Search

This paper proposes an image-based painterly rendering algorithm for automatically synthesizing an image with color ink diffusion. We suggest a mathematical model with a physical base to simulate the phenomenon of color colloidal ink diffusing into absorbent paper. Our algorithm contains three main parts: a feature extraction phase, a Kubelka-Munk (KM) color mixing phase, and a color ink diffusion synthesis phase. In the feature extraction phase, the information of the reference image is simplified by luminance division and color segmentation. In the color mixing phase, the KM theory is employed to approximate the result when one pigment is painted upon another pigment layer. Then, in the color ink diffusion synthesis phase, the physically-based model that we propose is employed to simulate the result of color ink diffusion in absorbent paper using a texture synthesis technique. Our image-based ink diffusing rendering (IBCIDR) algorithm eliminates the drawback of conventional Chinese ink simulations, which are limited to the black ink domain, and our approach demonstrates that, without using any strokes, a color image can be automatically converted to the diffused ink style with a visually pleasing appearance.

Digital video now plays an important role in supporting more profitable online patient training and counseling, and integration of patient training videos from multiple competitive organizations in the health care network will result in better offerings for patients. However, privacy concerns often prevent multiple competitive organizations from sharing and integrating their patient training videos. In addition, patients with infectious or chronic diseases may not want the online patient training organizations to identify who they are or even which video clips they are interested in. Thus, there is an urgent need to develop more effective techniques to protect both video content privacy and access privacy . In this paper, we have developed a new approach to construct a distributed Hippocratic video database system for supporting more profitable online patient training and counseling. First, a new database modeling approach is developed to support concept-oriented video database organization and assign a degree of privacy of the video content for each database level automatically. Second, a new algorithm is developed to protect the video content privacy at the level of individual video clip by filtering out the privacy-sensitive human objects automatically. In order to integrate the patient training videos from multiple competitive organizations for constructing a centralized video database indexing structure, a privacy-preserving video sharing scheme is developed to support privacy-preserving distributed classifier training and prevent the statistical inferences from the videos that are shared for cross-validation of video classifiers. Our experiments on large-scale video databases have also provided very convincing results.

IEEEWhile the zeitgeist of today prescribes localization as a key to eliminate the excessive transportation for manufacturing firms, firm-level practical and workable guidelines that go beyond maxims are scant. Laden with analytical methods, this paper offers a phronetic, customizable framework based on Lean Six Sigma (LSS). The framework mimics Define-Measure-Analyze-Improve-Control (DMAIC) process. The early phases of the framework, ex-ante, prescribe the use of a variety tools and techniques from Lean Six sigma (LSS) and employ fuzzy analytic hierarchy process (FAHP) to extract the vital few root-causes that may impede localization. The latter phases suggest the use of design of experiments (DOE) to evaluate the root-causes, ex-post. The paper outlines a clear and established procedure and demonstrates each analytical phase using evidence from a two-year real-world application in steel industry.

Available data in machine learning applications is becoming increasingly complex, due to higher dimensionality and difficult classes. There exists a wide variety of approaches to measuring complexity of labeled data, according to class overlap, separability or boundary shapes, as well as group morphology. Many techniques can transform the data in order to find better features, but few focus on specifically reducing data complexity. Most data transformation methods mainly treat the dimensionality aspect, leaving aside the available information within class labels which can be useful when classes are somehow complex. This paper proposes an autoencoder-based approach to complexity reduction, using class labels in order to inform the loss function about the adequacy of the generated variables. This leads to three different new feature learners, Scorer, Skaler and Slicer. They are based on Fisher's discriminant ratio, the Kullback-Leibler divergence and least-squares support vector machines, respectively. They can be applied as a preprocessing stage for a binary classification problem. A thorough experimentation across a collection of 27 datasets and a range of complexity and classification metrics shows that class-informed autoencoders perform better than 4 other popular unsupervised feature extraction techniques, especially when the final objective is using the data for a classification task., This paper has been accepted for publication by IEEE Transactions on Pattern Analysis and Machine Intelligence

In this paper, we consider the problem of descriptors construction for the task of content-based image retrieval using deep neural networks. The idea of neural codes, based on fully connected layers activations, is extended by incorporating the information contained in convolutional layers. It is known that the total number of neurons in the convolutional part of the network is large and the majority of them have little influence on the final classification decision. Therefore, in the paper we propose a novel algorithm that allows us to extract the most significant neuron activations and utilize this information to construct effective descriptors. The descriptors consisting of values taken from both the fully connected and convolutional layers perfectly represent the whole image content. The images retrieved using these descriptors match semantically very well to the query image, and also they are similar in other secondary image characteristics, like background, textures or color distribution. These features of the proposed descriptors are verified experimentally based on the IMAGENET1M dataset using the VGG16 neural network., Comment: 8

Deep learning models have been shown to be vulnerable to adversarial attacks. Adversarial attacks are imperceptible perturbations added to an image such that the deep learning model misclassifies the image with a high confidence. Existing adversarial defenses validate their performance using only the classification accuracy. However, classification accuracy by itself is not a reliable metric to determine if the resulting image is "adversarial-free". This is a foundational problem for online image recognition applications where the ground-truth of the incoming image is not known and hence we cannot compute the accuracy of the classifier or validate if the image is "adversarial-free" or not. This paper proposes a novel privacy preserving framework for defending Black box classifiers from adversarial attacks using an ensemble of iterative adversarial image purifiers whose performance is continuously validated in a loop using Bayesian uncertainties. The proposed approach can convert a single-step black box adversarial defense into an iterative defense and proposes three novel privacy preserving Knowledge Distillation (KD) approaches that use prior meta-information from various datasets to mimic the performance of the Black box classifier. Additionally, this paper proves the existence of an optimal distribution for the purified images that can reach a theoretical lower bound, beyond which the image can no longer be purified. Experimental results on six public benchmark datasets namely: 1) Fashion-MNIST, 2) CIFAR-10, 3) GTSRB, 4) MIO-TCD, 5) Tiny-ImageNet, and 6) MS-Celeb show that the proposed approach can consistently detect adversarial examples and purify or reject them against a variety of adversarial attacks.

Reinforcement learning (RL) has demonstrated its ability to solve high dimensional tasks by leveraging non-linear function approximators. However, these successes are mostly achieved by 'black-box' policies in simulated domains. When deploying RL to the real world, several concerns regarding the use of a 'black-box' policy might be raised. In order to make the learned policies more transparent, we propose in this paper a policy iteration scheme that retains a complex function approximator for its internal value predictions but constrains the policy to have a concise, hierarchical, and human-readable structure, based on a mixture of interpretable experts. Each expert selects a primitive action according to a distance to a prototypical state. A key design decision to keep such experts interpretable is to select the prototypical states from trajectory data. The main technical contribution of the paper is to address the challenges introduced by this non-differentiable prototypical state selection procedure. Experimentally, we show that our proposed algorithm can learn compelling policies on continuous action deep RL benchmarks, matching the performance of neural network based policies, but returning policies that are more amenable to human inspection than neural network or linear-in-feature policies.

This paper serves as a survey of recent advances in large margin training and its theoretical foundations, mostly for (nonlinear) deep neural networks (DNNs) that are probably the most prominent machine learning models for large-scale data in the community over the past decade. We generalize the formulation of classification margins from classical research to latest DNNs, summarize theoretical connections between the margin, network generalization, and robustness, and introduce recent efforts in enlarging the margins for DNNs comprehensively. Since the viewpoint of different methods is discrepant, we categorize them into groups for ease of comparison and discussion in the paper. Hopefully, our discussions and overview inspire new research work in the community that aim to improve the performance of DNNs, and we also point to directions where the large margin principle can be verified to provide theoretical evidence why certain regularizations for DNNs function well in practice. We managed to shorten the paper such that the crucial spirit of large margin learning and related methods are better emphasized., Accepted by TPAMI, 8 pages, 3 figures

Osteosarcoma is the most common malignant osteosarcoma, and most developing countries face great challenges in the diagnosis due to the lack of medical resources. Magnetic resonance imaging (MRI) has always been an important tool for the detection of osteosarcoma, but it is a time-consuming and labor-intensive task for doctors to manually identify MRI images. It is highly subjective and prone to misdiagnosis. Existing computer-aided diagnosis methods of osteosarcoma MRI images focus only on accuracy, ignoring the lack of computing resources in developing countries. In addition, the large amount of redundant and noisy data generated during imaging should also be considered. To alleviate the inefficiency of osteosarcoma diagnosis faced by developing countries, this paper proposed an artificial intelligence multiprocessing scheme for pre-screening, noise reduction, and segmentation of osteosarcoma MRI images. For pre-screening, we propose the Slide Block Filter to remove useless images. Next, we introduced a fast non-local means algorithm using integral images to denoise noisy images. We then segmented the filtered and denoised MRI images using a U-shaped network (ETUNet) embedded with a transformer layer, which enhances the functionality and robustness of the traditional U-shaped architecture. Finally, we further optimized the segmented tumor boundaries using conditional random fields. This paper conducted experiments on more than 70,000 MRI images of osteosarcoma from three hospitals in China. The experimental results show that our proposed methods have good results and better performance in pre-screening, noise reduction, and segmentation.

Single-guide RNA is a guide RNA (gRNA), which guides the insertion or deletion of uridine residues into kinetoplastid during RNA editing. It is a small non-coding RNA that can be combined with pre -mRNA pairing. SgRNA is a critical component of the CRISPR/Cas9 gene knockout system and play an important role in gene editing and gene regulation. It is important to accurately and quickly identify highly on-target activity sgRNAs. Due to its importance, several computational predictors have been proposed to predict sgRNAs on-target activity. All these methods have clearly contributed to the development of this very important field. However, they also have certain limitations. In the paper, we developed a new classifier SgRNA-RF, which extracts the features of nucleic acid composition and structure of on-target activity sgRNA sequence and identified by random forest algorithm. In addition to solving an imbalanced dataset, this paper proposed a new method called CS-Smote. We compared sgRNA-RF with state-of-the-art predictors on the five datasets, and found SgRNA-RF significantly improved the identification accuracy, with accuracies of 0.8636,0.9161,0.894,0.938,0.965,0.77,0.979,0.973, respectively. The user-friendly web server that implements sgRNA-RF is freely available at http://server.malab.cn/sgRNA-RF/.

Accompanied with the rapid increase of the demand for routine examination of leucorrhea, efficiency and accuracy become the primary task. However, in super depth of field (SDoF) system, the problem of automatic detection and localization of cells in leucorrhea micro-images is still a big challenge. The changing of the relative position between the cell center and focus plane of microscope lead to variable cell morphological structure in the two-dimensional image, which is an important reason for the low accuracy of current deep learning target detection algorithms. In this paper, an object detection method based on Retinanet in state of super depth of field is proposed, which can achieve high precision detecting of leucorrhea components by the SDoF feature aggregation module. Compared with the current mainstream algorithms, the mean average accuracy (mAP) index has been improved significantly, the mAP index is 82.7% for SDoF module and 83.0% for SDoF+ module, with an average increase of more than 10%. These improved features can significantly improve the efficiency and accuracy of the algorithm. The algorithm proposed in this paper can be integrated into the leucorrhea automatic detection system.

In Robot Assisted Minimally Invasive Surgery, discriminating critical subsurface structures is essential to make the surgical procedure safer and more efficient. In this paper, a novel robot assisted electrical bio-impedance scanning (RAEIS) system is developed and validated using a series of experiments. The proposed system constructs a tri-polar sensing configuration for tissue homogeneity inspection. Specifically, two robotic forceps are used as electrodes for applying electric current and measuring reciprocal voltages relative to a ground electrode which is placed distal from the measuring site. Compared to the other existing electrical bioimpedance sensing technology, the proposed system is able to use miniaturized electrodes to measure a site flexibly with enhanced subsurfacial detection capability. In this paper, we present the concept, the modeling of the sensing method, the hardware design, and the system calibration. Subsequently, a series of experiments are conducted for system evaluation including finite element simulation, saline solution bath experiments and experiments based on ex vivo animal tissues. The experimental results demonstrate that the proposed system can measure the resistivity of the material with high accuracy, and detect a subsurface non-homogeneous object with 100% success rate. The proposed parameters estimation algorithm is able to approximate the resistivity and the depth of the subsurface object effectively with one fast scanning.

The existing Human-Machine Interfaces (HMI) based on gesture recognition using surface electromyography (sEMG) have made significant progress. However, the sEMG has inherent limitations as well as the gesture classification and force estimation have not been effectively combined. There are limitations in applications such as prosthetic control and clinical rehabilitation, etc. In this paper, a grasping gesture and force recognition strategy based on wearable A-mode ultrasound and two-stage cascade model is proposed, which can simultaneously estimate the force while classifying the grasping gesture. This paper experiments five grasping gestures and four force levels (5-50%MVC). The results demonstrate that the performance of the proposed model is significantly better than that of the traditional model both in classification and regression (p0.001). Additionally, the two-stage cascade regression model (TSCRM) used the Gaussian Process regression model (GPR) with the mean and standard deviation (MSD) feature obtains excellent results, with normalized root-mean-square error (nRMSE) and correlation coefficient (CC) of 0.10490.0374 and 0.94610.0354, respectively. Besides, the latency of the model meets the requirement of real-time recognition (T15ms). Therefore, the research outcomes prove the feasibility of the proposed recognition strategy and provide a reference for the field of prosthetic control, etc.

In recent years, plenty of metrics have been proposed to identify networks that are free of gradient explosion and vanishing. However, due to the diversity of network components and complex serial-parallel hybrid connections in modern DNNs, the evaluation of existing metrics usually requires strong assumptions, complex statistical analysis, or has limited application fields, which constraints their spread in the community. In this paper, inspired by the Gradient Norm Equality and dynamical isometry, we first propose a novel metric called Block Dynamical Isometry, which measures the change of gradient norm in individual block. Because our Block Dynamical Isometry is norm-based, its evaluation needs weaker assumptions compared with the original dynamical isometry. To mitigate the challenging derivation, we propose a highly modularized statistical framework based on free probability. Our framework includes several key theorems to handle complex serial-parallel hybrid connections and a library to cover the diversity of network components. Besides, several sufficient prerequisites are provided. Powered by our metric and framework, we analyze extensive initialization, normalization, and network structures. We find that Gradient Norm Equality is a universal philosophy behind them. Then, we improve some existing methods based on our analysis, including an activation function selection strategy for initialization techniques, a new configuration for weight normalization, and a depth-aware way to derive coefficients in SeLU. Moreover, we propose a novel normalization technique named second moment normalization, which is theoretically 30% faster than batch normalization without accuracy loss. Last but not least, our conclusions and methods are evidenced by extensive experiments on multiple models over CIFAR10 and ImageNet., Comment: Under review as a regular paper in TPAMI

Three-dimensional Morphable Models (3DMMs) are powerful statistical tools for representing the 3D shapes and textures of an object class. Here we present the most complete 3DMM of the human head to date that includes face, cranium, ears, eyes, teeth and tongue. To achieve this, we propose two methods for combining existing 3DMMs of different overlapping head parts: i. use a regressor to complete missing parts of one model using the other, ii. use the Gaussian Process framework to blend covariance matrices from multiple models. Thus we build a new combined face-and-head shape model that blends the variability and facial detail of an existing face model (the LSFM) with the full head modelling capability of an existing head model (the LYHM). Then we construct and fuse a highly-detailed ear model to extend the variation of the ear shape. Eye and eye region models are incorporated into the head model, along with basic models of the teeth, tongue and inner mouth cavity. The new model achieves state-of-the-art performance. We use our model to reconstruct full head representations from single, unconstrained images allowing us to parameterize craniofacial shape and texture, along with the ear shape, eye gaze and eye color., 18 pages, 18 figures, submitted to Transactions on Pattern Analysis and Machine Intelligence (TPAMI) on the 9th of October as an extension paper of the original oral CVPR paper : arXiv:1903.03785

Recognition of the functional sites of genes, such as translation initiation sites, donor and acceptor splice sites and stop codons, is a relevant part of many current problems in bioinformatics. Recognition of the functional sites of genes is also a fundamental step in gene structure predictions in the most powerful programs. The best approaches to this type of recognition use sophisticated classifiers, such as support vector machines. However, with the rapid accumulation of sequence data, methods for combining many sources of evidence are necessary as it is unlikely that a single classifier can solve this type of problem with the best possible performance.A major issue is that the number of possible models to combine is large and the use of all of these models is impractical. In this paper, we present a framework that is based on floating search for combining as many classifiers as needed for the recognition of any functional sites of a gene. The methodology can be used for the recognition of translation initiation sites, donor and acceptor splice sites and stop codons. Furthermore, we can combine any number of classifiers that are trained on any species. The method is also scalable to large datasets, as is shown in experiments in which the whole human genome is used. The method is also applicable to other recognition tasks.We present experiments on the recognition of these four functional sites in the human genome, which is used as the target genome, and use another 20 species as sources of evidence. The proposed methodology shows significant improvement over state-of-the-art methods for use in a thorough evaluation process. The proposed method is also able to improve heuristic selection of species to be used as sources of evidence as the search finds the most useful datasets.Author summaryIn this paper we present a methodology for combining many sources of information to recognize some of the most important functional sites in a genomic sequence. The functional sites of the sequences, such as, translation start sites, translation initiation sites, acceptor and donor splice sites and stop codons, play a very relevant role in many Bioinformatics tasks. Their accurate recognition is an important task by itself and also as part of gene structure prediction programs.Our approach uses a methodology usually termed in Computer Science as “floating search”. This is a powerful heuristics applicable when the cost of evaluating each possible solution is high. The methodology is applied to the recognition of four different functional sites in the human genome using as additional sources of evidence the annotated genomes of other twenty different species.The results show an advantage of the proposed method and also challenge the standard assumption of using only genomes not very close and not very far from the human to improve the recognition of functional sites in the human genome.

This paper introduces a novel alignment-free sequence analysis methodology. Its main idea is based on introducing a new representation of the DNA-Sequence. This representation breaks the dependency between the DNA bases that exist in the traditional string presentation. We called it the Four-Lists-Representation (FLR). Based on the FLR, a series of revolutionary algorithms for searching, map-discovery, similarity-score analysis, and similarity-visualization have been developed. They are combined in what we call the FLR Methodology. The paper also studies most of the available similarity analysis techniques in a comprehensive state-of-art review. The conducted extensive simulation and theoretical studies confirm the outperformance of the whole set of FLR-based algorithms in terms of speed and memory consumption in comparison to a long list of available similarity analysis algorithms. The ability to provide a similarity-map, similarity-score, and similarity-graph as a set of evidence-based rationales makes the quality of results provided by the proposed methodology presents a new edge in this field and promises a new area of genome-based research.

Significance: A number of movement intent decoders exist in the literature that typically differ in the algorithms used and the nature of the outputs generated. Each approach comes with its own advantages and disadvantages. Combining the estimates of multiple algorithms may have better performance than any of the individual methods. Objective: This paper presents and evaluates a shared controller framework for prosthetic limbs based on multiple decoders of volitional movement intent. Methods: An algorithm to combine multiple estimates to control the prosthesis is developed in this paper. The capabilities of the approach are validated using a system that combines a Kalman filter-based decoder with a multilayer perceptron classifier-based decoder. The shared controller's performance is validated in online experiments where a virtual limb is controlled in real-time by amputee and intact-arm subjects. During the testing phase subjects controlled a virtual hand in real time to move digits to instructed positions using either a Kalman filter decoder, a multilayer perceptron decoder, or a linear combination of the two. Results: The shared controller results in statistically significant improvements over the component decoders. Specifically, certain degrees of shared control result in increases in the time-in-target metric and decreases in unintended movements. Conclusion: The shared controller of this paper combines the good qualities of component decoders tested in this paper. Herein, combining a Kalman filter decoder with a classifier-based decoder inherits the flexibility of the Kalman filter decoder and the limited unwanted movements from the classifier-based decoder, resulting in a system that may be able to perform the tasks of everyday life more naturally and reliably.

Person reidentification (Re-ID) aims to match observations of individuals across multiple nonoverlapping camera views. Recently, metric learning-based methods have played important roles in addressing this task. However, metrics are mostly learned in supervised manners, of which the performance relies heavily on the quantity and quality of manual annotations. Meanwhile, metric learning-based algorithms generally project person features into a common subspace, in which the extracted features are shared by all views. However, it may result in information loss since these algorithms neglect the view-specific features. Besides, they assume person samples of different views are taken from the same distribution. Conversely, these samples are more likely to obey different distributions due to view condition changes. To this end, this paper proposes an unsupervised cross-view metric learning method based on the properties of data distributions. Specifically, person samples in each view are taken from a mixture of two distributions: one models common prosperities among camera views and the other focuses on view-specific properties. Based on this, we introduce a shared mapping to explore the shared features. Meanwhile, we construct view-specific mappings to extract and project view-related features into a common subspace. As a result, samples in the transformed subspace follow the same distribution and are equipped with comprehensive representations. In this paper, these mappings are learned in an unsupervised manner by clustering samples in the projected space. Experimental results on five cross-view datasets validate the effectiveness of the proposed method.

Survival analysis is a popular branch of statistics. At present, many algorithms (like traditional multi-tasking learning model) cannot be applied well in practice because of censored data. Although using some model (like parametric regression model) can avoid it, they need strict assumptions. This undermines the very nature of things, which is very detrimental to the study of practical problems. The method proposed in this paper can apply well to the censored data, but does not need to make any additional assumptions about the original problem. It can be said that it breaks through the above two kinds of major limitations. The algorithm is a kind of inductive transfer learning method, which can fully obtain the information in the censored data, using domain-specific information implicit in each feature to enhance the generalization capability of the model. We also used two common performance metrics as criteria to judge the predictive performance differences between the models in this article and those of other mainstream models. The results show that the model proposed in this paper is 10 ∼ 15 percent higher than other mainstream models, which proves that our multi-task learning model has a great advantage in the survival analysis of cancer genes.

Conebeam CT using a circular trajectory is quite often used for various applications due to its relative simple geometry. For conebeam geometry, Feldkamp, Davis and Kress algorithm is regarded as the standard reconstruction method, but this algorithm suffers from so-called conebeam artifacts as the cone angle increases. Various model-based iterative reconstruction methods have been developed to reduce the cone-beam artifacts, but these algorithms usually require multiple applications of computational expensive forward and backprojections. In this paper, we develop a novel deep learning approach for accurate conebeam artifact removal. In particular, our deep network, designed on the differentiated backprojection domain, performs a data-driven inversion of an ill-posed deconvolution problem associated with the Hilbert transform. The reconstruction results along the coronal and sagittal directions are then combined using a spectral blending technique to minimize the spectral leakage. Experimental results show that our method outperforms the existing iterative methods despite significantly reduced runtime complexity., Comment: This paper is accepted for IEEE Trans. Medical Imaging

Photon-counting spectral computed tomography (CT) is capable of material characterization and can improve diagnostic performance over traditional clinical CT. However, it suffers from photon count starving for each individual energy channel which may cause severe artifacts in the reconstructed images. Furthermore, since the images in different energy channels describe the same object, there are high correlations among different channels. To make full use of the inter-channel correlations and minimize the count starving effect while maintaining clinically meaningful texture information, this paper combines a region-specific texture model with a low-rank correlation descriptor as an a priori regularization to explore a superior texture preserving Bayesian reconstruction of spectral CT. Specifically, the inter-channel correlations are characterized by the low-rank representation, and the inner-channel regional textures are modeled by a texture preserving Markov random field. In other words, this paper integrates the spectral and spatial information into a unified Bayesian reconstruction framework. The widely-used Split-Bregman algorithm is employed to minimize the objective function because of the non-differentiable property of the low-rank representation. To evaluate the tissue texture preserving performance of the proposed method for each channel, three references are built for comparison: one is the traditional CT image from energy integration detection. The second one is spectral images from dual-energy CT. The third one is individual channels images from custom-made photon-counting spectral CT. As expected, the proposed method produced promising results in terms of not only preserving texture features but also suppressing image noise in each channel, comparing to existing methods of total variation (TV), low-rank TV and tensor dictionary learning, by both visual inspection and quantitative indexes of root mean square error, peak signal to noise ratio, structural similarity and feature similarity.

In ultrasound (US) imaging, various types of adaptive beamforming techniques have been investigated to improve the resolution and contrast-to-noise ratio of the delay and sum (DAS) beamformers. Unfortunately, the performance of these adaptive beamforming approaches degrade when the underlying model is not sufficiently accurate and the number of channels decreases. To address this problem, here we propose a deep learning-based beamformer to generate significantly improved images over widely varying measurement conditions and channel subsampling patterns. In particular, our deep neural network is designed to directly process full or sub-sampled radio-frequency (RF) data acquired at various subsampling rates and detector configurations so that it can generate high quality ultrasound images using a single beamformer. The origin of such input-dependent adaptivity is also theoretically analyzed. Experimental results using B-mode focused ultrasound confirm the efficacy of the proposed methods., This is a significantly extended version of the original paper in arXiv:1901.01706. This paper is accepted for IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control

This paper considers implementation of artificial neural networks (ANNs) using molecular computing and DNA based on fractional coding. Prior work had addressed molecular two-layer ANNs with binary inputs and arbitrary weights. In prior work using fractional coding, a simple molecular perceptron that computes sigmoid of scaled weighted sum of the inputs was presented where the inputs and the weights lie between [-1, 1]. Even for computing the perceptron, the prior approach suffers from two major limitations. First, it cannot compute the sigmoid of the weighted sum, but only the sigmoid of the scaled weighted sum. Second, many machine learning applications require the coefficients to be arbitrarily positive and negative numbers that are not bounded between [-1, 1]; such numbers cannot be handled by the prior perceptron using fractional coding. This paper makes four contributions. First molecular perceptrons that can handle arbitrary weights and can compute sigmoid of the weighted sums are presented. Thus, these molecular perceptrons are ideal for regression applications and multi-layer ANNs. A new molecular divider is introduced and is used to compute sigmoid(ax) where a > 1. Second, based on fractional coding, a molecular artificial neural network (ANN) with one hidden layer is presented. Third, a trained ANN classifier with one hidden layer from seizure prediction application from electroencephalogram is mapped to molecular reactions and DNA and their performances are presented. Fourth, molecular activation functions for rectified linear unit (ReLU) and softmax are also presented., IEEE Transactions on Biomedical Circuits and Systems, 2020

This paper studies the fixed-time anti-synchronization (FTAS) of discontinuous reaction-diffusion neural networks (DRDNNs) with both time-varying coefficients and time delay. First, differential inclusion theory is used to deal with the influence caused by discontinuous activations. In addition, a new fixed-time convergence theorem is used to handle the time-varying coefficients. Second, a novel state-feedback control algorithm and integral state-feedback control algorithm are proposed to realize FTAS of DRDNNs. During the generalized (adaptive) pinning control strategy, a guideline is proposed to select neurons to pin the designed controller. Furthermore, we present several criteria on FTAS by using the generalized Lyapunov function method. Different from the traditional Lyapunov function with negative definite derivative, the derivative of the Lyapunov function can be positive in this paper. Finally, we give two numerical simulations to substantiate the merits of the obtained results.

Genome sequence data consists of DNA sequences or input sequences. Each one includes nucleotides with chemical structures presented as characters: ${^{\prime}\mathrm{A}^{\prime}, ^{\prime}\mathrm{C}^{\prime},^{\prime} \mathrm{G}^{\prime}}$ ' A ' , ' C ' , ' G ' , and ‘T’, and groups of motif sequences, called Transcription Factor Binding Sites (TFBSs), which are subsequences of DNA that lead to protein-synthesis. The detection of TFBSs is an important problem for bioinformatics research. With the similar patterns of motif sequences in TFBSs, computational algorithms for TFBSs detection have been improved to reduce resources used in laboratory setting. The metaheuristic algorithm is the important issue that has been continually improved to detect TFBSs with greater precision and recall. This paper proposes PSO_HD by applying Particle Swarm Optimization (PSO) as a pre-process and using Hamming distance to improve the efficiency of detecting TFBSs with more precision and recall. In order to measure its efficiency, the paper compares the TFBSs detection using PSO_HD algorithm with relevant algorithms in eight datasets. F -score is used as a measurement unit and compared to the related algorithms. The experimental results show that PSO_HD algorithm gives the highest average F -score, which can be indicated that the PSO_HD algorithm can improve the efficiency of detecting TFBSs with more precision and recall.

Objective: Radar technology promises to be a touchless and thereby burden-free method for continuous heart sound monitoring, which can be used to detect cardiovascular diseases. However, the first and most crucial step is to differentiate between high- and low-quality segments in a recording to assess their suitability for a subsequent automated analysis. This paper gives a comprehensive study on this task and first addresses the specific characteristics of radar-recorded heart sound signals. Methods: To gather heart sound signals recorded from radar, a bistatic radar system was built and installed at the university hospital. Under medical supervision, heart sound data were recorded from 30 healthy test subjects. The signals were segmented and labeled as high- or low-quality by a medical expert. Different state-of-the-art pattern classification algorithms were evaluated for the task of automated signal quality determination and the most promising one was optimized and evaluated using leave-one-subject-out cross validation. Results: The proposed classifier is able to achieve an accuracy of up to 96.36% and demonstrates a superior classification performance compared with the state-of-the-art classifier with a maximum accuracy of 76.00%. Conclusion: This paper introduces an ensemble classifier that is able to perform automated signal quality determination of radar-recorded heart sound signals with a high accuracy. Significance: Besides achieving a higher performance compared with state-of-the-art classifiers, this study is the first one to deal with the quality determination of heart sounds that are recorded by radar systems. The proposed method enables contactless and continuous heart sound monitoring for the detection of cardiovascular diseases.

Performing a fetal electrocardiogram (ECG) analysis, which contains important information about the status of a fetal, can help to detect fetus health even before birth. Since the fetal ECG extracted from the ECG signal recorded from the mother's abdomen, this extraction problem can be seen as a source separation problem, of recovering source signals from signal mixtures. In this paper, a method for separation of fetal ECG from abdominal ECG using activation scaled non-negative matrix factorization (NMF) is proposed. The performance of the proposed method is also compared with independent component analysis. The proposed method is tested under three different scenarios. First, the original abdominal ECG signal is used for fetal separation. Second, the recovered abdominal ECG after compression is used for separation. Third, the fetal ECG is extracted from the compressed domain of the abdominal ECG. We applied scaling on the activation matrix obtained using NMF for emphasizing the fetal ECG present in abdominal ECG. The improved-regularized least-squares $(l_p-\text{RLS})$ algorithm is used for signal reconstruction, which provides better reconstruction quality and less processing time in comparison with other existing methods. The proposed algorithm is evaluated and tested on real abdominal recordings obtained from two different datasets from Physionet. The first dataset used for this paper is Silesia dataset for abdominal and direct f-ECG, and the second dataset we considered is Set-A of the Physionet challenge. The obtained outcomes reveal that it is possible to separate fetal ECG from single-channel abdominal ECG signal, which can help us to achieve energy-efficient transmission, and cost-effective fetal ECG remote monitoring for Internet-of-Things applications, where device battery and computational capacity are limited.

Objective : Multi-modal brain functional connectivity (FC) data have shown great potential for providing insights into individual variations in behavioral and cognitive traits. The joint learning of multi-modal data can utilize intrinsic association, and thus can boost learning performance. Although several multi-task based learning models have already been proposed by viewing feature learning on each modality as one task, most of them ignore the structural information inherent across the modalities, which may play an important role in extracting discriminative features. Methods : In this paper, we propose a new manifold regularized multi-task learning model by simultaneously considering between-subject and between-modality relationships. Specifically, the $l_{2,1}$ -norm (i.e., group-sparsity) regularizer is enforced to jointly select a few common features across different modalities. A novelly designed manifold regularizer is further imposed as a crucial underpinning to preserve the structural information both within and between modalities. Such designed regularizers will make our model more adaptive to realistic neuroimaging data, which are usually of small sample size but high dimensional features. Results : Our model is validated on the Philadelphia Neurodevelopmental Cohort dataset, where our modalities are regarded as two types of functional MRI (fMRI) data collected under two paradigms. We conduct experimental studies on fMRI-based FC network data in two task conditions for intelligence quotient (IQ) prediction. The results show that our proposed model can not only achieve improved prediction performance, but also yield a set of IQ-relevant biomarkers. Conclusion and Significance : This paper develops a new multi-task learning model, enabling the discovery of significant biomarkers that may account for a proportion of the variance in human intelligence.

Through hemodynamic models, the change of neuronal state can be estimated from functional magnetic resonance imaging (fMRI) signals. Usually, there are confounds in the fMRI signal, which will degrade the performance of the estimation for the neuronal state change. For the reason, this paper introduces a state-space model with confounds, from a conventional hemodynamic model. In this model, a successive state estimation method requires a state value vector, an error covariance, an innovation covariance, and a cross covariance to be re-derived. Thus, a confounds square-root cubature Kalman smoothing (CSCKS) algorithm is proposed in this paper. We use a Balloon-Windkessel model to generate simulation data and add confounds signals to evaluate the performance of the proposed algorithm. The experiment results show that when the signal-to-interference ratio is less than 21 dB, the CSCKS proposed in this paper reduced estimation error to 16%, whereas the traditional algorithm reduced it to only 73%.

Previous work has shown that total variation superiorization (TVS) improves reconstructed image quality in proton computed tomography (pCT). The structure of the TVS algorithm has evolved since then and this paper investigated if this new algorithmic structure provides additional benefits to pCT image quality. Structural and parametric changes introduced to the original TVS algorithm included: (1) inclusion or exclusion of TV reduction requirement, (2) a variable number, $N$ , of TV perturbation steps per feasibility-seeking iteration, and (3) introduction of a perturbation kernel $0 . The structural change of excluding the TV reduction requirement check tended to have a beneficial effect for $3\le N\le 6$ and allows full parallelization of the TVS algorithm. Repeated perturbations per feasibility-seeking iterations reduced total variation (TV) and material dependent standard deviations for $3\le N\le 6$ . The perturbation kernel $\alpha $ , effectively equal to $\alpha =0.5$ in the original TVS algorithm, reduced TV and standard deviations as $\alpha $ was increased beyond $\alpha =0.5$ , but negatively impacted reconstructed relative stopping power (RSP) values for $\alpha >0.75$ . The reductions in TV and standard deviations allowed feasibility-seeking with a larger relaxation parameter $\lambda $ than previously used, without the corresponding increases in standard deviations experienced with the original TVS algorithm. This paper demonstrates that the modifications related to the evolution of the original TVS algorithm provide benefits in terms of both pCT image quality and computational efficiency for appropriately chosen parameter values.

This paper proposes a number of theorems and algorithms for the Chinese Remainder Theorem, which is used to solve a system of linear congruences, and the extended Rabin cryptosystem, which accepts a key composed of an arbitrary finite number of distinct primes. This paper further proposes methods to relax the condition on the primes with trade-offs in the time complexity. The proposed algorithms can be used to provide ciphertext indistinguishability. Finally, this paper conducts extensive experimental analysis on six large data sets. The experimental results show that the proposed algorithms are asymptotically tight to the existing decryption algorithm in the Rabin cryptosystem with the key composed of two distinct primes while maintaining increased generality.

This paper develops an effective distributed strategy for the solution of constrained multiagent stochastic optimization problems with coupled parameters across the agents. In this formulation, each agent is influenced by only a subset of the entries of a global parameter vector or model, and is subject to convex constraints that are only known locally. Problems of this type arise in several applications, most notably in disease propagation models, minimum-cost flow problems, distributed control formulations, and distributed power system monitoring. This paper focuses on stochastic settings, where a stochastic risk function is associated with each agent and the objective is to seek the minimizer of the aggregate sum of all risks subject to a set of constraints. Agents are not aware of the statistical distribution of the data and, therefore, can only rely on stochastic approximations in their learning strategies. We derive an effective distributed learning strategy that is able to track drifts in the underlying parameter model. A detailed performance and stability analysis is carried out showing that the resulting coupled diffusion strategy converges at a linear rate to an $O(\mu)$ neighborhood of the true penalized optimizer.

Voice disability is a barrier to effective communication. Around 1.2% of the World's population is facing some form of voice disability. Surgical procedures namely laryngoscopy, laryngeal electromyography, and stroboscopy are used for voice disability diagnosis. Researchers and practitioners have been working to find alternatives to these surgical procedures. Voice sample based diagnosis is one of them. The major steps followed by these works are (a) to extract voice features from voice samples and (b) to discriminate pathological voices from normal voices by using a classifier algorithm. However, there is no consensus about the voice feature and the classifier algorithm that can provide the best accuracy in screening voice disability. Moreover, some of the works use multiple voice features and multiple classifiers to ensure high reliability. In this paper, we address these issues. The motivation of the work is to address the need for non-invasive signal processing techniques to detect voice disability in the general population. This paper conducts a survey related to voice disability detection methods. The paper contains two main parts. In the first part, we present background information including causes of voice disability, current procedures and practices, voice features, and classifiers. In the second part, we present a comprehensive survey work on voice disability detection algorithms. The issues and challenges related to the selection of voice feature and classifier algorithms have been addressed at the end of this paper.

This paper focuses on constructing genotypic predictors for antiretroviral drug susceptibility of HIV. To this end, a method to recover the largest elements of an unknown vector in a least squares problem is developed. The proposed method introduces two novel ideas. The first idea is a novel forward stepwise selection procedure based on the magnitude of the estimates of the candidate variables. To implement this newly introduced procedure, we revise Tikhonov regularisation from a sparse representations' perspective. This analysis leads us to the second novel idea in the paper, which is the development of a new method to recover the largest elements of the unknown vector in the least squares problem. The method implements a sequence of Tikhonov regularisation problems which aim to recover the largest of the remaining elements of the unknown vector. Additionally, we derive sufficient conditions that ensure the recovery of the largest elements of the unknown vector. We perform numerical studies using simulated data and data from the Stanford HIV resistance database. The performance of the proposed method is compared against a state-of-the-art method.

Reconstructing large-scale gene regulatory networks (GRNs) is a challenging problem in the field of computational biology. Various methods for inferring GRNs have been developed, but they fail to accurately infer GRNs with a large number of genes. Additionally, the existing evaluation indexes for evaluating the constructed networks have obvious disadvantages because GRNs in most biological systems are sparse. In this paper, we develop a new method for inferring GRNs based on randomized singular value decomposition (RSVD) and ordinary differential equation (ODE)-based optimization, denoted as IGRSVD, from large-scale time series data with noise. The three major contributions of this paper are as follows. First, the IGRSVD algorithm uses the RSVD to handle the noise and reduce the original large-scale data into small-scale problems. Second, we propose two new evaluated indexes, the expected value accuracy (EVA) and the expected value error (EVE), to evaluate the performance of inferred networks by considering the sparse features in the network. Finally, the proposed IGRSVD algorithm is compared with the existing SVD algorithm and PCA_CMI algorithm using four subsets from E. coli and datasets from DREAM challenge. The experimental results demonstrate that the IGRSVD algorithm is effective and more suitable for reconstructing large-scale networks.

Objective. In the traditional sense, the modeling approaches can be divided into white-box (physics-based), black-box (data-driven), and gray-box (the combination of physics-based and data-driven). Because the human brain is a black box itself, the EEG-BCI algorithm is generally a data-driven approach. It generates a black-box or gray-box (e.g., "Visualizing convolutional networks") model. However, one black- or gray-box cannot completely explain the brain. This paper presents the first analytic "white-box" EEG-BCI algorithm using Gang neurons (EEGG).Approach. Independent and interactive components of neurons or brain regions can fully describe the brain. This paper constructed a relationship frame based on the independent and interactive compositions for intention recognition and analysis using a novel dendrite module of Gang neuron. A total of 4,906 EEG data about motor imagery (MI) of left-hand and right-hand movements from 26 subjects were obtained from GigaDB. Firstly, this paper explored EEGG's classification performance according to cross-subject accuracy. Secondly, this paper transformed the EEGG model into a relation spectrum expressing independent and interactive components of brain regions. Then, the relation spectrum was verified through the known ERD/ERS phenomenon. Finally, this paper explored the previously unreachable further BCI-based analysis of brain.Main results. (1) EEGG was more robust than typical "CSP+" algorithms for the data of poor quality [AUC:0.825±0.074(EEGG)>0.745±0.094(CSP+LDA)/0.591±0.104(CSP+Bayes)/0.750±0.091(CSP+SVM), pSignificance. EEGG decomposed the biological EEG-intention system of this paper into the relation spectrum inheriting Taylor series, rather than the fuzzy interpretation of outputs, which offers a novel frame for analysis of the brain.

For many cerebrovascular diseases both blood pressure (BP) and hemodynamic changes are important clinical variables. In this paper, we describe the development of a novel approach to noninvasively and simultaneously monitor cerebral hemodynamics, BP, and other important parameters at high temporal resolution (250 Hz sampling rate). In this approach, cerebral hemodynamics are acquired using near infrared spectroscopy based sensors and algorithms, whereas continuous BP is acquired by superficial temporal artery tonometry with pulse transit time based drift correction. The sensors, monitoring system, and data analysis algorithms used in the prototype for this approach are reported in detail in this paper. Preliminary performance tests demonstrated that we were able to simultaneously and noninvasively record and reveal cerebral hemodynamics and BP during people’s daily activity. As examples, we report dynamic cerebral hemodynamic and BP fluctuations during postural changes and micturition. These preliminary results demonstrate the feasibility of our approach, and its unique power in catching hemodynamics and BP fluctuations during transient symptoms (such as syncope) and revealing the dynamic features of related events.

The paper proposes an innovative deep convolutional neural network (DCNN) combined with texture map for detecting cancerous regions and marking the ROI in a single model automatically. The proposed DCNN model contains two collaborative branches, namely an upper branch to perform oral cancer detection, and a lower branch to perform semantic segmentation and ROI marking. With the upper branch the network model extracts the cancerous regions, and the lower branch makes the cancerous regions more precision. To make the features in the cancerous more regular, the network model extracts the texture images from the input image. A sliding window is then applied to compute the standard deviation values of the texture image. Finally, the standard deviation values are used to construct a texture map, which is partitioned into multiple patches and used as the input data to the deep convolutional network model. The method proposed by this paper is called texture-map-based branch-collaborative network. In the experimental result, the average sensitivity and specificity of detection are up to 0.9687 and 0.7129, respectively based on wavelet transform. And the average sensitivity and specificity of detection are up to 0.9314 and 0.9475, respectively based on Gabor filter.

The ratio of spectral power in two different bands and relative band power have been shown to be sometimes more discriminative features than the spectral power in a specific band for binary classification of a time series for seizure prediction. However, why and which ratio of spectral power and relative power features are better discriminators than a band power have not been understood. While general answers to why and which are difficult, this paper partially addresses the answer to these questions. Using auto-regressive modeling, this paper, for the first time, theoretically explains that for high signal-to-noise ratio (SNR) cases, the ratio features may sometime amplify the discriminability of one of the two states in a time series, as compared with a band power. This paper, also for the first time, introduces a novel frequency-domain model ratio (FDMR) that can be used to select the two frequency bands. The FDMR computes the ratio of the frequency responses of the two auto-regressive model filters that correspond to two different states. It is shown that the ratio implicitly cancels the effect of change of variance of the white noise that is input to the auto-regressive model in a non-stationary environment for high SNR conditions. It is also shown that under certain sufficient but not necessary conditions, the ratio of the spectral power and the relative band power, i.e., the band power divided by the total power spectral density, can be better discriminators than band power. Synthesized data and scalp EEG data from the MIT Physionet for patient-specific seizure prediction are used to explain why the ratios of spectral power obtained by a ranking algorithm in the prior literature satisfy the sufficient conditions for amplification of the ratio feature derived in this paper.

For patients with mechanical ventilation, secretions in airway are harmful and sometimes even mortal, it's of great significance to clear secretion timely and efficiently. In this paper, a new secretion clearance method for VCV (volume-controlled ventilation) ventilator is put forward, and a secretion clearance system with a VCV ventilator and double lungs is designed. Furthermore, the mathematical model of the secretion clearance system is built and verified via experimental study. Finally, to illustrate the influence of key parameters of respiratory system and secretion clearance system on the secretion clearance characteristics, coupling effects of two lungs on VCV secretion clearance system are studied by an orthogonal experiment, it can be obtained that rise of tidal volume adds to efficiency of secretion clearance while effect of area, compliance, and suction pressure on efficiency of secretion clearance needs further study. Rise of compliance improves bottom pressure of secretion clearance while rise of area, tidal volume, and suction pressure decreases bottom pressure of secretion clearance. This paper can be referred to in researches of secretion clearance for VCV.

One of the major limitations of motor imagery (MI)-based brain–computer interface (BCI) is its long calibration time. Due to between sessions/subjects variations in the properties of brain signals, typically, a large amount of training data needs to be collected at the beginning of each session to calibrate the parameters of the BCI system for the target user. In this paper, we propose a novel transfer learning approach on the classification domain to reduce the calibration time without sacrificing the classification accuracy of MI-BCI. Thus, when only few subject-specific trials are available for training, the estimation of the classification parameters is improved by incorporating previously recorded data from other users. For this purpose, a regularization parameter is added to the objective function of the classifier to make the classification parameters as close as possible to the classification parameters of the previous users who have feature spaces similar to that of the target subject. In this paper, a new similarity measure based on the Kullback–Leibler divergence (KL) is used to measure the similarity between two feature spaces obtained using subject-specific common spatial patterns (CSP). The proposed transfer learning approach is applied on the logistic regression classifier and evaluated using three datasets. The results showed that compared with the subject-specific classifier, the proposed weighted transfer learning classifier improved the classification results, particularly when few subject-specific trials were available for training ( p < 0.05). Importantly, this improvement was more pronounced for users with medium and poor accuracy. Moreover, the statistical results showed that the proposed weighted transfer learning classifier performed significantly better than the considered comparable baseline algorithms.

Objective: Ballistocardiogram (BCG) can be recorded using inexpensive and non-invasive hardware to estimate physiological changes in the heart. In this paper, a methodology is developed to evaluate the impact of additive noise on the BCG signal. Methods: A statistical model is built that incorporates subject-specific BCG morphology. BCG signals segmented by electrocardiogram RR intervals (BCG heartbeats) are averaged to estimate a parent template and subtemplates leveraging the quasi-periodic nature of the heart. Noise statistics are obtained for subtemplates with respect to the parent template. Then, a synthesis algorithm with adjustable additive noise is devised to generate subtemplates based on the individual's parent template and statistics. For the example use of the synthesis algorithm, the average correlation coefficient between subtemplates and the parent template (subtemplate versus parent template approach) is tested as a signal quality index. Results: A BCG heartbeat synthesis framework that incorporates an individual's BCG morphology and physiological variability was developed to quantify variations in the BCG signal against additive noise. The signal quality assessment of a person's BCG recording can be performed without requiring any a priori knowledge of the person's BCG morphology. A data-driven constraint on the required minimum number of heartbeats for a reliable template estimation was provided. Conclusion: The impact of additive noise on BCG morphology and estimated physiological parameters can be analyzed using the developed methodology without requiring prior statistics. Significance: This paper can facilitate the performance evaluation of BCG analysis algorithms against additive noise.

Duplication-Transfer-Loss (DTL) reconciliation is a powerful method for studying gene family evolution in the presence of horizontal gene transfer. DTL reconciliation seeks to reconcile gene trees with species trees by postulating speciation, duplication, transfer, and loss events. Efficient algorithms exist for finding optimal DTL reconciliations when the gene tree is binary. In practice, however, gene trees are often non-binary due to uncertainty in the gene tree topologies, and DTL reconciliation with non-binary gene trees is known to be NP-hard. In this paper, we present the first, exact algorithms for DTL reconciliation with non-binary gene trees. Specifically, we (i) show that the DTL reconciliation problem for non-binary gene trees is fixed-parameter tractable in the maximum degree of the gene tree, (ii) present an exponential-time, but in-practice efficient, algorithm to track and enumerate all optimal binary resolutions of an unresolved input gene tree, and (iii) apply our algorithms to a large empirical dataset of over 4700 gene trees from 100 species to study the impact of gene tree uncertainty on DTL-reconciliation and to demonstrate the applicability and utility of our algorithms. The new techniques and algorithms introduced in this paper make it possible, for the first time, to systematically calculate and negate the impact of gene tree uncertainty on reconciliation accuracy, and will help biologists avoid incorrect evolutionary inferences caused by gene tree uncertainty.

Medical image processing is often limited by the computational cost of the involved algorithms. Whereas dedicated computing devices (GPUs in particular) exist and do provide significant efficiency boosts, they have an extra cost of use in terms of housekeeping tasks (device selection and initialization, data streaming, synchronization with the CPU, and others), which may hinder developers from using them. This paper describes an OpenCL-based framework that is capable of handling dedicated computing devices seamlessly and that allows the developer to concentrate on image processing tasks. The framework handles automatically device discovery and initialization, data transfers to and from the device and the file system and kernel loading and compiling. Data structures need to be defined only once independently of the computing device; code is unique, consequently, for every device, including the host CPU. Pinned memory/buffer mapping is used to achieve maximum performance in data transfers. Code fragments included in the paper show how the computing device is almost immediately and effortlessly available to the users algorithms, so they can focus on productive work. Code required for device selection and initialization, data loading and streaming and kernel compilation is minimal and systematic. Algorithms can be thought of as mathematical operators (called processes), with input, output and parameters, and they may be chained one after another easily and efficiently. Also for efficiency, processes can have their initialization work split from their core workload, so process chains and loops do not incur in performance penalties. Algorithm code is independent of the device type targeted.

Objective: The aim of this paper is to improve the reconstruction accuracy in both position and source region of fluorescence molecular tomography (FMT). Methods: The reconstruction of the FMT is challenging due to its serious ill-posedness and ill-condition. Currently, to obtain the fluorescent sources accurately, more a priori information of the fluorescent sources is utilized and more efficient and practical methods are proposed. In this paper, we took the group sparsity of the fluorescent sources as a new type of priori information in the FMT, and proposed the fused LASSO method (FLM) for FMT. The FLM based on group sparsity prior not only takes advantage of the sparsity of the fluorescent sources, but also utilizes the structure of the sources, thus making the reconstruction results more accuracy and morphologically similar to the sources. To further improve the reconstruction efficiency, we adopt Nesterov's method to solve the FLM. Results: Both heterogeneous numerical simulation experiments and in vivo mouse experiments were carried out to verify the property of the FLM. The results have verified the superiority of the FLM over conventional methods in tumor detection and tumor morphological reconstruction. Furthermore, the in vivo experiments had demonstrated that the FLM has great potential in preclinical application of the FMT. Significance: The reconstruction method based on group sparsity prior has a great potential in the FMT study, it can further improve the reconstruction quality, which has practical significance in preclinical research.

This paper proposes a novel approach for online, individualized gait analysis, based on an adaptive periodic model of any gait signal. The proposed method learns a model of the gait cycle during online measurement, using a continuous representation that can adapt to inter- and intra-personal variability by creating an individualized model. Once the algorithm has converged to the input signal, key gait events can be identified based on the estimated gait phase and amplitude. The approach is implemented and tested on retirement home resident 6 min walk (6MW) data using wearable accelerometers at the ankle. The proposed approach converges within approximately four gait cycles and achieves 3% error in detecting initial swing events. 1 1 An early version of this work was presented in [1] . A more extensive description of related work and an extended method, including optimization of learning rates, were added to this paper. Further, this paper applies and evaluates the method to a new and much larger gait dataset taken from older adults who each have a variety of medical conditions. Therefore, the experimental protocol was also updated and the results are entirely novel.

This paper presents a wearable sensor architecture for frequency-multiplexed electrical impedance tomography (EIT) and synchronous multilead electrocardiogram (ECG) data acquisition. The system is based on a novel electronic sensing architecture, called cooperative sensors, that significantly reduces the cabling complexity and enables flexible EIT stimulation and measurement patterns. The cooperative-sensor architecture was initially designed for ECG and has been extended for multichannel bioimpedance measurement. This approach allows for an adjustable EIT stimulation pattern via frequency-division multiplexing. This paper also shows a calibration procedure as well as EIT system noise performance assessment. Preliminary measurements on a healthy volunteer showed the ability of the wearable system to measure EIT data synchronously with multilead ECG. Ventilation-related and heartbeat-related EIT images were reconstructed, demonstrating the feasibility of the proposed architecture for non-invasive cardiovascular monitoring.

This paper presents a QuadTree-based melanoma detection system inspired by dermatologists' color perception. Clinical color assessment in dermoscopy images is challenging because of subtle differences in shades, location-dependent color information, poor color contrast, and wide variation among images of the same class. To overcome these challenges, color enhancement and automatic color identification techniques, based on QuadTree segmentation and modeled after expert color assessments, are developed. The approach presented in this paper is shown to provide an accurate model of expert color assessment. Specifically, the proposed model is shown to: 1) identify significantly more colors in melanomas than in benign skin lesions; 2) identify a higher frequency in melanomas of three colors: blue-gray, black, and pink; and 3) delineate locations of melanoma colors by quintiles, specifically predilection for blue-gray and pink in the periphery and a trend for white and black in the lesion center. Performance of the proposed method is evaluated using four classifiers. The kernel support vector machine classifier is found to achieve the best results, with an area under the receiver operating characteristic (ROC) curve of 0.93, compared to average area under the ROC curve of 0.82 achieved by the dermatologists in this study. The results indicate that the biologically inspired method of automatic color detection proposed in this paper has the potential to play an important role in melanoma diagnosis in the clinic.

High-definition transcranial direct current stimulation (HD-tDCS) is a potential neuromodulation apparatus for stroke rehabilitation. However, its modulatory effects in stroke subjects is still not well understood. In this paper, the offline modulatory effects of HD-tDCS on the ipsilesional primary motor cortex were investigated by performing wrist isometric contraction tasks before and after HD-tDCS in eleven unilateral chronic stroke subjects using a synchronized HD-tDCS and electroencephalogram/electromyography measurement system. This paper is a randomized, single blinded, and sham-controlled crossover study. Each subject randomly received three HD-tDCS (anode, cathode, and sham) with at least one-week washout period. Online feedback-guided medium-level wrist isometric contraction tasks were conducted for the affected upper limbs before stimulation and 10, 30, and 50 min after the end of 10-min 1-mA HD-tDCS. The characteristics of corticomuscular coherence (CMC), cortical oscillation power spectral density, and power spectral entropy were analyzed during tasks and compared across all sessions and stimulation conditions. Anode HD-tDCS induced significant CMC changes in stroke subjects, while cathode and sham stimulation did not induce significant CMC changes. The largest neuromodulation effects were observed at 10 min immediately after anodal HD-tDCS.