The effectiveness and feasibility of ball-milled powdered activated carbon (BPAC) for removal of organic pesticides in conventional drinking water treatment process.

In the conventional drinking water treatment process (CDWTP), powdered activated carbon (PAC) is commonly used for removing organic pesticides, or other organic contaminants. However, the hydraulic retention time (HRT) in CDWTP is insufficient for fulfilling PAC adsorption equilibrium to realize its full capacity. This study examined the adsorption kinetics, adsorption thermal dynamics, and removal efficiency for six organic pesticides using the ball-milled PAC (BPAC) with varying particle sizes in CDWTP. Based on the experiments with the pesticides of atrazine, diazinon, dimethoate, fenitrothion, isoproturon and thiometon, the results indicated that as the particle size reduced from around 38 µm for the commercial PAC to 1 µm for the BPAC, the adsorption rates for hydrophobic pesticides increased up to twentyfold. Diffusional adsorption from the bulk solution to the external PAC surface is the most likely predominant mechanism. This could allow a sufficient pesticides' adsorption within the limited HRT and to achieve a great depth removal of these toxic compounds. However, the addition of BPAC with a diameter of 1 µm was observed to significantly increase residual particles in treated water after the conventional treatment process. With a further systematic evaluation of both adsorption rate and particle penetration, a particle size of around 6 µm BPAC was considered a practical compromise between the adsorption rate and particle penetration for real application. Results from five surface waters of different water quality indicated that, compared to commercial PAC, application of 6 µm BPAC could achieve up to a 75% reduction in adsorbent dosage while maintaining around the same pesticide removal efficiencies. Additionally, thermodynamic analyses suggest that adsorption of these pesticides could be enthalpically or entropically driven depending on the degree of pesticide hydrophobicity.

Multi-granularity learning of explicit geometric constraint and contrast for label-efficient medical image segmentation and differentiable clinical function assessment.

Automated segmentation is a challenging task in medical image analysis that usually requires a large amount of manually labeled data. However, most current supervised learning based algorithms suffer from insufficient manual annotations, posing a significant difficulty for accurate and robust segmentation. In addition, most current semi-supervised methods lack explicit representations of geometric structure and semantic information, restricting segmentation accuracy. In this work, we propose a hybrid framework to learn polygon vertices, region masks, and their boundaries in a weakly/semi-supervised manner that significantly advances geometric and semantic representations. Firstly, we propose multi-granularity learning of explicit geometric structure constraints via polygon vertices (PolyV) and pixel-wise region (PixelR) segmentation masks in a semi-supervised manner. Secondly, we propose eliminating boundary ambiguity by using an explicit contrastive objective to learn a discriminative feature space of boundary contours at the pixel level with limited annotations. Thirdly, we exploit the task-specific clinical domain knowledge to differentiate the clinical function assessment end-to-end. The ground truth of clinical function assessment, on the other hand, can serve as auxiliary weak supervision for PolyV and PixelR learning. We evaluate the proposed framework on two tasks, including optic disc (OD) and cup (OC) segmentation along with vertical cup-to-disc ratio (vCDR) estimation in fundus images; left ventricle (LV) segmentation at end-diastolic and end-systolic frames along with ejection fraction (LVEF) estimation in two-dimensional echocardiography images. Experiments on nine large-scale datasets of the two tasks under different label settings demonstrate our model's superior performance on segmentation and clinical function assessment.

Development of automated neural network prediction for echocardiographic left ventricular ejection fraction.

Introduction: The echocardiographic measurement of left ventricular ejection fraction (LVEF) is fundamental to the diagnosis and classification of patients with heart failure (HF). Methods: This paper aimed to quantify LVEF automatically and accurately with the proposed pipeline method based on deep neural networks and ensemble learning. Within the pipeline, an Atrous Convolutional Neural Network (ACNN) was first trained to segment the left ventricle (LV), before employing the area-length formulation based on the ellipsoid single-plane model to calculate LVEF values. This formulation required inputs of LV area, derived from segmentation using an improved Jeffrey's method, as well as LV length, derived from a novel ensemble learning model. To further improve the pipeline's accuracy, an automated peak detection algorithm was used to identify end-diastolic and end-systolic frames, avoiding issues with human error. Subsequently, single-beat LVEF values were averaged across all cardiac cycles to obtain the final LVEF. Results: This method was developed and internally validated in an open-source dataset containing 10,030 echocardiograms. The Pearson's correlation coefficient was 0.83 for LVEF prediction compared to expert human analysis (p < 0.001), with a subsequent area under the receiver operator curve (AUROC) of 0.98 (95% confidence interval 0.97 to 0.99) for categorisation of HF with reduced ejection (HFrEF; LVEF<40%). In an external dataset with 200 echocardiograms, this method achieved an AUC of 0.90 (95% confidence interval 0.88 to 0.91) for HFrEF assessment. Conclusion: The automated neural network-based calculation of LVEF is comparable to expert clinicians performing time-consuming, frame-by-frame manual evaluations of cardiac systolic function.

Gait Intention Prediction Using a Lower-Limb Musculoskeletal Model and Long Short-Term Memory Neural Networks.

The prediction of gait motion intention is essential for achieving intuitive control of assistive devices and diagnosing gait disorders. To reduce the cost associated with using multimodal signals and signal processing, we proposed a novel method that integrates machine learning with musculoskeletal modelling techniques for the prediction of time-series joint angles, using only kinematic signals. Additionally, we hypothesised that a stacked long short-term memory (LSTM) neural network architecture can perform the task without relying on any ahead-of-motion features typically provided by electromyography signals. Optical cameras and inertial measurement unit (IMU) sensors were used to track level gait kinematics. Joint angles were modelled using the musculoskeletal model. The optimal LSTM architecture in fulfilling the prediction task was determined. Joint angle predictions were performed for joints on the sagittal plane, benefiting from joint angle modelling using signals from optical cameras and IMU sensors. Our proposed method predicted the upcoming joint angles in the prediction time of 10 ms, with an averaged root mean square error of 5.3° and a coefficient of determination of 0.81. Moreover, in support of our hypothesis, the recurrent stacked LSTM network demonstrated its ability to predict intended motion accurately and efficiently in gait, outperforming two other neural network architectures: a feedforward MLP and a hybrid LSTM-MLP. The method paves the way for the development of a cost-effective, single-modal control system for assistive devices in gait rehabilitation.

Efficient Adsorption of Tebuconazole in Aqueous Solution by Calcium Modified Water Hyacinth-Based Biochar: Adsorption Kinetics, Mechanism, and Feasibility.

The application of fungicides (such as tebuconazole) can impose harmful impacts on the ecosystem and humans. In this study, a new calcium modified water hyacinth-based biochar (WHCBC) was prepared and its effectiveness for removing tebuconazole (TE) via adsorption from water was tested. The results showed that Ca was loaded chemically (CaC2O4) onto the surface of WHCBC. The adsorption capacity of the modified biochar increased by 2.5 times in comparison to that of the unmodified water hyacinth biochar. The enhanced adsorption was attributed to the improved chemical adsorption capacity of the biochar through calcium modification. The adsorption data were better fitted to the pseudo-second-order kinetics and the Langmuir isotherm model, indicating that the adsorption process was dominated by monolayer adsorption. It was found that liquid film diffusion was the main rate-limiting step in the adsorption process. The maximum adsorption capacity of WHCBC was 40.5 mg/g for TE. The results indicate that the absorption mechanisms involved surface complexation, hydrogen bonding, and π-π interactions. The inhibitory rate of Cu2+ and Ca2+ on the adsorption of TE by WHCBC were at 4.05-22.8%. In contrast, the presence of other coexisting cations (Cr6+, K+, Mg2+, Pb2+), as well as natural organic matter (humic acid), could promote the adsorption of TE by 4.45-20.9%. In addition, the regeneration rate of WHCBC was able to reach up to 83.3% after five regeneration cycles by desorption stirring with 0.2 mol/L HCl (t = 360 min). The results suggest that WHCBC has a potential in application for removing TE from water.

A Stable Fe-Zn Modified Sludge-Derived Biochar for Diuron Removal: Kinetics, Isotherms, Mechanism, and Practical Research.

To remove typical herbicide diuron effectively, a novel sludge-derived modified biochar (SDMBC600) was prepared using sludge-derived biochar (SDBC600) as raw material and Fe-Zn as an activator and modifier in this study. The physico-chemical properties of SDMBC600 and the adsorption behavior of diuron on the SDMBC600 were studied systematically. The adsorption mechanisms as well as practical applications of SDMBC600 were also investigated and examined. The results showed that the SDMBC600 was chemically loaded with Fe-Zn and SDMBC600 had a larger specific surface area (204 m2/g) and pore volume (0.0985 cm3/g). The adsorption of diuron on SDMBC600 followed pseudo-second-order kinetics and the Langmuir isotherm model, with a maximum diuron adsorption capacity of 17.7 mg/g. The biochar could maintain a good adsorption performance (8.88-12.9 mg/g) under wide water quality conditions, in the pH of 2-10 and with the presence of humic acid and six typical metallic ions of 0-20 mg/L. The adsorption mechanisms of SDMBC600 for diuron were found to include surface complexation, π-π binding, hydrogen bonding, as well as pore filling. Additionally, the SDMBC600 was tested to be very stable with very low Fe and Zn leaching concentration ≤0.203 mg/L in the wide pH range. In addition, the SDMBC600 could maintain a high adsorption capacity (99.6%) after four times of regeneration and therefore, SDMBC600 could have a promising application for diuron removal in water treatment.

Artificial intelligence to enhance clinical value across the spectrum of cardiovascular healthcare.

Artificial intelligence (AI) is increasingly being utilized in healthcare. This article provides clinicians and researchers with a step-wise foundation for high-value AI that can be applied to a variety of different data modalities. The aim is to improve the transparency and application of AI methods, with the potential to benefit patients in routine cardiovascular care. Following a clear research hypothesis, an AI-based workflow begins with data selection and pre-processing prior to analysis, with the type of data (structured, semi-structured, or unstructured) determining what type of pre-processing steps and machine-learning algorithms are required. Algorithmic and data validation should be performed to ensure the robustness of the chosen methodology, followed by an objective evaluation of performance. Seven case studies are provided to highlight the wide variety of data modalities and clinical questions that can benefit from modern AI techniques, with a focus on applying them to cardiovascular disease management. Despite the growing use of AI, further education for healthcare workers, researchers, and the public are needed to aid understanding of how AI works and to close the existing gap in knowledge. In addition, issues regarding data access, sharing, and security must be addressed to ensure full engagement by patients and the public. The application of AI within healthcare provides an opportunity for clinicians to deliver a more personalized approach to medical care by accounting for confounders, interactions, and the rising prevalence of multi-morbidity.

Learn2Reg: Comprehensive Multi-Task Medical Image Registration Challenge, Dataset and Evaluation in the Era of Deep Learning.

Image registration is a fundamental medical image analysis task, and a wide variety of approaches have been proposed. However, only a few studies have comprehensively compared medical image registration approaches on a wide range of clinically relevant tasks. This limits the development of registration methods, the adoption of research advances into practice, and a fair benchmark across competing approaches. The Learn2Reg challenge addresses these limitations by providing a multi-task medical image registration data set for comprehensive characterisation of deformable registration algorithms. A continuous evaluation will be possible at https://learn2reg.grand-challenge.org. Learn2Reg covers a wide range of anatomies (brain, abdomen, and thorax), modalities (ultrasound, CT, MR), availability of annotations, as well as intra- and inter-patient registration evaluation. We established an easily accessible framework for training and validation of 3D registration methods, which enabled the compilation of results of over 65 individual method submissions from more than 20 unique teams. We used a complementary set of metrics, including robustness, accuracy, plausibility, and runtime, enabling unique insight into the current state-of-the-art of medical image registration. This paper describes datasets, tasks, evaluation methods and results of the challenge, as well as results of further analysis of transferability to new datasets, the importance of label supervision, and resulting bias. While no single approach worked best across all tasks, many methodological aspects could be identified that push the performance of medical image registration to new state-of-the-art performance. Furthermore, we demystified the common belief that conventional registration methods have to be much slower than deep-learning-based methods.

DVS-Net: Dual-domain Variable Splitting Network for Accelerated Parallel MRI Data.

Parallel imaging is an important method to accel-erate the acquisition of magnetic resonance imaging data, which can shorten the breath-hold times and reduce motion artifacts. In this paper, we propose a joint frequency domain and image domain (dual-domain) reconstruction method by introducing the full sampling condition for the undersampled multi-coil MR data. The motivation is that the dual domain method can provide more information for accurate image reconstruction. An efficient iterative algorithm is developed based on the variable splitting technique and alternating direction method of multipliers, which is unrolled into an end-to-end trainable deep neural network. We evaluate the proposed network on complex valued multi-coil knee images for both 6-fold and 8-fold acceleration factors, and compare with both variational and deep learning based reconstruction algorithms. The numerical results demonstrate that our method provides better reconstruction accuracy and perceptual quality by making using of the dual domain information. Clinical relevance: This improves the reconstruction quality for accelerated parallel MRI data both visually and quantitatively.

Automatic Segmentation of Target Structures for Total Marrow and Lymphoid Irradiation in Bone Marrow Transplantation.

The use of total marrow and lymphoid irradiation (TMLI) as part of conditioning regimens for bone marrow transplantation is trending due to its advantages in disease control and low toxicity. Accurate contouring of target structures such as bone and lymph nodes plays an important role in irradiation planning. However, this process is often time-consuming and prone to inter-observer variation. Recently, deep learning methods such as convolutional neural networks (CNNs) and vision transformers have achieved tremendous success in medical image segmentation, therefore enabling fast semiautomatic radiotherapy planning. In this paper, we propose a dual-encoder U-shaped model named DE-Net, to automatically segment the target structures for TMLI. To enhance the learned features, the encoder of DE-Net is composed of parallel CNNs and vision transformers, which can model both local and global contexts. The multi-level features from the two branches are progressively fused by intermediate modules, therefore effectively preserving low-level details. Our experiments demonstrate that the proposed method achieves state-of-the-art results and a significant improvement in lymph node segmentation compared with existing methods.

Degradation of the typical herbicide atrazine by UV/persulfate: kinetics and mechanisms.

Atrazine (ATZ), a widely used herbicide, had received a significant amount of attention due to its widespread detection in aquatic environments as well as its potential risks to human health. UV/persulfate (PS) process is an emerging technology for degrading organic pollutants in water. Thus, the degradation of ATZ by a UV/PS process was investigated in this study. The results showed that the removal rate of ATZ was 98.4% with a PS dosage of 2 mg/L and an initial ATZ concentration of 0.1 mg/L. In addition, a relatively high degradation efficiency was obtained under pH = 7. However, the addition of humic acid (HA) reduced the removal rate of ATZ. Hydroxyl radicals (•OH) and sulfate radicals (•SO4-) respectively contributed to 21.7% and 29% of the ATZ degradation. The ATZ degradation pathway was proposed, and the main reactions of ATZ in this UV/PS process included dechlorination, demethylation, and deethylation. Moreover, the toxicity of ATZ and its degradation products was assessed using the Toxicity Estimation Software Tool (TEST), and the results showed that the toxicity of the ATZ solution was reduced after the UV/PS process. These results indicate that UV/PS shows good promise as a remediation technique for the treatment of persistent herbicides such as ATZ in contaminated water.

A sensitive approach for screening acetylcholinesterase inhibition of water samples using ultra-performance liquid chromatography-tandem mass spectrometry.

A sensitive assay was developed to evaluate inhibitory effects of aqueous solution on acetylcholinesterase (AChE) activity via measuring hydrolysis rates of acetylcholine (ACh) based on ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Upon having identified precursor ions and product ions of the ACh and its hydrolysis products choline (Ch), the separation chromatogram for these two analytes has been established using a 50 mm reverse-phase BEH Shield RP18 column. The total chromatographic separation time is 7 min; limits of detection (LODs) for ACh and Ch are 0.14 µg L-1 and 0.12 µg L-1, respectively. A simple method for inactivation of AChE and optimization of operational parameters were then sequentially performed. It was found that adjusting solution pH to 2.5 not only can terminate the enzymatic reaction but also solve band shifting and broadening caused by aqueous matrices in chromatographic separation during UPLC-MS/MS detection. Under conditions of 0.00075 U mL-1 AChE, initial concentration of ACh at 100 µg L-1 and 20 min observation time, IC50 values of the proposed assay for chlorpyrifos-oxon, diazoxon, malaoxon, methidathion oxon, omethoate and paraoxon were 3.5 nM, 16.8 nM, 2.4 nM, 6.8 nM, 270 nM and 36.9 nM, respectively. They are 4.5-51.9 times smaller than those reported in a LC-MS based method, and >120 times lower than those obtained by the traditional Ellman method. The results suggested that, the proposed assay significantly increases the sensitivity of commercial AChE. In addition, inhibition efficiencies of three surface waters, a groundwater and four commercial brands of bottled drinking water samples on AChE activity were firstly measured using this UPLC-MS/MS based method. These water samples were proved to have different inhibitory effects on AChE activity, and the inhibition efficiencies dependent on concentrations of dissolved organic carbon (DOC) but are independent of UV absorbance at 254 nm (UV254) values. These results indicate that the proposed method has advantages of high sensitivity over all other conventional methods. It may become a promising AChE inhibition assay for assessing toxicity of aqueous solution containing neurotoxicity contaminants such as organophosphorus pesticides (OPPs) at low levels, or used to evaluate potential inhibition effects of natural waters on AChE activity.

Learning a Model-Driven Variational Network for Deformable Image Registration.

Data-driven deep learning approaches to image registration can be less accurate than conventional iterative approaches, especially when training data is limited. To address this issue and meanwhile retain the fast inference speed of deep learning, we propose VR-Net, a novel cascaded variational network for unsupervised deformable image registration. Using a variable splitting optimization scheme, we first convert the image registration problem, established in a generic variational framework, into two sub-problems, one with a point-wise, closed-form solution and the other one being a denoising problem. We then propose two neural layers (i.e. warping layer and intensity consistency layer) to model the analytical solution and a residual U-Net (termed generalized denoising layer) to formulate the denoising problem. Finally, we cascade the three neural layers multiple times to form our VR-Net. Extensive experiments on three (two 2D and one 3D) cardiac magnetic resonance imaging datasets show that VR-Net outperforms state-of-the-art deep learning methods on registration accuracy, whilst maintaining the fast inference speed of deep learning and the data-efficiency of variational models.

Phenotypic Expression and Outcomes in Individuals With Rare Genetic Variants of Hypertrophic Cardiomyopathy.

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is caused by rare variants in sarcomere-encoding genes, but little is known about the clinical significance of these variants in the general population. OBJECTIVES: The goal of this study was to compare lifetime outcomes and cardiovascular phenotypes according to the presence of rare variants in sarcomere-encoding genes among middle-aged adults. METHODS: This study analyzed whole exome sequencing and cardiac magnetic resonance imaging in UK Biobank participants stratified according to sarcomere-encoding variant status. RESULTS: The prevalence of rare variants (allele frequency <0.00004) in HCM-associated sarcomere-encoding genes in 200,584 participants was 2.9% (n = 5,712; 1 in 35), and the prevalence of variants pathogenic or likely pathogenic for HCM (SARC-HCM-P/LP) was 0.25% (n = 493; 1 in 407). SARC-HCM-P/LP variants were associated with an increased risk of death or major adverse cardiac events compared with controls (hazard ratio: 1.69; 95% confidence interval [CI]: 1.38-2.07; P < 0.001), mainly due to heart failure endpoints (hazard ratio: 4.23; 95% CI: 3.07-5.83; P < 0.001). In 21,322 participants with both cardiac magnetic resonance imaging and whole exome sequencing, SARC-HCM-P/LP variants were associated with an asymmetric increase in left ventricular maximum wall thickness (10.9 ± 2.7 mm vs 9.4 ± 1.6 mm; P < 0.001), but hypertrophy (≥13 mm) was only present in 18.4% (n = 9 of 49; 95% CI: 9%-32%). SARC-HCM-P/LP variants were still associated with heart failure after adjustment for wall thickness (hazard ratio: 6.74; 95% CI: 2.43-18.7; P < 0.001). CONCLUSIONS: In this population of middle-aged adults, SARC-HCM-P/LP variants have low aggregate penetrance for overt HCM but are associated with an increased risk of adverse cardiovascular outcomes and an attenuated cardiomyopathic phenotype. Although absolute event rates are low, identification of these variants may enhance risk stratification beyond familial disease.

Redefining ß-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis.

BACKGROUND: Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of ß-blocker efficacy in patients with sinus rhythm and atrial fibrillation. METHODS: Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of ß blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). FINDINGS: 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from ß blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of ß blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with ß blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p<0·0001 vs random), and cluster membership was externally validated across the nine independent trials. INTERPRETATION: An artificial intelligence-based clustering approach was able to distinguish prognostic response from ß blockers in patients with heart failure and reduced LVEF. This included patients in sinus rhythm with suboptimal efficacy, as well as a cluster of patients with atrial fibrillation where ß blockers did reduce mortality. FUNDING: Medical Research Council, UK, and EU/EFPIA Innovative Medicines Initiative BigData@Heart.

Complementary time-frequency domain networks for dynamic parallel MR image reconstruction.

PURPOSE: To introduce a novel deep learning-based approach for fast and high-quality dynamic multicoil MR reconstruction by learning a complementary time-frequency domain network that exploits spatiotemporal correlations simultaneously from complementary domains. THEORY AND METHODS: Dynamic parallel MR image reconstruction is formulated as a multivariable minimization problem, where the data are regularized in combined temporal Fourier and spatial (x-f) domain as well as in spatiotemporal image (x-t) domain. An iterative algorithm based on variable splitting technique is derived, which alternates among signal de-aliasing steps in x-f and x-t spaces, a closed-form point-wise data consistency step and a weighted coupling step. The iterative model is embedded into a deep recurrent neural network which learns to recover the image via exploiting spatiotemporal redundancies in complementary domains. RESULTS: Experiments were performed on two datasets of highly undersampled multicoil short-axis cardiac cine MRI scans. Results demonstrate that our proposed method outperforms the current state-of-the-art approaches both quantitatively and qualitatively. The proposed model can also generalize well to data acquired from a different scanner and data with pathologies that were not seen in the training set. CONCLUSION: The work shows the benefit of reconstructing dynamic parallel MRI in complementary time-frequency domains with deep neural networks. The method can effectively and robustly reconstruct high-quality images from highly undersampled dynamic multicoil data ( 16× and 24× yielding 15 s and 10 s scan times respectively) with fast reconstruction speed (2.8 seconds). This could potentially facilitate achieving fast single-breath-hold clinical 2D cardiac cine imaging.

Systematic evaluation of iterative deep neural networks for fast parallel MRI reconstruction with sensitivity-weighted coil combination.

PURPOSE: To systematically investigate the influence of various data consistency layers and regularization networks with respect to variations in the training and test data domain, for sensitivity-encoded accelerated parallel MR image reconstruction. THEORY AND METHODS: Magnetic resonance (MR) image reconstruction is formulated as a learned unrolled optimization scheme with a down-up network as regularization and varying data consistency layers. The proposed networks are compared to other state-of-the-art approaches on the publicly available fastMRI knee and neuro dataset and tested for stability across different training configurations regarding anatomy and number of training samples. RESULTS: Data consistency layers and expressive regularization networks, such as the proposed down-up networks, form the cornerstone for robust MR image reconstruction. Physics-based reconstruction networks outperform post-processing methods substantially for R = 4 in all cases and for R = 8 when the training and test data are aligned. At R = 8, aligning training and test data is more important than architectural choices. CONCLUSION: In this work, we study how dataset sizes affect single-anatomy and cross-anatomy training of neural networks for MRI reconstruction. The study provides insights into the robustness, properties, and acceleration limits of state-of-the-art networks, and our proposed down-up networks. These key insights provide essential aspects to successfully translate learning-based MRI reconstruction to clinical practice, where we are confronted with limited datasets and various imaged anatomies.

Adaptive Weighting of Handcrafted Feature Losses for Facial Expression Recognition.

Due to the importance of facial expressions in human-machine interaction, a number of handcrafted features and deep neural networks have been developed for facial expression recognition. While a few studies have shown the similarity between the handcrafted features and the features learned by deep network, a new feature loss is proposed to use feature bias constraint of handcrafted and deep features to guide the deep feature learning during the early training of network. The feature maps learned with and without the proposed feature loss for a toy network suggest that our approach can fully explore the complementarity between handcrafted features and deep features. Based on the feature loss, a general framework for embedding the traditional feature information into deep network training was developed and tested using the FER2013, CK+, Oulu-CASIA, and MMI datasets. Moreover, adaptive loss weighting strategies are proposed to balance the influence of different losses for different expression databases. The experimental results show that the proposed feature loss with adaptive weighting achieves much better accuracy than the original handcrafted feature and the network trained without using our feature loss. Meanwhile, the feature loss with adaptive weighting can provide complementary information to compensate for the deficiency of a single feature.

Deep Learning for Cardiac Image Segmentation: A Review.

Deep learning has become the most widely used approach for cardiac image segmentation in recent years. In this paper, we provide a review of over 100 cardiac image segmentation papers using deep learning, which covers common imaging modalities including magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound and major anatomical structures of interest (ventricles, atria, and vessels). In addition, a summary of publicly available cardiac image datasets and code repositories are included to provide a base for encouraging reproducible research. Finally, we discuss the challenges and limitations with current deep learning-based approaches (scarcity of labels, model generalizability across different domains, interpretability) and suggest potential directions for future research.

Paradoxical Higher Myocardial Wall Stress and Increased Cardiac Remodeling Despite Lower Mass in Females.

Background Increased left ventricular (LV) mass is characterized by increased myocardial wall thickness and/or ventricular dilatation that is associated with worse outcomes. We aim to comprehensively compare sex-stratified associations between measures of LV remodeling and increasing LV mass in the general population. Methods and Results Participants were prospectively recruited in the National Heart Center Singapore Biobank to examine health and cardiovascular risk factors in the general population. Cardiovascular magnetic resonance was performed in all individuals. Participants with established cardiovascular diseases and abnormal cardiovascular magnetic resonance scan results were excluded. Global and regional measures of LV remodeling (geometry, function, interstitial volumes, and wall stress) were performed using conventional image analysis and novel 3-dimensional machine learning phenotyping. Sex-stratified analyses were performed in 1005 participants (57% males; 53±13 years). Age and prevalence of cardiovascular risk factors were well-matched in both sexes (P>0.05 for all). Progressive increase in LV mass was associated with increased concentricity in either sex, but to a greater extent in females. Compared with males, females had higher wall stress (mean difference: 170 mm Hg, P<0.0001) despite smaller LV mass (42.4±8.2 versus 55.6±14.2 g/m2, P<0.0001), lower blood pressures (P<0.0001), and higher LV ejection fraction (61.9±5.9% versus 58.6±6.4%, P<0.0001). The regions of increased concentric remodeling corresponded to regions of increased wall stress. Compared with males, females had increased extracellular volume fraction (27.1±2.4% versus 25.1±2.9%, P<0.0001). Conclusions Compared with males, females have lower LV mass, increased wall stress, and concentric remodeling. These findings provide mechanistic insights that females are susceptible to particular cardiovascular complications.