Evaluation of wave delivery methodology for brain MRE: Insights from computational simulations.

PURPOSE: MR elastography (MRE) of the brain is being explored as a biomarker of neurodegenerative disease such as dementia. However, MRE measures for healthy brain have varied widely. Differing wave delivery methodologies may have influenced this, hence finite element-based simulations were performed to explore this possibility. METHODS: The natural frequencies of a series of cranial models were calculated, and MRE-associated vibration was simulated for different wave delivery methods at varying frequency, using simple isotropic viscoelastic material models for the brain. Displacement fields and the corresponding brain constitutive properties estimated by standard inversion techniques were compared across delivery methods and frequencies. RESULTS: The delivery methods produced widely different MRE displacement fields and inversions. Furthermore, resonances at natural frequencies influenced the displacement patterns. Consequently, some delivery methods led to lower inversion errors than others, and the error on the storage modulus varied by up to 11% between methods. CONCLUSION: Wave delivery has a considerable impact on brain MRE reliability. Assuming small variations in brain biomechanics, as recently reported to accompany neurodegenerative disease (e.g., 7% for Alzheimer's disease), the effect of wave delivery is important. Hence, a consensus should be established on a consistent methodology to ensure diagnostic and prognostic consistency. Magn Reson Med 78:341-356, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Magnetic resonance elastography of the brain: An in silico study to determine the influence of cranial anatomy.

PURPOSE: Magnetic resonance elastography (MRE) of the brain has demonstrated potential as a biomarker of neurodegenerative disease such as dementia but requires further evaluation. Cranial anatomical features such as the falx cerebri and tentorium cerebelli membranes may influence MRE measurements through wave reflection and interference and tissue heterogeneity at their boundaries. We sought to determine the influence of these effects via simulation. METHODS: MRE-associated mechanical stimulation of the brain was simulated using steady state harmonic finite element analysis. Simulations of geometrical models and anthropomorphic brain models derived from anatomical MRI data of healthy individuals were compared. Constitutive parameters were taken from MRE measurements for healthy brain. Viscoelastic moduli were reconstructed from the simulated displacement fields and compared with ground truth. RESULTS: Interference patterns from reflections and heterogeneity resulted in artifacts in the reconstructions of viscoelastic moduli. Artifacts typically occurred in the vicinity of boundaries between different tissues within the cranium, with a magnitude of 10%-20%. CONCLUSION: Given that MRE studies for neurodegenerative disease have reported only marginal variations in brain elasticity between controls and patients (e.g., 7% for Alzheimer's disease), the predicted errors are a potential confound to the development of MRE as a biomarker of dementia and other neurodegenerative diseases. Magn Reson Med 76:645-662, 2016. © 2015 Wiley Periodicals, Inc.

Multi-centre benchmarking of deep learning models for COVID-19 detection in chest x-rays.

Introduction: This study is a retrospective evaluation of the performance of deep learning models that were developed for the detection of COVID-19 from chest x-rays, undertaken with the goal of assessing the suitability of such systems as clinical decision support tools. Methods: Models were trained on the National COVID-19 Chest Imaging Database (NCCID), a UK-wide multi-centre dataset from 26 different NHS hospitals and evaluated on independent multi-national clinical datasets. The evaluation considers clinical and technical contributors to model error and potential model bias. Model predictions are examined for spurious feature correlations using techniques for explainable prediction. Results: Models performed adequately on NHS populations, with performance comparable to radiologists, but generalised poorly to international populations. Models performed better in males than females, and performance varied across age groups. Alarmingly, models routinely failed when applied to complex clinical cases with confounding pathologies and when applied to radiologist defined "mild" cases. Discussion: This comprehensive benchmarking study examines the pitfalls in current practices that have led to impractical model development. Key findings highlight the need for clinician involvement at all stages of model development, from data curation and label definition, to model evaluation, to ensure that all clinical factors and disease features are appropriately considered during model design. This is imperative to ensure automated approaches developed for disease detection are fit-for-purpose in a clinical setting.

A Generative Shape Compositional Framework to Synthesize Populations of Virtual Chimeras.

Generating virtual organ populations that capture sufficient variability while remaining plausible is essential to conduct in silico trials (ISTs) of medical devices. However, not all anatomical shapes of interest are always available for each individual in a population. The imaging examinations and modalities used can vary between subjects depending on their individualized clinical pathways. Different imaging modalities may have various fields of view and are sensitive to signals from other tissues/organs, or both. Hence, missing/partially overlapping anatomical information is often available across individuals. We introduce a generative shape model for multipart anatomical structures, learnable from sets of unpaired datasets, i.e., where each substructure in the shape assembly comes from datasets with missing or partially overlapping substructures from disjoint subjects of the same population. The proposed generative model can synthesize complete multipart shape assemblies coined virtual chimeras (VCs). We applied this framework to build VCs from databases of whole-heart shape assemblies that each contribute samples for heart substructures. Specifically, we propose a graph neural network-based generative shape compositional framework, which comprises two components, a part-aware generative shape model that captures the variability in shape observed for each structure of interest in the training population and a spatial composition network that assembles/composes the structures synthesized by the former into multipart shape assemblies (i.e., VCs). We also propose a novel self-supervised learning scheme that enables the spatial composition network to be trained with partially overlapping data and weak labels. We trained and validated our approach using shapes of cardiac structures derived from cardiac magnetic resonance (MR) images in the UK Biobank (UKBB). When trained with complete and partially overlapping data, our approach significantly outperforms a principal component analysis (PCA)-based shape model (trained with complete data) in terms of generalizability and specificity. This demonstrates the superiority of the proposed method, as the synthesized cardiac virtual populations are more plausible and capture a greater degree of shape variability than those generated by the PCA-based shape model.

Joint shape/texture representation learning for cardiovascular disease diagnosis from magnetic resonance imaging.

Aims: Cardiovascular diseases (CVDs) are the leading cause of mortality worldwide. Cardiac image and mesh are two primary modalities to present the shape and structure of the heart and have been demonstrated to be efficient in CVD prediction and diagnosis. However, previous research has been generally focussed on a single modality (image or mesh), and few of them have tried to jointly consider the image and mesh representations of heart. To obtain efficient and explainable biomarkers for CVD prediction and diagnosis, it is needed to jointly consider both representations. Methods and results: We design a novel multi-channel variational auto-encoder, mesh-image variational auto-encoder, to learn joint representation of paired mesh and image. After training, the shape-aware image representation (SAIR) can be learned directly from the raw images and applied for further CVD prediction and diagnosis. We demonstrate our method on data from UK Biobank study and two other datasets via extensive experiments. In acute myocardial infarction prediction, SAIR achieves 81.43% accuracy, significantly higher than traditional biomarkers like metadata and clinical indices (left ventricle and right ventricle clinical indices of cardiac function like chamber volume, mass, and ejection fraction). Conclusion: Our mesh-image variational auto-encoder provides a novel approach for 3D cardiac mesh reconstruction from images. The extraction of SAIR is fast and without need of segmentation masks, and its focussing can be visualized in the corresponding cardiac meshes. SAIR archives better performance than traditional biomarkers and can be applied as an efficient supplement to them, which is of significant potential in CVD analysis.

Beyond images: an integrative multi-modal approach to chest x-ray report generation.

Image-to-text radiology report generation aims to automatically produce radiology reports that describe the findings in medical images. Most existing methods focus solely on the image data, disregarding the other patient information accessible to radiologists. In this paper, we present a novel multi-modal deep neural network framework for generating chest x-rays reports by integrating structured patient data, such as vital signs and symptoms, alongside unstructured clinical notes. We introduce a conditioned cross-multi-head attention module to fuse these heterogeneous data modalities, bridging the semantic gap between visual and textual data. Experiments demonstrate substantial improvements from using additional modalities compared to relying on images alone. Notably, our model achieves the highest reported performance on the ROUGE-L metric compared to relevant state-of-the-art models in the literature. Furthermore, we employed both human evaluation and clinical semantic similarity measurement alongside word-overlap metrics to improve the depth of quantitative analysis. A human evaluation, conducted by a board-certified radiologist, confirms the model's accuracy in identifying high-level findings, however, it also highlights that more improvement is needed to capture nuanced details and clinical context.

Retinal imaging for the assessment of stroke risk: a systematic review.

BACKGROUND: Stroke is a leading cause of morbidity and mortality. Retinal imaging allows non-invasive assessment of the microvasculature. Consequently, retinal imaging is a technology which is garnering increasing attention as a means of assessing cardiovascular health and stroke risk. METHODS: A biomedical literature search was performed to identify prospective studies that assess the role of retinal imaging derived biomarkers as indicators of stroke risk. RESULTS: Twenty-four studies were included in this systematic review. The available evidence suggests that wider retinal venules, lower fractal dimension, increased arteriolar tortuosity, presence of retinopathy, and presence of retinal emboli are associated with increased likelihood of stroke. There is weaker evidence to suggest that narrower arterioles and the presence of individual retinopathy traits such as microaneurysms and arteriovenous nicking indicate increased stroke risk. Our review identified three models utilizing artificial intelligence algorithms for the analysis of retinal images to predict stroke. Two of these focused on fundus photographs, whilst one also utilized optical coherence tomography (OCT) technology images. The constructed models performed similarly to conventional risk scores but did not significantly exceed their performance. Only two studies identified in this review used OCT imaging, despite the higher dimensionality of this data. CONCLUSION: Whilst there is strong evidence that retinal imaging features can be used to indicate stroke risk, there is currently no predictive model which significantly outperforms conventional risk scores. To develop clinically useful tools, future research should focus on utilization of deep learning algorithms, validation in external cohorts, and analysis of OCT images.

Unsupervised ensemble-based phenotyping enhances discoverability of genes related to left-ventricular morphology.

Recent genome-wide association studies have successfully identified associations between genetic variants and simple cardiac morphological parameters derived from cardiac magnetic resonance images. However, the emergence of large databases, including genetic data linked to cardiac magnetic resonance facilitates the investigation of more nuanced patterns of cardiac shape variability than those studied so far. Here we propose a framework for gene discovery coined unsupervised phenotype ensembles. The unsupervised phenotype ensemble builds a redundant yet highly expressive representation by pooling a set of phenotypes learnt in an unsupervised manner, using deep learning models trained with different hyperparameters. These phenotypes are then analysed via genome-wide association studies, retaining only highly confident and stable associations across the ensemble. We applied our approach to the UK Biobank database to extract geometric features of the left ventricle from image-derived three-dimensional meshes. We demonstrate that our approach greatly improves the discoverability of genes that influence left ventricle shape, identifying 49 loci with study-wide significance and 25 with suggestive significance. We argue that our approach would enable more extensive discovery of gene associations with image-derived phenotypes for other organs or image modalities.

Virtual high-resolution MR angiography from non-angiographic multi-contrast MRIs: synthetic vascular model populations for in-silico trials.

Despite success on multi-contrast MR image synthesis, generating specific modalities remains challenging. Those include Magnetic Resonance Angiography (MRA) that highlights details of vascular anatomy using specialised imaging sequences for emphasising inflow effect. This work proposes an end-to-end generative adversarial network that can synthesise anatomically plausible, high-resolution 3D MRA images using commonly acquired multi-contrast MR images (e.g. T1/T2/PD-weighted MR images) for the same subject whilst preserving the continuity of vascular anatomy. A reliable technique for MRA synthesis would unleash the research potential of very few population databases with imaging modalities (such as MRA) that enable quantitative characterisation of whole-brain vasculature. Our work is motivated by the need to generate digital twins and virtual patients of cerebrovascular anatomy for in-silico studies and/or in-silico trials. We propose a dedicated generator and discriminator that leverage the shared and complementary features of multi-source images. We design a composite loss function for emphasising vascular properties by minimising the statistical difference between the feature representations of the target images and the synthesised outputs in both 3D volumetric and 2D projection domains. Experimental results show that the proposed method can synthesise high-quality MRA images and outperform the state-of-the-art generative models both qualitatively and quantitatively. The importance assessment reveals that T2 and PD-weighted images are better predictors of MRA images than T1; and PD-weighted images contribute to better visibility of small vessel branches towards the peripheral regions. In addition, the proposed approach can generalise to unseen data acquired at different imaging centres with different scanners, whilst synthesising MRAs and vascular geometries that maintain vessel continuity. The results show the potential for use of the proposed approach to generating digital twin cohorts of cerebrovascular anatomy at scale from structural MR images typically acquired in population imaging initiatives.

Radiomics in the evaluation of ovarian masses - a systematic review.

OBJECTIVES: The study aim was to conduct a systematic review of the literature reporting the application of radiomics to imaging techniques in patients with ovarian lesions. METHODS: MEDLINE/PubMed, Web of Science, Scopus, EMBASE, Ovid and ClinicalTrials.gov were searched for relevant articles. Using PRISMA criteria, data were extracted from short-listed studies. Validity and bias were assessed independently by 2 researchers in consensus using the Quality in Prognosis Studies (QUIPS) tool. Radiomic Quality Score (RQS) was utilised to assess radiomic methodology. RESULTS: After duplicate removal, 63 articles were identified, of which 33 were eligible. Fifteen assessed lesion classifications, 10 treatment outcomes, 5 outcome predictions, 2 metastatic disease predictions and 1 classification/outcome prediction. The sample size ranged from 28 to 501 patients. Twelve studies investigated CT, 11 MRI, 4 ultrasound and 1 FDG PET-CT. Twenty-three studies (70%) incorporated 3D segmentation. Various modelling methods were used, most commonly LASSO (least absolute shrinkage and selection operator) (10/33). Five studies (15%) compared radiomic models to radiologist interpretation, all demonstrating superior performance. Only 6 studies (18%) included external validation. Five studies (15%) had a low overall risk of bias, 9 (27%) moderate, and 19 (58%) high risk of bias. The highest RQS achieved was 61.1%, and the lowest was - 16.7%. CONCLUSION: Radiomics has the potential as a clinical diagnostic tool in patients with ovarian masses and may allow better lesion stratification, guiding more personalised patient care in the future. Standardisation of the feature extraction methodology, larger and more diverse patient cohorts and real-world evaluation is required before clinical translation. CLINICAL RELEVANCE STATEMENT: Radiomics shows promising results in improving lesion stratification, treatment selection and outcome prediction. Modelling with larger cohorts and real-world evaluation is required before clinical translation. KEY POINTS: • Radiomics is emerging as a tool for enhancing clinical decisions in patients with ovarian masses. • Radiomics shows promising results in improving lesion stratification, treatment selection and outcome prediction. • Modelling with larger cohorts and real-world evaluation is required before clinical translation.

High-throughput 3DRA segmentation of brain vasculature and aneurysms using deep learning.

BACKGROUND AND OBJECTIVES: Automatic segmentation of the cerebral vasculature and aneurysms facilitates incidental detection of aneurysms. The assessment of aneurysm rupture risk assists with pre-operative treatment planning and enables in-silico investigation of cerebral hemodynamics within and in the vicinity of aneurysms. However, ensuring precise and robust segmentation of cerebral vessels and aneurysms in neuroimaging modalities such as three-dimensional rotational angiography (3DRA) is challenging. The vasculature constitutes a small proportion of the image volume, resulting in a large class imbalance (relative to surrounding brain tissue). Additionally, aneurysms and vessels have similar image/appearance characteristics, making it challenging to distinguish the aneurysm sac from the vessel lumen. METHODS: We propose a novel multi-class convolutional neural network to tackle these challenges and facilitate the automatic segmentation of cerebral vessels and aneurysms in 3DRA images. The proposed model is trained and evaluated on an internal multi-center dataset and an external publicly available challenge dataset. RESULTS: On the internal clinical dataset, our method consistently outperformed several state-of-the-art approaches for vessel and aneurysm segmentation, achieving an average Dice score of 0.81 (0.15 higher than nnUNet) and an average surface-to-surface error of 0.20 mm (less than the in-plane resolution (0.35 mm/pixel)) for aneurysm segmentation; and an average Dice score of 0.91 and average surface-to-surface error of 0.25 mm for vessel segmentation. In 223 cases of a clinical dataset, our method accurately segmented 190 aneurysm cases. CONCLUSIONS: The proposed approach can help address class imbalance problems and inter-class interference problems in multi-class segmentation. Besides, this method performs consistently on clinical datasets from four different sources and the generated results are qualified for hemodynamic simulation. Code available at https://github.com/cistib/vessel-aneurysm-segmentation.

Contribution of Shape Features to Intradiscal Pressure and Facets Contact Pressure in L4/L5 FSUs: An In-Silico Study.

Finite element models (FEMs) of the spine commonly use a limited number of simplified geometries. Nevertheless, the geometric features of the spine are important in determining its FEM outcomes. The link between a spinal segment's shape and its biomechanical response has been studied, but the co-variances of the shape features have been omitted. We used a principal component (PCA)-based statistical shape modelling (SSM) approach to investigate the contribution of shape features to the intradiscal pressure (IDP) and the facets contact pressure (FCP) in a cohort of synthetic L4/L5 functional spinal units under axial compression. We quantified the uncertainty in the FEM results, and the contribution of individual shape modes to these results. This parameterisation approach is able to capture the variability in the correlated anatomical features in a real population and sample plausible synthetic geometries. The first shape mode ([Formula: see text]) explained 22.6% of the shape variation in the subject-specific cohort used to train the SSM, and had the largest correlation with, and contribution to IDP (17%) and FCP (11%). The largest geometric variation in ([Formula: see text]) was in the annulus-nucleus ratio.

Artificial intelligence in ovarian cancer histopathology: a systematic review.

This study evaluates the quality of published research using artificial intelligence (AI) for ovarian cancer diagnosis or prognosis using histopathology data. A systematic search of PubMed, Scopus, Web of Science, Cochrane CENTRAL, and WHO-ICTRP was conducted up to May 19, 2023. Inclusion criteria required that AI was used for prognostic or diagnostic inferences in human ovarian cancer histopathology images. Risk of bias was assessed using PROBAST. Information about each model was tabulated and summary statistics were reported. The study was registered on PROSPERO (CRD42022334730) and PRISMA 2020 reporting guidelines were followed. Searches identified 1573 records, of which 45 were eligible for inclusion. These studies contained 80 models of interest, including 37 diagnostic models, 22 prognostic models, and 21 other diagnostically relevant models. Common tasks included treatment response prediction (11/80), malignancy status classification (10/80), stain quantification (9/80), and histological subtyping (7/80). Models were developed using 1-1375 histopathology slides from 1-776 ovarian cancer patients. A high or unclear risk of bias was found in all studies, most frequently due to limited analysis and incomplete reporting regarding participant recruitment. Limited research has been conducted on the application of AI to histopathology images for diagnostic or prognostic purposes in ovarian cancer, and none of the models have been demonstrated to be ready for real-world implementation. Key aspects to accelerate clinical translation include transparent and comprehensive reporting of data provenance and modelling approaches, and improved quantitative evaluation using cross-validation and external validations. This work was funded by the Engineering and Physical Sciences Research Council.

RecON: Online learning for sensorless freehand 3D ultrasound reconstruction.

Sensorless freehand 3D ultrasound (US) reconstruction based on deep networks shows promising advantages, such as large field of view, relatively high resolution, low cost, and ease of use. However, existing methods mainly consider vanilla scan strategies with limited inter-frame variations. These methods thus are degraded on complex but routine scan sequences in clinics. In this context, we propose a novel online learning framework for freehand 3D US reconstruction under complex scan strategies with diverse scanning velocities and poses. First, we devise a motion-weighted training loss in training phase to regularize the scan variation frame-by-frame and better mitigate the negative effects of uneven inter-frame velocity. Second, we effectively drive online learning with local-to-global pseudo supervisions. It mines both the frame-level contextual consistency and the path-level similarity constraint to improve the inter-frame transformation estimation. We explore a global adversarial shape before transferring the latent anatomical prior as supervision. Third, we build a feasible differentiable reconstruction approximation to enable the end-to-end optimization of our online learning. Experimental results illustrate that our freehand 3D US reconstruction framework outperformed current methods on two large, simulated datasets and one real dataset. In addition, we applied the proposed framework to clinical scan videos to further validate its effectiveness and generalizability.

Hemodynamics of thrombus formation in intracranial aneurysms: An in silico observational study.

How prevalent is spontaneous thrombosis in a population containing all sizes of intracranial aneurysms? How can we calibrate computational models of thrombosis based on published data? How does spontaneous thrombosis differ in normo- and hypertensive subjects? We address the first question through a thorough analysis of published datasets that provide spontaneous thrombosis rates across different aneurysm characteristics. This analysis provides data for a subgroup of the general population of aneurysms, namely, those of large and giant size (>10 mm). Based on these observed spontaneous thrombosis rates, our computational modeling platform enables the first in silico observational study of spontaneous thrombosis prevalence across a broader set of aneurysm phenotypes. We generate 109 virtual patients and use a novel approach to calibrate two trigger thresholds: residence time and shear rate, thus addressing the second question. We then address the third question by utilizing this calibrated model to provide new insight into the effects of hypertension on spontaneous thrombosis. We demonstrate how a mechanistic thrombosis model calibrated on an intracranial aneurysm cohort can help estimate spontaneous thrombosis prevalence in a broader aneurysm population. This study is enabled through a fully automatic multi-scale modeling pipeline. We use the clinical spontaneous thrombosis data as an indirect population-level validation of a complex computational modeling framework. Furthermore, our framework allows exploration of the influence of hypertension in spontaneous thrombosis. This lays the foundation for in silico clinical trials of cerebrovascular devices in high-risk populations, e.g., assessing the performance of flow diverters in aneurysms for hypertensive patients.

An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images.

The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segmented by convolutional neural network (CNN) on FDG PET-CT of aortitis and control patients. The FDG PET-CT dataset was split into training (43 aortitis:21 control), test (12 aortitis:5 control) and validation (24 aortitis:14 control) cohorts. Radiomic features (RF), including SUV metrics, were extracted from the segmented data and harmonized. Three radiomic fingerprints were constructed: A-RFs with high diagnostic utility removing highly correlated RFs; B used principal component analysis (PCA); C-Random Forest intrinsic feature selection. The diagnostic utility was evaluated with accuracy and area under the receiver operating characteristic curve (AUC). Several RFs and Fingerprints had high AUC values (AUC > 0.8), confirmed by balanced accuracy, across training, test and external validation datasets. Good diagnostic performance achieved across several multi-centre datasets suggests that a radiomic pipeline can be generalizable. These findings could be used to build an automated clinical decision tool to facilitate objective and standardized assessment regardless of observer experience.

Mitosis domain generalization in histopathology images - The MIDOG challenge.

The density of mitotic figures (MF) within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of MF by pathologists is subject to a strong inter-rater bias, limiting its prognostic value. State-of-the-art deep learning methods can support experts but have been observed to strongly deteriorate when applied in a different clinical environment. The variability caused by using different whole slide scanners has been identified as one decisive component in the underlying domain shift. The goal of the MICCAI MIDOG 2021 challenge was the creation of scanner-agnostic MF detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were provided. In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance. The winning algorithm yielded an F1 score of 0.748 (CI95: 0.704-0.781), exceeding the performance of six experts on the same task.

Learning to complete incomplete hearts for population analysis of cardiac MR images.

Cardiac MR acquisition with complete coverage from base to apex is required to ensure accurate subsequent analyses, such as volumetric and functional measurements. However, this requirement cannot be guaranteed when acquiring images in the presence of motion induced by cardiac muscle contraction and respiration. To address this problem, we propose an effective two-stage pipeline for detecting and synthesising absent slices in both the apical and basal region. The detection model comprises several dense blocks containing convolutional long short-term memory (ConvLSTM) layers, to leverage through-plane contextual and sequential ordering information of slices in cine MR data and achieve reliable classification results. The imputation network is based on a dedicated conditional generative adversarial network (GAN) that helps retain key visual cues and fine structural details in the synthesised image slices. The proposed network can infer multiple missing slices that are anatomically plausible and lead to improved accuracy of subsequent analyses on cardiac MRIs, e.g., ventricle segmentation, cardiac quantification compared to those derived from incomplete cardiac MR datasets. For instance, the results obtained when compensating for the absence of two basal slices show that the mean differences to the reference of stroke volume and ejection fraction are only -1.3 mL and -1.0%, respectively, which are significantly smaller than those calculated from the incomplete data (-26.8 mL and -6.7%). The proposed approach can improve the reliability of high-throughput image analysis in large-scale population studies, minimising the need for re-scanning patients or discarding incomplete acquisitions.

A probabilistic deep motion model for unsupervised cardiac shape anomaly assessment.

Automatic shape anomaly detection in large-scale imaging data can be useful for screening suboptimal segmentations and pathologies altering the cardiac morphology without intensive manual labour. We propose a deep probabilistic model for local anomaly detection in sequences of heart shapes, modelled as point sets, in a cardiac cycle. A deep recurrent encoder-decoder network captures the spatio-temporal dependencies to predict the next shape in the cycle and thus derive the outlier points that are attributed to excessive deviations from the network prediction. A predictive mixture distribution models the inlier and outlier classes via Gaussian and uniform distributions, respectively. A Gibbs sampling Expectation-Maximisation (EM) algorithm computes soft anomaly scores of the points via the posterior probabilities of each class in the E-step and estimates the parameters of the network and the predictive distribution in the M-step. We demonstrate the versatility of the method using two shape datasets derived from: (i) one million biventricular CMR images from 20,000 participants in the UK Biobank (UKB), and (ii) routine diagnostic imaging from Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac Image (M&Ms). Experiments show that the detected shape anomalies in the UKB dataset are mostly associated with poor segmentation quality, and the predicted shape sequences show significant improvement over the input sequences. Furthermore, evaluations on U-Net based shapes from the M&Ms dataset reveals that the anomalies are attributable to the underlying pathologies that affect the ventricles. The proposed model can therefore be used as an effective mechanism to sift shape anomalies in large-scale cardiac imaging pipelines for further analysis.

The Pitfalls of Using Open Data to Develop Deep Learning Solutions for COVID-19 Detection in Chest X-Rays.

Since the emergence of COVID-19, deep learning models have been developed to identify COVID-19 from chest X-rays. With little to no direct access to hospital data, the AI community relies heavily on public data comprising numerous data sources. Model performance results have been exceptional when training and testing on open-source data, surpassing the reported capabilities of AI in pneumonia-detection prior to the COVID-19 outbreak. In this study impactful models are trained on a widely used open-source data and tested on an external test set and a hospital dataset, for the task of classifying chest X-rays into one of three classes: COVID-19, non-COVID pneumonia and no-pneumonia. Classification performance of the models investigated is evaluated through ROC curves, confusion matrices and standard classification metrics. Explainability modules are implemented to explore the image features most important to classification. Data analysis and model evalutions show that the popular open-source dataset COVIDx is not representative of the real clinical problem and that results from testing on this are inflated. Dependence on open-source data can leave models vulnerable to bias and confounding variables, requiring careful analysis to develop clinically useful/viable AI tools for COVID-19 detection in chest X-rays.