (absolute certainty of LGE), with 50 representing clinical equipoise. The scored images were split into two classes-either "high certainty" of whether LGE was present or not, or "low certainty." The dataset was split into training, validation, and test sets (70:15:15). A deep learning binary classifier based on the EfficientNetV2 convolutional neural network architecture was trained to distinguish between these categories. Classifier performance on the test set was evaluated by calculating the accuracy, precision, recall, F1-score, and area under the receiver operating characteristics curve (ROC AUC). Performance was also evaluated on an external test set of images from a different center. RESULTS: One thousand six hundred and forty-five images (from 272 patients) were labeled and split at the patient level into training (1151 images), validation (247 images), and test (247 images) sets for the deep learning binary classifier. Of these, 1208 images were "high certainty" (255 for LGE, 953 for no LGE), and 437 were "low certainty". An external test comprising 247 images from 41 patients from another center was also employed. After 100 epochs, the performance on the internal test set was accuracy = 0.94, recall = 0.80, precision = 0.97, F1-score = 0.87, and ROC AUC = 0.94. The classifier also performed robustly on the external test set (accuracy = 0.91, recall = 0.73, precision = 0.93, F1-score = 0.82, and ROC AUC = 0.91). These results were benchmarked against a reference inter-expert accuracy of 0.86. CONCLUSION: Deep learning shows potential to automate quality control of late gadolinium imaging in CMR. The ability to identify short-axis images with intermediate LGE likelihood in real-time may serve as a useful decision-support tool. This approach has the potential to guide immediate further imaging while the patient is still in the scanner, thereby reducing the frequency of recalls and inconclusive reports due to diagnostic indecision.

Aortic stenosis assessment from the 3-chamber cine: Ratio of balanced steady-state-free-precession (bSSFP) blood signal between the aorta and left ventricle predicts severity.

BACKGROUND: Cardiovascular magnetic resonance (CMR) imaging is an important tool for evaluating the severity of aortic stenosis (AS), co-existing aortic disease, and concurrent myocardial abnormalities. Acquiring this additional information requires protocol adaptations and additional scanner time, but is not necessary for the majority of patients who do not have AS. We observed that the relative signal intensity of blood in the ascending aorta on a balanced steady state free precession (bSSFP) 3-chamber cine was often reduced in those with significant aortic stenosis. We investigated whether this effect could be quantified and used to predict AS severity in comparison to existing gold-standard measurements. METHODS: Multi-centre, multi-vendor retrospective analysis of patients with AS undergoing CMR and transthoracic echocardiography (TTE). Blood signal intensity was measured in a ∼1 cm2 region of interest (ROI) in the aorta and left ventricle (LV) in the 3-chamber bSSFP cine. Because signal intensity varied across patients and scanner vendors, a ratio of the mean signal intensity in the aorta ROI to the LV ROI (Ao:LV) was used. This ratio was compared using Pearson correlations against TTE parameters of AS severity: aortic valve peak velocity, mean pressure gradient and the dimensionless index. The study also assessed whether field strength (1.5 T vs. 3 T) and patient characteristics (presence of bicuspid aortic valves (BAV), dilated aortic root and low flow states) altered this signal relationship. RESULTS: 314 patients (median age 69 [IQR 57-77], 64% male) who had undergone both CMR and TTE were studied; 84 had severe AS, 78 had moderate AS, 66 had mild AS and 86 without AS were studied as a comparator group. The median time between CMR and TTE was 12 weeks (IQR 4-26). The Ao:LV ratio at 1.5 T strongly correlated with peak velocity (r = -0.796, p = 0.001), peak gradient (r = -0.772, p = 0.001) and dimensionless index (r = 0.743, p = 0.001). An Ao:LV ratio of < 0.86 was 84% sensitive and 82% specific for detecting AS of any severity and a ratio of 0.58 was 83% sensitive and 92% specific for severe AS. The ability of Ao:LV ratio to predict AS severity remained for patients with bicuspid aortic valves, dilated aortic root or low indexed stroke volume. The relationship between Ao:LV ratio and AS severity was weaker at 3 T. CONCLUSIONS: The Ao:LV ratio, derived from bSSFP 3-chamber cine images, shows a good correlation with existing measures of AS severity. It demonstrates utility at 1.5 T and offers an easily calculable metric that can be used at the time of scanning or automated to identify on an adaptive basis which patients benefit from dedicated imaging to assess which patients should have additional sequences to assess AS.

Efficient labelling for efficient deep learning: the benefit of a multiple-image-ranking method to generate high volume training data applied to ventricular slice level classification in cardiac MRI.

Background: Getting the most value from expert clinicians' limited labelling time is a major challenge for artificial intelligence (AI) development in clinical imaging. We present a novel method for ground-truth labelling of cardiac magnetic resonance imaging (CMR) image data by leveraging multiple clinician experts ranking multiple images on a single ordinal axis, rather than manual labelling of one image at a time. We apply this strategy to train a deep learning (DL) model to classify the anatomical position of CMR images. This allows the automated removal of slices that do not contain the left ventricular (LV) myocardium. Methods: Anonymised LV short-axis slices from 300 random scans (3,552 individual images) were extracted. Each image's anatomical position relative to the LV was labelled using two different strategies performed for 5 hours each: (I) 'one-image-at-a-time': each image labelled according to its position: 'too basal', 'LV', or 'too apical' individually by one of three experts; and (II) 'multiple-image-ranking': three independent experts ordered slices according to their relative position from 'most-basal' to 'most apical' in batches of eight until each image had been viewed at least 3 times. Two convolutional neural networks were trained for a three-way classification task (each model using data from one labelling strategy). The models' performance was evaluated by accuracy, F1-score, and area under the receiver operating characteristics curve (ROC AUC). Results: After excluding images with artefact, 3,323 images were labelled by both strategies. The model trained using labels from the 'multiple-image-ranking strategy' performed better than the model using the 'one-image-at-a-time' labelling strategy (accuracy 86% vs. 72%, P=0.02; F1-score 0.86 vs. 0.75; ROC AUC 0.95 vs. 0.86). For expert clinicians performing this task manually the intra-observer variability was low (Cohen's κ=0.90), but the inter-observer variability was higher (Cohen's κ=0.77). Conclusions: We present proof of concept that, given the same clinician labelling effort, comparing multiple images side-by-side using a 'multiple-image-ranking' strategy achieves ground truth labels for DL more accurately than by classifying images individually. We demonstrate a potential clinical application: the automatic removal of unrequired CMR images. This leads to increased efficiency by focussing human and machine attention on images which are needed to answer clinical questions.

TMS-Net: A segmentation network coupled with a run-time quality control method for robust cardiac image segmentation.

Recently, deep networks have shown impressive performance for the segmentation of cardiac Magnetic Resonance Imaging (MRI) images. However, their achievement is proving slow to transition to widespread use in medical clinics because of robustness issues leading to low trust of clinicians to their results. Predicting run-time quality of segmentation masks can be useful to warn clinicians against poor results. Despite its importance, there are few studies on this problem. To address this gap, we propose a quality control method based on the agreement across decoders of a multi-view network, TMS-Net, measured by the cosine similarity. The network takes three view inputs resliced from the same 3D image along different axes. Different from previous multi-view networks, TMS-Net has a single encoder and three decoders, leading to better noise robustness, segmentation performance and run-time quality estimation in our experiments on the segmentation of the left atrium on STACOM 2013 and STACOM 2018 challenge datasets. We also present a way to generate poor segmentation masks by using noisy images generated with engineered noise and Rician noise to simulate undertraining, high anisotropy and poor imaging settings problems. Our run-time quality estimation method show a good classification of poor and good quality segmentation masks with an AUC reaching to 0.97 on STACOM 2018. We believe that TMS-Net and our run-time quality estimation method has a high potential to increase the thrust of clinicians to automatic image analysis tools.

Validity of Self-testing at Home With Rapid Severe Acute Respiratory Syndrome Coronavirus 2 Antibody Detection by Lateral Flow Immunoassay.

BACKGROUND: We explore severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody lateral flow immunoassay (LFIA) performance under field conditions compared to laboratory-based electrochemiluminescence immunoassay (ECLIA) and live virus neutralization. METHODS: In July 2021, 3758 participants performed, at home, a self-administered Fortress LFIA on finger-prick blood, reported and submitted a photograph of the result, and provided a self-collected capillary blood sample for assessment of immunoglobulin G (IgG) antibodies using the Roche Elecsys® Anti-SARS-CoV-2 ECLIA. We compared the self-reported LFIA result to the quantitative ECLIA and checked the reading of the LFIA result with an automated image analysis (ALFA). In a subsample of 250 participants, we compared the results to live virus neutralization. RESULTS: Almost all participants (3593/3758, 95.6%) had been vaccinated or reported prior infection. Overall, 2777/3758 (73.9%) were positive on self-reported LFIA, 2811/3457 (81.3%) positive by LFIA when ALFA-reported, and 3622/3758 (96.4%) positive on ECLIA (using the manufacturer reference standard threshold for positivity of 0.8 U mL-1). Live virus neutralization was detected in 169 of 250 randomly selected samples (67.6%); 133/169 were positive with self-reported LFIA (sensitivity 78.7%; 95% confidence interval [CI]: 71.8, 84.6), 142/155 (91.6%; 95% CI: 86.1, 95.5) with ALFA, and 169 (100%; 95% CI: 97.8, 100.0) with ECLIA. There were 81 samples with no detectable virus neutralization; 47/81 were negative with self-reported LFIA (specificity 58.0%; 95% CI: 46.5, 68.9), 34/75 (45.3%; 95% CI: 33.8, 57.3) with ALFA, and 0/81 (0%; 95% CI: 0, 4.5) with ECLIA. CONCLUSIONS: Self-administered LFIA is less sensitive than a quantitative antibody test, but the positivity in LFIA correlates better than the quantitative ECLIA with virus neutralization.

Validation of two-dimensional video-based inference of finger kinematics with pose estimation.

Accurate capture finger of movements for biomechanical assessments has typically been achieved within laboratory environments through the use of physical markers attached to a participant's hands. However, such requirements can narrow the broader adoption of movement tracking for kinematic assessment outside these laboratory settings, such as in the home. Thus, there is the need for markerless hand motion capture techniques that are easy to use and accurate enough to evaluate the complex movements of the human hand. Several recent studies have validated lower-limb kinematics obtained with a marker-free technique, OpenPose. This investigation examines the accuracy of OpenPose, when applied to images from single RGB cameras, against a 'gold standard' marker-based optical motion capture system that is commonly used for hand kinematics estimation. Participants completed four single-handed activities with right and left hands, including hand abduction and adduction, radial walking, metacarpophalangeal (MCP) joint flexion, and thumb opposition. The accuracy of finger kinematics was assessed using the root mean square error. Mean total active flexion was compared using the Bland-Altman approach, and the coefficient of determination of linear regression. Results showed good agreement for abduction and adduction and thumb opposition activities. Lower agreement between the two methods was observed for radial walking (mean difference between the methods of 5.03°) and MCP flexion (mean difference of 6.82°) activities, due to occlusion. This investigation demonstrated that OpenPose, applied to videos captured with monocular cameras, can be used for markerless motion capture for finger tracking with an error below 11° and on the order of that which is accepted clinically.

Machine learning to support visual auditing of home-based lateral flow immunoassay self-test results for SARS-CoV-2 antibodies.

Background: Lateral flow immunoassays (LFIAs) are being used worldwide for COVID-19 mass testing and antibody prevalence studies. Relatively simple to use and low cost, these tests can be self-administered at home, but rely on subjective interpretation of a test line by eye, risking false positives and false negatives. Here, we report on the development of ALFA (Automated Lateral Flow Analysis) to improve reported sensitivity and specificity. Methods: Our computational pipeline uses machine learning, computer vision techniques and signal processing algorithms to analyse images of the Fortress LFIA SARS-CoV-2 antibody self-test, and subsequently classify results as invalid, IgG negative and IgG positive. A large image library of 595,339 participant-submitted test photographs was created as part of the REACT-2 community SARS-CoV-2 antibody prevalence study in England, UK. Alongside ALFA, we developed an analysis toolkit which could also detect device blood leakage issues. Results: Automated analysis showed substantial agreement with human experts (Cohen's kappa 0.90-0.97) and performed consistently better than study participants, particularly for weak positive IgG results. Specificity (98.7-99.4%) and sensitivity (90.1-97.1%) were high compared with visual interpretation by human experts (ranges due to the varying prevalence of weak positive IgG tests in datasets). Conclusions: Given the potential for LFIAs to be used at scale in the COVID-19 response (for both antibody and antigen testing), even a small improvement in the accuracy of the algorithms could impact the lives of millions of people by reducing the risk of false-positive and false-negative result read-outs by members of the public. Our findings support the use of machine learning-enabled automated reading of at-home antibody lateral flow tests as a tool for improved accuracy for population-level community surveillance.

Automatic Diagnosis Labeling of Cardiovascular MRI by Using Semisupervised Natural Language Processing of Text Reports.

PURPOSE: To assess whether the semisupervised natural language processing (NLP) of text from clinical radiology reports could provide useful automated diagnosis categorization for ground truth labeling to overcome manual labeling bottlenecks in the machine learning pipeline. MATERIALS AND METHODS: In this retrospective study, 1503 text cardiac MRI reports from 2016 to 2019 were manually annotated for five diagnoses by clinicians: normal, dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy, myocardial infarction (MI), and myocarditis. A semisupervised method that uses bidirectional encoder representations from transformers (BERT) pretrained on 1.14 million scientific publications was fine-tuned by using the manually extracted labels, with a report dataset split into groups of 801 for training, 302 for validation, and 400 for testing. The model's performance was compared with two traditional NLP models: a rule-based model and a support vector machine (SVM) model. The models' F1 scores and receiver operating characteristic curves were used to analyze performance. RESULTS: After 15 epochs, the F1 scores on the test set of 400 reports were as follows: normal, 84%; DCM, 79%; hypertrophic cardiomyopathy, 86%; MI, 91%; and myocarditis, 86%. The pooled F1 score and area under the receiver operating curve were 86% and 0.96, respectively. On the same test set, the BERT model had a higher performance than the rule-based model (F1 score, 42%) and SVM model (F1 score, 82%). Diagnosis categories classified by using the BERT model performed the labeling of 1000 MR images in 0.2 second. CONCLUSION: The developed model used labels extracted from radiology reports to provide automated diagnosis categorization of MR images with a high level of performance.Keywords: Semisupervised Learning, Diagnosis/Classification/Application Domain, Named Entity Recognition, MRI Supplemental material is available for this article. © RSNA, 2021.

LA-Net: A Multi-Task Deep Network for the Segmentation of the Left Atrium.

Although atrial fibrillation (AF) is the most common sustained atrial arrhythmia, treatment success for this condition remains suboptimal. Information from magnetic resonance imaging (MRI) has the potential to improve treatment efficacy, but there are currently few automatic tools for the segmentation of the atria in MR images. In the study, we propose a LA-Net, a multi-task network optimised to simultaneously generate left atrial segmentation and edge masks from MRI. LA-Net includes cross attention modules (CAMs) and enhanced decoder modules (EDMs) to purposefully select the most meaningful edge information for segmentation and smoothly incorporate it into segmentation masks at multiple-scales. We evaluate the performance of LA-Net on two MR sequences: late gadolinium enhanced (LGE) atrial MRI and atrial short axis balanced steady state free precession (bSSFP) MRI. LA-Net gives Hausdorff distances of 12.43 mm and Dice scores of 0.92 on the LGE (STACOM 2018) dataset and Hausdorff distances of 17.41 mm and Dice scores of 0.90 on the bSSFP (in-house) dataset without any post-processing, surpassing previously proposed segmentation networks, including U-Net and SEGANet. Our method allows automatic extraction of information about the LA from MR images, which can play an important role in the management of AF patients.

Automated Identification of Orthopedic Implants on Radiographs Using Deep Learning.

Accurate identification of metallic orthopedic implant design is important for preoperative planning of revision arthroplasty. Surgical records of implant models are frequently unavailable. The aim of this study was to develop and evaluate a convolutional neural network for identifying orthopedic implant models using radiographs. In this retrospective study, 427 knee and 922 hip unilateral anteroposterior radiographs, including 12 implant models from 650 patients, were collated from an orthopedic center between March 2015 and November 2019 to develop classification networks. A total of 198 images paired with autogenerated image masks were used to develop a U-Net segmentation network to automatically zero-mask around the implants on the radiographs. Classification networks processing original radiographs, and two-channel conjoined original and zero-masked radiographs, were ensembled to provide a consensus prediction. Accuracies of five senior orthopedic specialists assisted by a reference radiographic gallery were compared with network accuracy using McNemar exact test. When evaluated on a balanced unseen dataset of 180 radiographs, the final network achieved a 98.9% accuracy (178 of 180) and 100% top-three accuracy (180 of 180). The network performed superiorly to all five specialists (76.1% [137 of 180] median accuracy and 85.6% [154 of 180] best accuracy; both P < .001), with robustness to scan quality variation and difficult to distinguish implants. A neural network model was developed that outperformed senior orthopedic specialists at identifying implant models on radiographs; real-world application can now be readily realized through training on a broader range of implants and joints, supported by all code and radiographs being made freely available. Supplemental material is available for this article. Keywords: Neural Networks, Skeletal-Appendicular, Knee, Hip, Computer Applications-General (Informatics), Prostheses, Technology Assess-ment, Observer Performance © RSNA, 2021.

Efficient Source Camera Identification with Diversity-Enhanced Patch Selection and Deep Residual Prediction.

Source camera identification has long been a hot topic in the field of image forensics. Besides conventional feature engineering algorithms developed based on studying the traces left upon shooting, several deep-learning-based methods have also emerged recently. However, identification performance is susceptible to image content and is far from satisfactory for small image patches in real demanding applications. In this paper, an efficient patch-level source camera identification method is proposed based on a convolutional neural network. First, in order to obtain improved robustness with reduced training cost, representative patches are selected according to multiple criteria for enhanced diversity in training data. Second, a fine-grained multiscale deep residual prediction module is proposed to reduce the impact of scene content. Finally, a modified VGG network is proposed for source camera identification at brand, model, and instance levels. A more critical patch-level evaluation protocol is also proposed for fair performance comparison. Abundant experimental results show that the proposed method achieves better results as compared with the state-of-the-art algorithms.

EP-PINNs: Cardiac Electrophysiology Characterisation Using Physics-Informed Neural Networks.

Accurately inferring underlying electrophysiological (EP) tissue properties from action potential recordings is expected to be clinically useful in the diagnosis and treatment of arrhythmias such as atrial fibrillation. It is, however, notoriously difficult to perform. We present EP-PINNs (Physics Informed Neural Networks), a novel tool for accurate action potential simulation and EP parameter estimation from sparse amounts of EP data. We demonstrate, using 1D and 2D in silico data, how EP-PINNs are able to reconstruct the spatio-temporal evolution of action potentials, whilst predicting parameters related to action potential duration (APD), excitability and diffusion coefficients. EP-PINNs are additionally able to identify heterogeneities in EP properties, making them potentially useful for the detection of fibrosis and other localised pathology linked to arrhythmias. Finally, we show EP-PINNs effectiveness on biological in vitro preparations, by characterising the effect of anti-arrhythmic drugs on APD using optical mapping data. EP-PINNs are a promising clinical tool for the characterisation and potential treatment guidance of arrhythmias.

Automated analysis and detection of abnormalities in transaxial anatomical cardiovascular magnetic resonance images: a proof of concept study with potential to optimize image acquisition.

The large number of available MRI sequences means patients cannot realistically undergo them all, so the range of sequences to be acquired during a scan are protocolled based on clinical details. Adapting this to unexpected findings identified early on in the scan requires experience and vigilance. We investigated whether deep learning of the images acquired in the first few minutes of a scan could provide an automated early alert of abnormal features. Anatomy sequences from 375 CMR scans were used as a training set. From these, we annotated 1500 individual slices and used these to train a convolutional neural network to perform automatic segmentation of the cardiac chambers, great vessels and any pleural effusions. 200 scans were used as a testing set. The system then assembled a 3D model of the thorax from which it made clinical measurements to identify important abnormalities. The system was successful in segmenting the anatomy slices (Dice 0.910) and identified multiple features which may guide further image acquisition. Diagnostic accuracy was 90.5% and 85.5% for left and right ventricular dilatation, 85% for left ventricular hypertrophy and 94.4% for ascending aorta dilatation. The area under ROC curve for diagnosing pleural effusions was 0.91. We present proof-of-concept that a neural network can segment and derive accurate clinical measurements from a 3D model of the thorax made from transaxial anatomy images acquired in the first few minutes of a scan. This early information could lead to dynamic adaptive scanning protocols, and by focusing scanner time appropriately and prioritizing cases for supervision and early reporting, improve patient experience and efficiency.

Dynamic changes in cell size and corresponding cell fate after optic nerve injury.

Identifying disease-specific patterns of retinal cell loss in pathological conditions has been highlighted by the emergence of techniques such as Detection of Apoptotic Retinal Cells and Adaptive Optics confocal Scanning Laser Ophthalmoscopy which have enabled single-cell visualisation in vivo. Cell size has previously been used to stratify Retinal Ganglion Cell (RGC) populations in histological samples of optic neuropathies, and early work in this field suggested that larger RGCs are more susceptible to early loss than smaller RGCs. More recently, however, it has been proposed that RGC soma and axon size may be dynamic and change in response to injury. To address this unresolved controversy, we applied recent advances in maximising information extraction from RGC populations in retinal whole mounts to evaluate the changes in RGC size distribution over time, using three well-established rodent models of optic nerve injury. In contrast to previous studies based on sampling approaches, we examined the whole Brn3a-positive RGC population at multiple time points over the natural history of these models. The morphology of over 4 million RGCs was thus assessed to glean novel insights from this dataset. RGC subpopulations were found to both increase and decrease in size over time, supporting the notion that RGC cell size is dynamic in response to injury. However, this study presents compelling evidence that smaller RGCs are lost more rapidly than larger RGCs despite the dynamism. Finally, using a bootstrap approach, the data strongly suggests that disease-associated changes in RGC spatial distribution and morphology could have potential as novel diagnostic indicators.

Strain maps of the left atrium imaged with a novel high-resolution CINE MRI protocol.

To date, regional atrial strains have not been imaged in vivo, despite their potential to provide useful clinical information. To address this gap, we present a novel CINE MRI protocol capable of imaging the entire left atrium at an isotropic 2-mm resolution in one single breath-hold. As proof of principle, we acquired data in 10 healthy volunteers and 2 cardiovascular patients using this technique. We also demonstrated how regional atrial strains can be estimated from this data following a manual segmentation of the left atrium using automatic image tracking techniques. The estimated principal strains vary smoothly across the left atrium and have a similar magnitude to estimates reported in the literature.

A Semi-Automatic Method To Segment The Left Atrium in MR Volumes With Varying Slice Numbers.

Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with dramatic increases in mortality and morbidity. Atrial cine MR images are increasingly used in the management of this condition, but there are few specific tools to aid in the segmentation of such data. Some characteristics of atrial cine MR (thick slices, variable number of slices in a volume) preclude the direct use of traditional segmentation tools. When combined with scarcity of labelled data and similarity of the intensity and texture of the left atrium (LA) to other cardiac structures, the segmentation of the LA in CINE MRI becomes a difficult task. To deal with these challenges, we propose a semi-automatic method to segment the left atrium (LA) in MR images, which requires an initial user click per volume. The manually given location information is used to generate a chamber location map to roughly locate the LA, which is then used as an input to a deep network with slightly over 0.5 million parameters. A tracking method is introduced to pass the location information across a volume and to remove unwanted structures in segmentation maps. According to the results of our experiments conducted in an in-house MRI dataset, the proposed method outperforms the U-Net [1] with a margin of 20 mm on Hausdorff distance and 0.17 on Dice score, with limited manual interaction.

Development of a pro-arrhythmic ex vivo intact human and porcine model: cardiac electrophysiological changes associated with cellular uncoupling.

We describe a human and large animal Langendorff experimental apparatus for live electrophysiological studies and measure the electrophysiological changes due to gap junction uncoupling in human and porcine hearts. The resultant ex vivo intact human and porcine model can bridge the translational gap between smaller simple laboratory models and clinical research. In particular, electrophysiological models would benefit from the greater myocardial mass of a large heart due to its effects on far-field signal, electrode contact issues and motion artefacts, consequently more closely mimicking the clinical setting. Porcine (n = 9) and human (n = 4) donor hearts were perfused on a custom-designed Langendorff apparatus. Epicardial electrograms were collected at 16 sites across the left atrium and left ventricle. A total of 1 mM of carbenoxolone was administered at 5 ml/min to induce cellular uncoupling, and then recordings were repeated at the same sites. Changes in electrogram characteristics were analysed. We demonstrate the viability of a controlled ex vivo model of intact porcine and human hearts for electrophysiology with pharmacological modulation. Carbenoxolone reduces cellular coupling and changes contact electrogram features. The time from stimulus artefact to (-dV/dt)max increased between baseline and carbenoxolone (47.9 ± 4.1-67.2 ± 2.7 ms) indicating conduction slowing. The features with the largest percentage change between baseline and carbenoxolone were fractionation + 185.3%, endpoint amplitude - 106.9%, S-endpoint gradient + 54.9%, S point - 39.4%, RS ratio + 38.6% and (-dV/dt)max - 20.9%. The physiological relevance of this methodological tool is that it provides a model to further investigate pharmacologically induced pro-arrhythmic substrates.

Denoising Adversarial Autoencoders.

Unsupervised learning is of growing interest because it unlocks the potential held in vast amounts of unlabeled data to learn useful representations for inference. Autoencoders, a form of generative model, may be trained by learning to reconstruct unlabeled input data from a latent representation space. More robust representations may be produced by an autoencoder if it learns to recover clean input samples from corrupted ones. Representations may be further improved by introducing regularization during training to shape the distribution of the encoded data in the latent space. We suggest denoising adversarial autoencoders (AAEs), which combine denoising and regularization, shaping the distribution of latent space using adversarial training. We introduce a novel analysis that shows how denoising may be incorporated into the training and sampling of AAEs. Experiments are performed to assess the contributions that denoising makes to the learning of representations for classification and sample synthesis. Our results suggest that autoencoders trained using a denoising criterion achieve higher classification performance and can synthesize samples that are more consistent with the input data than those trained without a corruption process.

Rethinking multiscale cardiac electrophysiology with machine learning and predictive modelling.

We review some of the latest approaches to analysing cardiac electrophysiology data using machine learning and predictive modelling. Cardiac arrhythmias, particularly atrial fibrillation, are a major global healthcare challenge. Treatment is often through catheter ablation, which involves the targeted localised destruction of regions of the myocardium responsible for initiating or perpetuating the arrhythmia. Ablation targets are either anatomically defined, or identified based on their functional properties as determined through the analysis of contact intracardiac electrograms acquired with increasing spatial density by modern electroanatomic mapping systems. While numerous quantitative approaches have been investigated over the past decades for identifying these critical curative sites, few have provided a reliable and reproducible advance in success rates. Machine learning techniques, including recent deep-learning approaches, offer a potential route to gaining new insight from this wealth of highly complex spatio-temporal information that existing methods struggle to analyse. Coupled with predictive modelling, these techniques offer exciting opportunities to advance the field and produce more accurate diagnoses and robust personalised treatment. We outline some of these methods and illustrate their use in making predictions from the contact electrogram and augmenting predictive modelling tools, both by more rapidly predicting future states of the system and by inferring the parameters of these models from experimental observations.