Search

Coronary artery disease (CAD) is a cardiovascular condition with high morbidity and mortality. Intravascular optical coherence tomography (IVOCT) has been considered as an optimal imagining system for the diagnosis and treatment of CAD. Constrained by Nyquist theorem, dense sampling in IVOCT attains high resolving power to delineate cellular structures/features. There is a trade-off between high spatial resolution and fast scanning rate for coronary imaging. In this paper, we propose a viable spectral-spatial acquisition method that down-scales the sampling process in both spectral and spatial domain while maintaining high quality in image reconstruction. The down-scaling schedule boosts data acquisition speed without any hardware modifications. Additionally, we propose a unified multi-scale reconstruction framework, namely Multiscale-Spectral-Spatial-Magnification Network (MSSMN), to resolve highly down-scaled (compressed) OCT images with flexible magnification factors. We incorporate the proposed methods into Spectral Domain OCT (SD-OCT) imaging of human coronary samples with clinical features such as stent and calcified lesions. Our experimental results demonstrate that spectral-spatial down-scaled data can be better reconstructed than data that are down-scaled solely in either spectral or spatial domain. Moreover, we observe better reconstruction performance using MSSMN than using existing reconstruction methods. Our acquisition method and multi-scale reconstruction framework, in combination, may allow faster SD-OCT inspection with high resolution during coronary intervention. [ABSTRACT FROM AUTHOR]

Modern sleep monitoring development is shifting towards the use of unobtrusive sensors combined with algorithms for automatic sleep scoring. Many different combinations of wet and dry electrodes, ear-centered, forehead-mounted or headband-inspired designs have been proposed, alongside an ever growing variety of machine learning algorithms for automatic sleep scoring. Objective: Among candidate positions, those in the facial area and around the ears have the benefit of being relatively hairless, and in our view deserve extra attention. In this paper, we seek to determine the limits to sleep monitoring quality within this spatial constraint. Methods: We compare 13 different, realistic sensor setups derived from the same data set and analysed with the same pipeline. Results: All setups which include both a lateral and an EOG derivation show similar, state-of-the-art performance, with average Cohen’s kappa values of at least 0.80. Conclusion: If large electrode distances are used, positioning is not critical for achieving large sleep-related signal-to-noise-ratio, and hence accurate sleep scoring. Significance: We argue that with the current competitive performance of automated staging approaches, there is a need for establishing an improved benchmark beyond current single human rater scoring. [ABSTRACT FROM AUTHOR]

Machine learning techniques used in computer-aided medical image analysis usually suffer from the domain shift problem caused by different distributions between source/reference data and target data. As a promising solution, domain adaptation has attracted considerable attention in recent years. The aim of this paper is to survey the recent advances of domain adaptation methods in medical image analysis. We first present the motivation of introducing domain adaptation techniques to tackle domain heterogeneity issues for medical image analysis. Then we provide a review of recent domain adaptation models in various medical image analysis tasks. We categorize the existing methods into shallow and deep models, and each of them is further divided into supervised, semi-supervised and unsupervised methods. We also provide a brief summary of the benchmark medical image datasets that support current domain adaptation research. This survey will enable researchers to gain a better understanding of the current status, challenges and future directions of this energetic research field. [ABSTRACT FROM AUTHOR]

Significance: The performance of traditional approaches to decoding movement intent from electromyograms (EMGs) and other biological signals commonly degrade over time. Furthermore, conventional algorithms for training neural network based decoders may not perform well outside the domain of the state transitions observed during training. The work presented in this paper mitigates both these problems, resulting in an approach that has the potential to substantially improve the quality of life of the people with limb loss. Objective: This paper presents and evaluates the performance of four decoding methods for volitional movement intent from intramuscular EMG signals. Methods: The decoders are trained using the dataset aggregation (DAgger) algorithm, in which the training dataset is augmented during each training iteration based on the decoded estimates from previous iterations. Four competing decoding methods, namely polynomial Kalman filters (KFs), multilayer perceptron (MLP) networks, convolutional neural networks (CNN), and long short-term memory (LSTM) networks, were developed. The performances of the four decoding methods were evaluated using EMG datasets recorded from two human volunteers with transradial amputation. Short-term analyses, in which the training and cross-validation data came from the same dataset, and long-term analyses, in which the training and testing were done in different datasets, were performed. Results: Short-term analyses of the decoders demonstrated that CNN and MLP decoders performed significantly better than KF and LSTM decoders, showing an improvement of up to 60% in the normalized mean-square decoding error in cross-validation tests. Long-term analyses indicated that the CNN, MLP, and LSTM decoders performed significantly better than a KF-based decoder at most analyzed cases of temporal separations (0–150 days) between the acquisition of the training and testing datasets. Conclusion: The short-term and long-term performances of MLP- and CNN-based decoders trained with DAgger demonstrated their potential to provide more accurate and naturalistic control of prosthetic hands than alternate approaches. [ABSTRACT FROM AUTHOR]

Objective: This paper describes how non-invasive wearable sensors can be used in combination with deep learning to classify artificially induced gait alterations without the requirement for a medical professional or gait analyst to be present. This approach is motivated by the goal of diagnosing gait abnormalities on a symptom-by-symptom basis, irrespective of other neuromuscular movement disorders the patients may be affected by. This could lead to improvements in treatment and offer a greater insight into movement disorders. Methods: In-shoe pressure was measured for 12 able-bodied participants, each subject to eight artificially induced gait alterations, achieved by modifying the underside of the shoe. The data were recorded at 100 Hz over 2520 data channels and were analyzed using the deep learning architecture and the long term short term memory networks. Additionally, the rationale for the decision-making process of these networks was investigated. Conclusion: Long term short term memory networks are applicable to the classification of the gait function. The classifications can be made using only 2 s of sparse data (82.0% accuracy over 96 000 instances of test data) from participants who were not a part of the training set. Significance: This paper provides potential for the gait function to be accurately classified using non-invasive techniques, and at more regular intervals, outside of a clinical setting, without the need for healthcare professionals to be present. [ABSTRACT FROM AUTHOR]

Breast histology image classification is a crucial step in the early diagnosis of breast cancer. In breast pathological diagnosis, Convolutional Neural Networks (CNNs) have demonstrated great success using digitized histology slides. However, tissue classification is still challenging due to the high visual variability of the large-sized digitized samples and the lack of contextual information. In this paper, we propose a novel CNN, called Multi-level Context and Uncertainty aware (MCUa) dynamic deep learning ensemble model. MCUa model consists of several multi-level context-aware models to learn the spatial dependency between image patches in a layer-wise fashion. It exploits the high sensitivity to the multi-level contextual information using an uncertainty quantification component to accomplish a novel dynamic ensemble model. MCUa model has achieved a high accuracy of 98.11% on a breast cancer histology image dataset. Experimental results show the superior effectiveness of the proposed solution compared to the state-of-the-art histology classification models. [ABSTRACT FROM AUTHOR]

Using deep neural networks (DNNs) to explore spatial patterns and temporal dynamics of human brain activities has been an important yet challenging problem because the artificial neural networks are hard to be designed manually. There have been several promising deep learning methods, e.g., deep belief network (DBN), convolutional neural network (CNN), and deep sparse recurrent auto-encoder (DSRAE), that can decompose neuroscientific and meaningful spatiotemporal patterns from 4D functional Magnetic Resonance Imaging (fMRI) data. However, those previous studies still depend on hand-crafted neural network architectures and hyperparameters, which are not optimal in various senses. In this paper, we employ the evolutionary algorithms (EA) to optimize the deep neural architecture of DSRAE by minimizing the expected loss of initialized models, named eNAS-DSRAE (evolutionary Neural Architecture Search on Deep Sparse Recurrent Auto-Encoder). Also, validation experiments are designed and performed on the publicly available human connectome project (HCP) 900 datasets, and the results achieved by the optimized eNAS-DSRAE suggested that our framework can successfully identify the spatiotemporal features and perform better than the hand-crafted neural network models. To our best knowledge, the proposed eNAS-DSRAE is not only among the earliest NAS models that can extract connectome-scale meaningful spatiotemporal brain networks from 4D fMRI data, but also is an effective framework to optimize the RNN-based models. [ABSTRACT FROM AUTHOR]

Seizure prediction for drug-refractory epilepsy patients can improve their quality of life, reduce their anxiety, and help them take the necessary precautions. Nowadays, numerous deep learning algorithms have been proposed to predict seizure onset and obtain better performance than traditional machine learning methods. However, these methods require a large set of parameters and large hardware resources; they also have high energy consumption. Therefore, these methods cannot be implemented on compact, low-power wearable, or implantable medical devices. The devices should operate on a real-time basis to continually inform the epileptic patients. In this paper, we describe energy-efficient and hardware-friendly methods to predict the epileptic seizures. A model of only 45 kB was obtained by the neural architecture search and was evaluated across three datasets. The overall sensitivity, false prediction rate, and area under receiver operating characteristic curve were 99.81%, 0.005/h, 1 and 93.48%, 0.063/h, 0.977 and 85.19%, 0.116/h, 0.933, respectively, for three public datasets. This model was further reduced with several methods for model compression. The performances for the model sizes less than 50 kB for scalp EEG data and less than 10 kB for intracranial EEG data outperformed all the other models of similar model sizes. In particular, the energy consumption estimation was less than 10 mJ per inference for scalp EEG signal and less than 0.5 mJ per inference for intracranial EEG signal, which meet the requirements for low-power wearable and implantable devices, respectively. [ABSTRACT FROM AUTHOR]

Objective: To propose a deep-learning network for the diagnosis of two corneal diseases: Fuchs’ endothlelial dystrophy and keratoconus, based on optical coherence tomography (OCT) images of the cornea. Methods: In this paper, we propose a novel network with parallel resolution-specific encoders and composite classification features to directly diagnose Fuchs’ endothelial dystrophy and keratoconus using OCT images. Our proposed network consists of a multi-resolution input, multiple parallel encoders, and a composite of convolutional and dense features for classification. The purpose of using parallel resolution-specific encoders is to perform multi-resolution feature fusion. Also, using composite classification features enhances the dense feature learning. We implemented other related networks for comparison with our network and performed k-fold cross-validation on a dataset of 16,721 OCT images. We used saliency maps and sensitivity analysis to visualize our proposed network. Results: The proposed network outperformed other networks with an image classification accuracy of 0.91 and a scan classification accuracy of 0.94. The visualizations show that our network learned better features than other networks. Significance: The proposed methods can potentially be a step towards the early diagnosis of corneal diseases, which is necessary to prevent their progression, hence, prevent loss of vision. [ABSTRACT FROM AUTHOR]

Objective: Multi-modal functional magnetic resonance imaging (fMRI) can be used to make predictions about individual behavioral and cognitive traits based on brain connectivity networks. Methods: To take advantage of complementary information from multi-modal fMRI, we propose an interpretable multi-modal graph convolutional network (MGCN) model, incorporating both fMRI time series and functional connectivity (FC) between each pair of brain regions. Specifically, our model learns a graph embedding from individual brain networks derived from multi-modal data. A manifold-based regularization term is enforced to consider the relationships of subjects both within and between modalities. Furthermore, we propose the gradient-weighted regression activation mapping (Grad-RAM) and the edge mask learning to interpret the model, which is then used to identify significant cognition-related biomarkers. Results: We validate our MGCN model on the Philadelphia Neurodevelopmental Cohort to predict individual wide range achievement test (WRAT) score. Our model obtains superior predictive performance over GCN with a single modality and other competing approaches. The identified biomarkers are cross-validated from different approaches. Conclusion and Significance: This paper develops a new interpretable graph deep learning framework for cognition prediction, with the potential to overcome the limitations of several current data-fusion models. The results demonstrate the power of MGCN in analyzing multi-modal fMRI and discovering significant biomarkers for human brain studies. [ABSTRACT FROM AUTHOR]

Objective: Glaucoma is the second leading cause of blindness worldwide. Glaucomatous progression can be easily monitored by analyzing the degeneration of retinal ganglion cells (RGCs). Many researchers have screened glaucoma by measuring cup-to-disc ratios from fundus and optical coherence tomography scans. However, this paper presents a novel strategy that pays attention to the RGC atrophy for screening glaucomatous pathologies and grading their severity. Methods: The proposed framework encompasses a hybrid convolutional network that extracts the retinal nerve fiber layer, ganglion cell with the inner plexiform layer and ganglion cell complex regions, allowing thus a quantitative screening of glaucomatous subjects. Furthermore, the severity of glaucoma in screened cases is objectively graded by analyzing the thickness of these regions. Results: The proposed framework is rigorously tested on publicly available Armed Forces Institute of Ophthalmology (AFIO) dataset, where it achieved the F1 score of 0.9577 for diagnosing glaucoma, a mean dice coefficient score of 0.8697 for extracting the RGC regions and an accuracy of 0.9117 for grading glaucomatous progression. Furthermore, the performance of the proposed framework is clinically verified with the markings of four expert ophthalmologists, achieving a statistically significant Pearson correlation coefficient of 0.9236. Conclusion: An automated assessment of RGC degeneration yields better glaucomatous screening and grading as compared to the state-of-the-art solutions. Significance: An RGC-aware system not only screens glaucoma but can also grade its severity and here we present an end-to-end solution that is thoroughly evaluated on a standardized dataset and is clinically validated for analyzing glaucomatous pathologies. [ABSTRACT FROM AUTHOR]

Arrhythmia detection and classification is a crucial step for diagnosing cardiovascular diseases. However, deep learning models that are commonly used and trained in end-to-end fashion are not able to provide good interpretability. In this paper, we address this deficiency by proposing the first novel interpretable arrhythmia classification approach based on a human-machine collaborative knowledge representation. Our approach first employs an AutoEncoder to encode electrocardiogram signals into two parts: hand-encoding knowledge and machine-encoding knowledge. A classifier then takes as input the encoded knowledge to classify arrhythmia heartbeats with or without human in the loop (HIL). Experiments and evaluation on the MIT-BIH Arrhythmia Database demonstrate that our new approach not only can effectively classify arrhythmia while offering interpretability, but also can improve the classification accuracy by adjusting the hand-encoding knowledge with our HIL mechanism. [ABSTRACT FROM AUTHOR]

Recognizing activities of daily living (ADL) provides vital contextual information that enhances the effectiveness of various mobile health and wellness applications. Development of wearable motion sensors along with machine learning algorithms offer a great opportunity for ADL recognition. However, the performance of the ADL recognition systems may significantly degrade when they are used by a new user due to inter-subject variability. This issue limits the usability of these systems. In this paper, we propose a deep learning assisted personalization framework for ADL recognition with the aim to maximize the personalization performance while minimizing solicitation of inputs or labels from the user to reduce user's burden. The proposed framework consists of unsupervised retraining of automatic feature extraction layers and supervised fine-tuning of classification layers through a novel active learning model based on a given model's uncertainty. We design a Bayesian deep convolutional neural network with stochastic latent variables that allows us to estimate both aleatoric (data-dependent) and epistemic (model-dependent) uncertainties in recognition task. In this study, for the first time, we show how distinguishing between the two aforementioned sources of uncertainty leads to more effective active learning. The experimental results show that our proposed method improves the accuracy of ADL recognition on a new user by 25% on average compared to the case of using a model for a new user with no personalization with an average final accuracy of 89.2%. Moreover, our method achieves higher personalization accuracy while significantly reducing user's burden in terms of soliciting inputs and labels compared to other methods. [ABSTRACT FROM AUTHOR]

Objective: Ultrasound elastography is gaining traction as an accessible and useful diagnostic tool for things such as, cancer detection and differentiation and thyroid disease diagnostics. Unfortunately, state-of-the-art shear wave imaging techniques, essential to promote this goal, are limited to high-end ultrasound hardware due to high-power requirements, and are extremely sensitive to patient and sonographer motion, and generally suffer from low frame rates. Motivated by research and theory showing that longitudinal wave sound speed carries similar diagnostic abilities to shear wave imaging, we present an alternative approach using single-sided pressure-wave sound speed measurements from channel data. Methods: In this paper, we present a single-sided sound speed inversion solution using a fully convolutional deep neural network. We use simulations for training, allowing the generation of limitless ground truth data. Results: We show that it is possible to invert for longitudinal sound speed in soft tissue at high frame rates. We validate the method on simulated data. We present highly encouraging results on limited real data. Conclusion: Sound speed inversion on channel data has made significant potential possible in real time with deep learning technologies. Significance: Specialized shear wave ultrasound systems remain inaccessible in many locations. Longitudinal sound speed and deep learning technologies enable an alternative approach to diagnosis based on tissue elasticity. High frame rates are also possible. [ABSTRACT FROM AUTHOR]

Objective: This paper addresses two key problems of skin lesion classification. The first problem is the effective use of high-resolution images with pretrained standard architectures for image classification. The second problem is the high-class imbalance encountered in real-world multi-class datasets. Methods: To use high-resolution images, we propose a novel patch-based attention architecture that provides global context between small, high-resolution patches. We modify three pretrained architectures and study the performance of patch-based attention. To counter class imbalance problems, we compare oversampling, balanced batch sampling, and class-specific loss weighting. Additionally, we propose a novel diagnosis-guided loss weighting method that takes the method used for ground-truth annotation into account. Results: Our patch-based attention mechanism outperforms previous methods and improves the mean sensitivity by $\text{7}\%$. Class balancing significantly improves the mean sensitivity and we show that our diagnosis-guided loss weighting method improves the mean sensitivity by $\text{3}\%$ over normal loss balancing. Conclusion: The novel patch-based attention mechanism can be integrated into pretrained architectures and provides global context between local patches while outperforming other patch-based methods. Hence, pretrained architectures can be readily used with high-resolution images without downsampling. The new diagnosis-guided loss weighting method outperforms other methods and allows for effective training when facing class imbalance. Significance: The proposed methods improve automatic skin lesion classification. They can be extended to other clinical applications where high-resolution image data and class imbalance are relevant. [ABSTRACT FROM AUTHOR]

Currently available risk prediction methods are limited in their ability to deal with complex, heterogeneous, and longitudinal data such as that available in primary care records, or in their ability to deal with multiple competing risks. This paper develops a novel deep learning approach that is able to successfully address current limitations of standard statistical approaches such as landmarking and joint modeling. Our approach, which we call Dynamic-DeepHit, flexibly incorporates the available longitudinal data comprising various repeated measurements (rather than only the last available measurements) in order to issue dynamically updated survival predictions for one or multiple competing risk(s). Dynamic-DeepHit learns the time-to-event distributions without the need to make any assumptions about the underlying stochastic models for the longitudinal and the time-to-event processes. Thus, unlike existing works in statistics, our method is able to learn data-driven associations between the longitudinal data and the various associated risks without underlying model specifications. We demonstrate the power of our approach by applying it to a real-world longitudinal dataset from the U.K. Cystic Fibrosis Registry, which includes a heterogeneous cohort of 5883 adult patients with annual follow-ups between 2009 to 2015. The results show that Dynamic-DeepHit provides a drastic improvement in discriminating individual risks of different forms of failures due to cystic fibrosis. Furthermore, our analysis utilizes post-processing statistics that provide clinical insight by measuring the influence of each covariate on risk predictions and the temporal importance of longitudinal measurements, thereby enabling us to identify covariates that are influential for different competing risks. [ABSTRACT FROM AUTHOR]

Objective: Multi-modal functional magnetic resonance imaging has been widely used for brain research. Conventional data-fusion methods cannot capture complex relationship (e.g., nonlinear predictive relationship) between multiple data. This paper aims to develop a neural network framework to extract phenotype related cross-data relationships and use it to study the brain development. Methods: We propose a novel method, deep collaborative learning (DCL), to address the limitation of existing methods. DCL first uses a deep network to represent original data and then seeks their correlations, while also linking the data representation with phenotypical information. Results: We studied the difference of functional connectivity (FCs) between different age groups and also use FCs as a fingerprint to predict cognitive abilities. Our experiments demonstrated higher accuracy of using DCL over other conventional models when classifying populations of different ages and cognitive scores. Moreover, DCL revealed that brain connections became stronger at adolescence stage. Furthermore, DCL detected strong correlations between default mode network and other networks which were overlooked by linear canonical correlation analysis, demonstrating DCL's ability of detecting nonlinear correlations. Conclusion: The results verified the superiority of DCL over conventional data-fusion methods. In addition, the stronger brain connection demonstrated the importance of adolescence stage for brain development. Significance: DCL can better combine complex correlations between multiple data sets in addition to their fitting to phenotypes, with the potential to overcome the limitations of several current data-fusion models. [ABSTRACT FROM AUTHOR]

In this paper, we present a deep learning framework “Rehab-Net” for effectively classifying three upper limb movements of the human arm, involving extension, flexion, and rotation of the forearm, which, over the time, could provide a measure of rehabilitation progress. The proposed framework, Rehab-Net is formulated with a personalized, light weight and low-complex, customized convolutional neural network (CNN) model, using two-layers of CNN, interleaved with pooling layers, followed by a fully connected layer that classifies the three movements from tri-axial acceleration input data collected from the wrist. The proposed Rehab-Net framework was validated on sensor data collected in two situations: 1) semi-naturalistic environment involving an archetypal activity of “making-tea” with four stroke survivors and 2) natural environment, where ten stroke survivors were free to perform any desired arm movement for the duration of 120 min. We achieved an overall accuracy of 97.89% on semi-naturalistic data and 88.87% on naturalistic data which exceeded state-of-the-art learning algorithms namely, linear discriminant analysis, support vector machines, and k-means clustering with an average accuracy of 48.89%, 44.14%, and 27.64%. Subsequently, a computational complexity analysis of the proposed model has been discussed with an eye toward hardware implementation. The clinical significance of this study is to accurately monitor the clinical progress of the rehabilitated subjects under the ambulatory settings. [ABSTRACT FROM AUTHOR]

Gesture recognition using sparse multichannel surface electromyography (sEMG) is a challenging problem, and the solutions are far from optimal from the point of view of muscle–computer interface. In this paper, we address this problem from the context of multi-view deep learning. A novel multi-view convolutional neural network (CNN) framework is proposed by combining classical sEMG feature sets with a CNN-based deep learning model. The framework consists of two parts. In the first part, multi-view representations of sEMG are modeled in parallel by a multistream CNN, and a performance-based view construction strategy is proposed to choose the most discriminative views from classical feature sets for sEMG-based gesture recognition. In the second part, the learned multi-view deep features are fused through a view aggregation network composed of early and late fusion subnetworks, taking advantage of both early and late fusion of learned multi-view deep features. Evaluations on 11 sparse multichannel sEMG databases as well as five databases with both sEMG and inertial measurement unit data demonstrate that our multi-view framework outperforms single-view methods on both unimodal and multimodal sEMG data streams. [ABSTRACT FROM AUTHOR]

Objective: Deep learning has recently been applied to electrical impedance tomography (EIT) imaging. Nevertheless, there are still many challenges that this approach has to face, e.g., targets with sharp corners or edges cannot be well recovered when using circular inclusion training data. This paper proposes an iterative-based inversion method and a convolutional neural network (CNN) based inversion method to recover some challenging inclusions such as triangular, rectangular, or lung shapes, where the CNN-based method uses only random circle or ellipse training data. Methods: First, the iterative method, i.e., bases-expansion subspace optimization method (BE-SOM), is proposed based on a concept of induced contrast current (ICC) with total variation regularization. Second, the theoretical analysis of BE-SOM and the physical concepts introduced there motivate us to propose a dominant-current deep learning scheme for EIT imaging, in which dominant parts of ICC are utilized to generate multi-channel inputs of CNN. Results: The proposed methods are tested with both numerical and experimental data, where several realistic phantoms including simulated pneumothorax and pleural effusion pathologies are also considered. Conclusions and Significance: Significant performance improvements of the proposed methods are shown in reconstructing targets with sharp corners or edges. It is also demonstrated that the proposed methods are capable of fast, stable, and high-quality EIT imaging, which is promising in providing quantitative images for potential clinical applications. [ABSTRACT FROM AUTHOR]

T1-weighted image (T1WI) and T2-weighted image (T2WI) are the two routinely acquired magnetic resonance (MR) modalities that can provide complementary information for clinical and research usages. However, the relatively long acquisition time makes the acquired image vulnerable to motion artifacts. To speed up the imaging process, various algorithms have been proposed to reconstruct high-quality images from under-sampled k-space data. However, most of the existing algorithms only rely on mono-modality acquisition for the image reconstruction. In this paper, we propose to combine complementary MR acquisitions (i.e., T1WI and under-sampled T2WI particularly) to reconstruct the high-quality image (i.e., corresponding to the fully sampled T2WI). To the best of our knowledge, this is the first work to fuse multi-modal MR acquisitions through deep learning to speed up the reconstruction of a certain target image. Specifically, we present a novel deep learning approach, namely Dense-Unet, to accomplish the reconstruction task. The proposed Dense-Unet requires fewer parameters and less computation, while achieving promising performance. Our results have shown that Dense-Unet can reconstruct a three-dimensional T2WI volume in less than 10 s with an under-sampling rate of 8 for the k-space and negligible aliasing artifacts or signal-noise-ratio loss. Experiments also demonstrate excellent transferring capability of Dense-Unet when applied to the datasets acquired by different MR scanners. The above-mentioned results imply great potential of our method in many clinical scenarios. [ABSTRACT FROM AUTHOR]

Correctly identifying sleep stages is important in diagnosing and treating sleep disorders. This paper proposes a joint classification-and-prediction framework based on convolutional neural networks (CNNs) for automatic sleep staging, and, subsequently, introduces a simple yet efficient CNN architecture to power the framework. Given a single input epoch, the novel framework jointly determines its label (classification) and its neighboring epochs’ labels (prediction) in the contextual output. While the proposed framework is orthogonal to the widely adopted classification schemes, which take one or multiple epochs as contextual inputs and produce a single classification decision on the target epoch, we demonstrate its advantages in several ways. First, it leverages the dependency among consecutive sleep epochs while surpassing the problems experienced with the common classification schemes. Second, even with a single model, the framework has the capacity to produce multiple decisions, which are essential in obtaining a good performance as in ensemble-of-models methods, with very little induced computational overhead. Probabilistic aggregation techniques are then proposed to leverage the availability of multiple decisions. To illustrate the efficacy of the proposed framework, we conducted experiments on two public datasets: Sleep-EDF Expanded (Sleep-EDF), which consists of 20 subjects, and Montreal Archive of Sleep Studies (MASS) dataset, which consists of 200 subjects. The proposed framework yields an overall classification accuracy of 82.3% and 83.6%, respectively. We also show that the proposed framework not only is superior to the baselines based on the common classification schemes but also outperforms existing deep-learning approaches. To our knowledge, this is the first work going beyond the standard single-output classification to consider multitask neural networks for automatic sleep staging. This framework provides avenues for further studies of different neural-network architectures for automatic sleep staging. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel framework for dermoscopy image recognition via both a deep learning method and a local descriptor encoding strategy. Specifically, deep representations of a rescaled dermoscopy image are first extracted via a very deep residual neural network pretrained on a large natural image dataset. Then these local deep descriptors are aggregated by orderless visual statistic features based on Fisher vector (FV) encoding to build a global image representation. Finally, the FV encoded representations are used to classify melanoma images using a support vector machine with a Chi-squared kernel. Our proposed method is capable of generating more discriminative features to deal with large variations within melanoma classes, as well as small variations between melanoma and nonmelanoma classes with limited training data. Extensive experiments are performed to demonstrate the effectiveness of our proposed method. Comparisons with state-of-the-art methods show the superiority of our method using the publicly available ISBI 2016 Skin lesion challenge dataset. [ABSTRACT FROM AUTHOR]

Objective: Early diagnosis of acute renal transplant rejection (ARTR) is critical for accurate treatment. Although the current gold standard, diagnostic technique is renal biopsy, it is not preferred due to its invasiveness, long recovery time (1–2 weeks), and potential for complications, e.g., bleeding and/or infection. Methods: This paper presents a computer-aided diagnostic (CAD) system for early ARTR detection using (3D + $b$ -value) diffusion-weighted (DW) magnetic resonance imaging (MRI) data. The CAD process starts from kidney tissue segmentation with an evolving geometric (level-set-based) deformable model. The evolution is guided by a voxel-wise stochastic speed function, which follows from a joint kidney-background Markov–Gibbs random field model accounting for an adaptive kidney shape prior and on-going kidney-background visual appearances. A B-spline-based three-dimensional data alignment is employed to handle local deviations due to breathing and heart beating. Then, empirical cumulative distribution functions of apparent diffusion coefficients of the segmented DW-MRI at different $b$ -values are collected as discriminatory transplant status features. Finally, a deep-learning-based classifier with stacked nonnegative constrained autoencoders is employed to distinguish between rejected and nonrejected renal transplants. Results: In our initial “leave-one-subject-out” experiment on 100 subjects, $\text{97.0}\%$ of the subjects were correctly classified. The subsequent four-fold and ten-fold cross-validations gave the average accuracy of $\text{96.0}\%$ and $\text{94.0}\%$ , respectively. Conclusion: These results demonstrate the promise of this new CAD system to reliably diagnose renal transplant rejection. Significance: The technology presented here can significantly impact the quality of care of renal transplant patients since it has the potential to replace the gold standard in kidney diagnosis, biopsy. [ABSTRACT FROM AUTHOR]

Medical imaging plays a critical role in various clinical applications. However, due to multiple considerations such as cost and radiation dose, the acquisition of certain image modalities may be limited. Thus, medical image synthesis can be of great benefit by estimating a desired imaging modality without incurring an actual scan. In this paper, we propose a generative adversarial approach to address this challenging problem. Specifically, we train a fully convolutional network (FCN) to generate a target image given a source image. To better model a nonlinear mapping from source to target and to produce more realistic target images, we propose to use the adversarial learning strategy to better model the FCN. Moreover, the FCN is designed to incorporate an image-gradient-difference-based loss function to avoid generating blurry target images. Long-term residual unit is also explored to help the training of the network. We further apply Auto-Context Model to implement a context-aware deep convolutional adversarial network. Experimental results show that our method is accurate and robust for synthesizing target images from the corresponding source images. In particular, we evaluate our method on three datasets, to address the tasks of generating CT from MRI and generating 7T MRI from 3T MRI images. Our method outperforms the state-of-the-art methods under comparison in all datasets and tasks. [ABSTRACT FROM AUTHOR]

Goal: Automatic segmentation of brainstem gliomas and prediction of genotype (H3 K27M) mutation status based on magnetic resonance (MR) images are crucial but challenging tasks for computer-aided diagnosis in neurosurgery. In this paper, we present a novel cascaded deep convolutional neural network (CNN) to address these two challenging tasks simultaneously. Methods: Our novel segmentation task contains two feature-fusion modules: the Gaussian-pyramid multiscale input features-fusion technique and the brainstem-region feature enhancement. The aim is to resolve very difficult problems in brainstem glioma segmentation. Our prediction model combines CNN features and support-vector-machine classifier to automatically predict genotypes without region-of-interest labeled-MR images and is learned jointly with the segmentation task. First, Gaussian-pyramid multiscale input feature fusion is added to our glioma-segmentation task to solve the problems of size variety and weak brainstem-gliomas boundaries. Second, the two feature-fusion modules provide both local and global contexts to retain higher frequency details for sharper tumor boundaries, handling the problem of the large variation of tumor shape, and volume resolution. Results and Conclusion: Experiments demonstrate that our cascaded CNN method achieves not only a good tumor segmentation result with a high Dice similarity coefficient of 77.03%, but also a competitive genotype prediction result with an average accuracy of 94.85% upon fivefold cross-validation. [ABSTRACT FROM AUTHOR]

Objective: In this paper, we accurately detect the state-sequence first heart sound (S1)–systole–second heart sound (S2)–diastole, i.e., the positions of S1 and S2, in heart sound recordings. We propose an event detection approach without explicitly incorporating a priori information of the state duration. This renders it also applicable to recordings with cardiac arrhythmia and extendable to the detection of extra heart sounds (third and fourth heart sound), heart murmurs, as well as other acoustic events. Methods: We use data from the 2016 PhysioNet/CinC Challenge, containing heart sound recordings and annotations of the heart sound states. From the recordings, we extract spectral and envelope features and investigate the performance of different deep recurrent neural network (DRNN) architectures to detect the state sequence. We use virtual adversarial training, dropout, and data augmentation for regularization. Results: We compare our results with the state-of-the-art method and achieve an average score for the four events of the state sequence of ${\bf F}_{1}\approx 96$ % on an independent test set. Conclusion: Our approach shows state-of-the-art performance carefully evaluated on the 2016 PhysioNet/CinC Challenge dataset. Significance: In this work, we introduce a new methodology for the segmentation of heart sounds, suggesting an event detection approach with DRNNs using spectral or envelope features. [ABSTRACT FROM AUTHOR]

Objective: Deep learning based methods for retinal vessel segmentation are usually trained based on pixel-wise losses, which treat all vessel pixels with equal importance in pixel-to-pixel matching between a predicted probability map and the corresponding manually annotated segmentation. However, due to the highly imbalanced pixel ratio between thick and thin vessels in fundus images, a pixel-wise loss would limit deep learning models to learn features for accurate segmentation of thin vessels, which is an important task for clinical diagnosis of eye-related diseases. Methods: In this paper, we propose a new segment-level loss which emphasizes more on the thickness consistency of thin vessels in the training process. By jointly adopting both the segment-level and the pixel-wise losses, the importance between thick and thin vessels in the loss calculation would be more balanced. As a result, more effective features can be learned for vessel segmentation without increasing the overall model complexity. Results: Experimental results on public data sets demonstrate that the model trained by the joint losses outperforms the current state-of-the-art methods in both separate-training and cross-training evaluations. Conclusion: Compared to the pixel-wise loss, utilizing the proposed joint-loss framework is able to learn more distinguishable features for vessel segmentation. In addition, the segment-level loss can bring consistent performance improvement for both deep and shallow network architectures. Significance: The findings from this study of using joint losses can be applied to other deep learning models for performance improvement without significantly changing the network architectures. [ABSTRACT FROM AUTHOR]

Objective: This paper investigates the hypothesis that focal seizures can be predicted using scalp electroencephalogram (EEG) data. Our first aim is to learn features that distinguish between the interictal and preictal regions. The second aim is to define a prediction horizon in which the prediction is as accurate and as early as possible, clearly two competing objectives. Methods: Convolutional filters on the wavelet transformation of the EEG signal are used to define and learn quantitative signatures for each period: interictal, preictal, and ictal. The optimal seizure prediction horizon is also learned from the data as opposed to making an a priori assumption. Results: Computational solutions to the optimization problem indicate a 10-min seizure prediction horizon. This result is verified by measuring Kullback–Leibler divergence on the distributions of the automatically extracted features. Conclusion: The results on the EEG database of 204 recordings demonstrate that (i) the preictal phase transition occurs approximately ten minutes before seizure onset, and (ii) the prediction results on the test set are promising, with a sensitivity of 87.8% and a low false prediction rate of 0.142 FP/h. Our results significantly outperform a random predictor and other seizure prediction algorithms. Significance: We demonstrate that a robust set of features can be learned from scalp EEG that characterize the preictal state of focal seizures. [ABSTRACT FROM AUTHOR]

Objective: False positive reduction is one of the most crucial components in an automated pulmonary nodule detection system, which plays an important role in lung cancer diagnosis and early treatment. The objective of this paper is to effectively address the challenges in this task and therefore to accurately discriminate the true nodules from a large number of candidates. Methods: We propose a novel method employing three-dimensional (3-D) convolutional neural networks (CNNs) for false positive reduction in automated pulmonary nodule detection from volumetric computed tomography (CT) scans. Compared with its 2-D counterparts, the 3-D CNNs can encode richer spatial information and extract more representative features via their hierarchical architecture trained with 3-D samples. More importantly, we further propose a simple yet effective strategy to encode multilevel contextual information to meet the challenges coming with the large variations and hard mimics of pulmonary nodules. Results: The proposed framework has been extensively validated in the LUNA16 challenge held in conjunction with ISBI 2016, where we achieved the highest competition performance metric (CPM) score in the false positive reduction track. Conclusion: Experimental results demonstrated the importance and effectiveness of integrating multilevel contextual information into 3-D CNN framework for automated pulmonary nodule detection in volumetric CT data. Significance: While our method is tailored for pulmonary nodule detection, the proposed framework is general and can be easily extended to many other 3-D object detection tasks from volumetric medical images, where the targeting objects have large variations and are accompanied by a number of hard mimics. [ABSTRACT FROM AUTHOR]

In this paper, a multiscale convolutional network (MSCN) and graph-partitioning-based method is proposed for accurate segmentation of cervical cytoplasm and nuclei. Specifically, deep learning via the MSCN is explored to extract scale invariant features, and then, segment regions centered at each pixel. The coarse segmentation is refined by an automated graph partitioning method based on the pretrained feature. The texture, shape, and contextual information of the target objects are learned to localize the appearance of distinctive boundary, which is also explored to generate markers to split the touching nuclei. For further refinement of the segmentation, a coarse-to-fine nucleus segmentation framework is developed. The computational complexity of the segmentation is reduced by using superpixel instead of raw pixels. Extensive experimental results demonstrate that the proposed cervical nucleus cell segmentation delivers promising results and outperforms existing methods. [ABSTRACT FROM PUBLISHER]