Search

Objective: We propose a MATLAB-based tool to convert electrocardiography (ECG) waveforms from paper-based ECG records into digitized ECG signals that is vendor-agnostic. The tool is packaged as an open source standalone graphical user interface (GUI) based application. Methods and procedures: To reach this objective we: (1) preprocess the ECG records, which includes skew correction, background grid removal and linear filtering; (2) segment ECG signals using Connected Components Analysis (CCA); (3) implement Optical Character Recognition (OCR) for removal of overlapping ECG lead characters and for interfacing of patients’ demographic information with their research records or their electronic medical record (EMR). The ECG digitization results are validated through a reader study where clinically salient features, such as intervals of QRST complex, between the paper ECG records and the digitized ECG records are compared. Results: Comparison of clinically important features between the paper-based ECG records and the digitized ECG signals, reveals intra- and inter-observer correlations of 0.86–0.99 and 0.79–0.94, respectively. The kappa statistic was found to average at 0.86 and 0.72 for intra- and inter-observer correlations, respectively. Conclusion: The clinically salient features of the ECG waveforms such as the intervals of QRST complex, are preserved during the digitization procedure. Clinical and Healthcare Impact: This open-source digitization tool can be used as a research resource to digitize paper ECG records thereby enabling development of new prediction algorithms to risk stratify individuals with cardiovascular disease, and/or allow for development of ECG-based cardiovascular diagnoses relying upon automated digital algorithms.

This paper considers the precoder designs for successive zero-forcing dirty paper coding (SZF-DPC), a suboptimal transmission technique for MIMO broadcast channels (MIMO BCs). Existing precoder designs for SZF-DPC often consider a sum power constraint. In this paper, we address the precoder design for SZF-DPC with per-antenna power constraints (PAPCs), which has not been well studied. First, we formulate the precoder design as a rank-constrained optimization problem, which is generally difficult to handle. To solve this problem, we follow a relaxation approach, and prove that the optimal solution of the relaxed problem is also optimal for the original problem. Considering the relaxed problem, we propose a numerically efficient algorithm to find the optimal solution, which exhibits a fast convergence rate. Suboptimal precoder designs, with lower computational complexity, are also presented, and compared with the optimal ones in terms of achievable sum rate and computational complexity. © 2012 IEEE.

Two wheeled Self balancing robots have become more popular in recent times and used in many applications, commonly used in transportation as these balancing robots are faster and apparently stable. Balancing is achieved by using controllers tuned for the system. Inverted pendulum principle is used in these robots. This paper proposes the use of two control methods, namely Proportional-Integral-Derivative (PID) and Extended PID (EPID), for maintaining the balance of two-wheeled and self-balancing robots, even when slightly tilted, thus preventing them from falling. PID control theory generally depends up on transfer functions, this paper extends the theory using EPID. EPID provides a new perspective on the working principle of PID control and its relationship with other control theories, such as tracking control and disturbance observer. This approach is illustrated using the fundamental principle of inverted pendulum technology. The key advantage of these robots over other mobile robots is their increased stability due to zero turning radius, which allows for smoother rotations during changes in direction.

© 2022 IEEE.Deep transfer learning is one of the common techniques used to classify different types of cancer. The goal of this research is to focus on and adopt a fast, accurate, suitable, and reliable for classification of colorectal cancer. Digital histology images are adjustable to the application of convolutional neural networks (CNNs) for analysis and classification, due to the sheer size of pixel data present in them. Which can provide a lot of information about colorectal cancer. We used ten different types of pr-trained models with two type method of classification techniques namely (normal classification and k-fold crosse validation) to classify the tumor tissue, we used two different kinds of datasets were these datasets consisting of three classes (normal, low tumors, and high tumors). Among all these eight models of deep transfer learning, the highest accuracy achieved was 96.6% with Darknet53 for 5-Fold and for normal classification the highest results obtained was 98.7% for ResNet50. Moreover, we compared our result with many other papers in stat-of-the-art, the results obtained show clearly the proposed method was outperformed the other papers.

This paper reviews the recent literature on technologies and methodologies for quantitative human gait analysis in the context of neurodegnerative diseases. The use of technological instruments can be of great support in both clinical diagnosis and severity assessment of these pathologies. In this paper, sensors, features and processing methodologies have been reviewed in order to provide a highly consistent work that explores the issues related to gait analysis. First, the phases of the human gait cycle are briefly explained, along with some non-normal gait patterns (gait abnormalities) typical of some neurodegenerative diseases. The work continues with a survey on the publicly available datasets principally used for comparing results. Then the paper reports the most common processing techniques for both feature selection and extraction and for classification and clustering. Finally, a conclusive discussion on current open problems and future directions is outlined.

© 2021 IEEE.Coronavirus usually transmits from the animal to the human, but now, the virus transmission is between persons. Therefore, scientists and researchers are trying to develop several types of machine learning methods to defend against COVID-19. Medical images play a significant role in this time due to they can be used to recognize COVID-19 accurately. However, in this paper, we used X-Ray images, the images undergone to sharpening techniques to increase the results further. The texture techniques named local binary pattern (LBP) have been used in order to extract features. The features obtained were applied to the support vector machine (SVM). The results we achieved were 100% for all performance measurements. Our results were conspicuously superior compared to the state-of-the-art papers published.

An intelligent speed advisory system can be used to recommend speed for vehicles travelling in a given road network in cities. In this paper, we extend our previous work where a distributed speed advisory system has been devised to recommend an optimal consensus speed for a fleet of Internal Combustion Engine Vehicles (ICEVs) in a highway scenario. In particular, we propose a novel optimisation framework where the exact format of each vehicle's cost function can be implicit, and our algorithm can be used to recommend multiple consensus speeds for vehicles travelling on different lanes in an urban highway scenario. Our studies show that the proposed scheme based on an improved whale optimisation algorithm can effectively reduce CO2 emission generated from ICEVs while providing different recommended speed options for groups of vehicles., This paper has been accepted by IEEE VTC2021-Spring for presentation

This paper tries to establish the most important parameters of readability when it comes to choosing reading books for students in the second and third stages of primary and compulsory secondary education (CSE). A computational system is postulated, through data-mining techniques, capable of automatically classify reading texts by virtue of syntactic, lexical, semantics, and topological parameters regarding their overall content. The validity of this process is ensured through a careful selection by editorial experts used to accomplish the same validation task among different texts. In this paper, the results in the automatic classification of the degree of readability reached are promising; they allow to continue with a deeper investigation in this field of analysis with the purpose of progressing in the age rank studied, as well as in the text themes defining the corpus.

A novel respiratory signal detection system capable of simultaneously tracking the position of the subject and detecting his or her respiratory signal is described. The monitoring system consists of depth camera with ultra wide band radar device. Both sensors are connected through a mini computer, which performs data acquisition and storage. In this paper, we propose a method to locate the position of the subject where he or she is lying in the bed covered with blanket. Mask R-CNN is used to help segment upper-body's silhouette and give out the center point distance. The distance between the camera and the subject is then converted into a range bin of the radar and the breath-like signal is extracted from that range- bin. Additional contribution of this paper is that we developed a classifier to classify the whether the extracted signal in the selected range bin is indeed a breathing signal or not.

In this paper a new compression technique based on the discrete Tchebichef transform is presented. To comply with strict on-implant hardware implementation requirements, such as low power dissipation and small silicon area consumption, the discrete Tchebichef transform is modified and truncated. An algorithm is proposed to generate approximate transform matrices capable of truncation without suffering from destructive energy leakage among the coefficients. This is achieved by preserving orthogonality of the basis functions that convey majority portion of the signal energy. Based on the presented algorithm, a new truncated transformation matrix is proposed, which reduces the hardware complexity by up to 74% compared to that of the original transform. Hardware implementation of the proposed neural signal compression technique is prototyped using standard digital hardware. With pre-recorded neural signals as the input, compression rate of 26.15 is achieved while the root-mean-square of error is kept as low as 1.1%.Clinical Relevance- This paper proposes a technique for data compression in high-density neural recording brain implants, along with a power- and area-efficient hardware implementation. From among clinical applications of such implants one can point to neuro-prostheses, and brain-machine interfaces for therapeutic purposes.

Rehabilitation level evaluation is an important part of the automatic rehabilitation training system. As a general rule, this process is manually performed by rehabilitation doctors using chart-based ordinal scales which can be both subjective and inefficient. In this paper, a novel approach based on ensemble learning is proposed which automatically evaluates stroke patients' rehabilitation level using multi-channel sEMG signals to this problem. The correlation between rehabilitation levels and rehabilitation training actions is investigated and actions suitable for rehabilitation assessment are selected. Then, features are extracted from the selected actions. Finally, the features are used to train the stacking classification model. Experiments using sEMG data collected from 24 stroke patients have been carried out to examine the validity and feasibility of the proposed method. The experiment results show that the algorithm proposed in this paper can improve the classification accuracy of 6 Brunnstrom stages to 94.36%, which can promote the application of home-based rehabilitation training in practice.

In clinical conversational applications, extracted entities tend to capture the main subject of a patient's complaint, namely symptoms or diseases. However, they mostly fail to recognize the characterizations of a complaint such as the time, the onset, and the severity. For example, if the input is "I have a headache and it is extreme", state-of-the-art models only recognize the main symptom entity - headache, but ignore the severity factor of "extreme", that characterizes headache. In this paper, we design a two-stage approach to detect the characterizations of entities like symptoms presented by general users in contexts where they would describe their symptoms to a clinician. We use Word2Vec and BERT to encode clinical text given by the patients. We transform the output and re-frame the task as multi-label classification problem. Finally, we combine the processed encodings with the Linear Discriminant Analysis (LDA) algorithm to classify the characterizations of the main entity. Experimental results demonstrate that our method achieves 40-50% improvement on the accuracy over the state-of-the-art models., Comment: This paper has been communicated to 42nd IEEE Annual Conference of Engineering in Medicine and Biology Society, Montreal, Canada. It is 5 pages long. It has 2 figures

This paper has been presented at the 2020 IEEE International Conference on Communications 5G networks aim to provide orchestration of services across multiple administrative domains through the concept of federation. In this paper, we are exploring the federation feature of a platform for 5G transport network of vertical services. Then we formulate the decision problem that directly impacts the revenue of 5G administrative domains, and we propose as solution a Q-learning algorithm. The simulation results show near optimum profit maximization and a well-trained Q-learning algorithm can outperform the intuitive "greedy" approach in a realistic scenario. This work has been partially funded by the EC H2020 5G-TRANSFORMER Project (grant no. 761536) and the EU H2020 5GROWTH Project (grant no. 856709).

© 2022 IEEE.Businesses are concerned about energy losses. Electronic gadgets have become more prevalent as a result of their adoption. The frequency with which home electricity usage data is collected has grown, allowing for sophisticated data analysis that was previously unavailable. Indeed, adopting Smart Grid (SG) networks, which are freshly improved networks of linked devices, may considerably enhance the existing energy infrastructure\"s dependability, economy, and durability. The SG involves sharing a lot of data, including information on specific users\" power use. And using this information, machine learning and deep learning algorithms may be able to detect power theft users reliably. This paper presented a Convolutional Neural Network (CNN)-based model for automated network-based vulnerability scanning that has excellent classification performance in many categories. Testing from iteration two to four iterations, this study examines research to discover the ideal configuration of the sequential model (SM) for categorization. The method is validated using a two-layer design, including an efficiency of 0.92, the whole first layer is composed of 128 nodes while the second level consists of 64 nodes. This allows for the development of a higher-level classifier for electrical signals, which may be employed in a number of applications. CNN was used to create electrical signal detectors, and SM was used to extract data from an electricity usage dataset. Furthermore, the number of features in the data set can be reduced using the Blue Monkey (BM) approach, and these results are then used to develop high-performance models. In this regard, the focus of this study has been on lowering the amount of needed features in the dataset in order to establish a rising classification algorithm for electrical signals. Experiments have applied the proposed systems\" fantastic performance, with just 666 characteristics required to combine the CNN and BM methods. Comparative to 1035 traits when CNN was used alone. This shows that the CNN and BM models are better than the CNN model in terms of lowering sufficient know while maintaining the same reliability.

Despite all the recent developments of using the surface electromyography (sEMG) as a control signal, reliable classifications still remain an arduous task due to overlapping classes and classification ripples. In this paper, we present a straightforward approach to avoid classification ripple based on smoothing the arg max value of an Extreme Learning Machine (ELM) classifier. We compare the baseline accuracy of the classifier with an arg max filtered by a traditional Exponential Smoothing Filter (ESF) and our adaptation of Antonyan Vardan Transform (AVT). The classifiers were evaluated using sEMG data acquired through 12 channels from four subjects performing 17 different movements of forearm and fingers with three repetitions each. In the best scenario, our methods reached results higher than 96% and 82% of overall and weighted accuracy, respectively. Those results match or outperform similar papers of the literature using a simpler model, which may help the application of the techniques on embedded platforms and make the practical use of such devices more feasible.

Clinical assessment of abnormal neuromechanics is typically performed by manipulation of the affected limbs; a process with low inter- and intra-rater reliability. This paper aims at formalizing a framework that closes the loop between a clinician's expertise and computational algorithms, to enhance the clinician's diagnostic capabilities during physical manipulation. The framework's premise is that the dynamics that can be measured by manipulation of a limb are distinct between movement disorders. An a priori database contains measurements encoded in a space called the information map. Based on this map, a computational algorithm identifies which probing motions are more likely to yield distinguishing information about a patient's movement disorder. The clinician executes this movement and the resulting dynamics, combined with clinician input, is used by the algorithm to estimate which of the movement disorders in the database are most probable. This is recursively repeated until a diagnosis can be confidently made. The main contributions of this paper are the formalization of the framework and the addition of the information map to select informative movements. The establishment of the framework provides a foundation for a standardized assessment of movement disorders and future work will aim at testing the framework's efficacy.

This paper presents a voice control interface prototype for assistive robots aiming to help people living with upper limb disabilities to perform daily activities autonomously. Assistive robotic devices can be used to help people with upper-body disabilities gain more autonomy in their daily life. However, it is very difficult or even impossible for certain users to control the robot with conventional control systems (e.g. joystick, sip-and-puff). This paper presents the design and preliminary evaluation of a voice command system prototype for the control of assistive robotic arms' movements. This work aims at making the control of assistive robots more intuitive and fluid, and to perform various tasks in less time and with a lesser effort. The prototype of the voice command interface developed is first presented, followed by two experiments with five able-bodied subjects in order to assess the system's performance and guide future development.

This paper presents the development of a new active assistive eating device, which aims to stabilize the movements of people living with movement disorders, such as spasticity and ataxia. Many people living with upper-body incapacities are unable to eat on their own, due to movement disorders (ex. tremors, spastic motions, lack of muscular tone), resulting from various ailments like Cerebral palsy, Parkinson's disease, Dystonia, Multiple sclerosis, strokes, and Muscular dystrophy). Our past work focused on the development of a purely mechanical device, which involved damping of the system via passive mechanical dampers. This paper extends said work by using active stabilization of user movements. The active assistance enables the design of intelligent algorithms that can assist human movements more efficiently. This active version has the benefits of being easily adjustable; the level of damping can be adjusted in real-time, depending on the user movement; different control modes are offered, and the guiding of user movements is also allowed. Firstly, the mechanical design of the device is presented, followed by the damping arrangement, the electronic design, the control algorithms and finally, the preliminary experiments are mentioned.

Alzheimer's disease (AD) accounts for 60%-70% of all dementia cases, and clinical diagnosis at its early stage is extremely difficult. As several new drugs aiming to modify disease progression or alleviate symptoms are being developed, to assess their efficacy, novel robust biomarkers of brain function are urgently required. This paper aims to explore a routine to gain such biomarkers using the quantitative analysis of electroencephalography (QEEG). This paper proposes a supervised classification framework that uses EEG signals to classify healthy controls (HC) and AD participants. The framework consists of data augmentation, feature extraction, K-nearest neighbor (KNN) classification, quantitative evaluation, and topographic visualization. Considering the human brain either as a stationary or a dynamical system, both the frequency-based and time-frequency-based features were tested in 40 participants. The results show that: 1) the proposed method can achieve up to a 99% classification accuracy on short (4s) eyes open EEG epochs, with the KNN algorithm that has best performance when compared with alternative machine learning approaches 2) the features extracted using the wavelet transform produced better classification performance in comparison to the features based on FFT and 3) in the spatial domain, the temporal and parietal areas offer the best distinction between healthy controls and AD. The proposed framework can effectively classify HC and AD participants with high accuracy, meanwhile offering identification and the localization of significant QEEG features. These important findings and the proposed classification framework could be used for the development of a biomarker for the diagnosis and monitoring of disease progression in AD. 27 5 826 835

Locating the position of a remote node on a wireless network is becoming more relevant, as we move forward in the Internet of things and in autonomous vehicles. This paper proposes a new system to implement the location of remote nodes. A new prototype Android application has been developed to collect real measurements and to study the performance of several smartphone's sensors and location algorithms, including an innovative one, based on the second order cone programming (SOCP) relaxation. The application collects the WiFi access points information and the terminal location. An internal odometry module developed for the prototype is used when Android's service is unavailable. This paper compares the performance of existing location estimators given in closed form, an existing SOCP one, and the new SOCP location estimator proposed, which has reduced complexity. An algorithm to merge measurements from non-identical terminals is also proposed. Cooperative and terminal stand-alone operations are compared, showing a higher performance for SOCP-based ones, that are capable of estimating the path loss exponent and the transmission power. The heterogeneous terminals were also used in the tests. Our results show that the accurate positioning of static remote entities can be achieved using a single smartphone. On the other hand, the accurate real-time positioning of the mobile terminal is provided when three or more scattered terminal nodes cooperate sharing the samples taken synchronously.

Objective: To remove blood from an incision and find the incision spot is a key task during surgery, or else over discharge of blood will endanger a patient’s life. However, the repetitive manual blood removal involves plenty of workload contributing fatigue of surgeons. Thus, it is valuable to design a robotic system which can automatically remove blood on the incision surface. Methods: In this paper, we design a robotic system to fulfill the surgical task of the blood removal. The system consists of a pair of dual cameras, a 6-DoF robotic arm, an aspirator whose handle is fixed to a robotic arm, and a pump connected to the aspirator. Further, a path-planning algorithm is designed to generate a path, which the aspirator tip should follow to remove blood. Results: In a group of simulating bleeding experiments on ex vivo porcine tissue, the contour of the blood region is detected, and the reconstructed spatial coordinates of the detected blood contour is obtained afterward. The BRR robot cleans thoroughly the blood running out the incision. Conclusions: This study contributes the first result on designing an autonomous blood removal medical robot. The skill of the surgical blood removal operation, which is manually operated by surgeons nowadays, is alternatively grasped by the proposed BRR medical robot.

Smart electric vehicles (EVs) are attractive because of their clean, zero-emission, low impact on the environment whilst providing a safer and smoother riding experience. To provide the latter, driving control requires appropriate systems and algorithms to optimize smart vehicle performance, maximize vehicle stability and protection, minimize accident probability, heighten driving comfort, and optimize transportation costs. Despite advancements in these areas, the realization of optimal smart EVs still requires considerable effort. This paper reviews driving control systems and algorithms for smart EVs, including the advanced driving assistant system, implementation of sensors, vehicle dynamics, and control algorithms. The major contribution of this review is to identify promising work to assist researchers with the most advanced trends in this area for prospective regulations.

The multipath challenge is a research line in continuous development because of its multiple benefits, however, these benefits are overshadowed by scalability, which goes down considerably when the paths are multiple and disjoint. The disjointness aggregates an extra value to the multiple paths, but it also implies more complex mathematical operations that increase the computational cost. In fact, diverse proposals exist that try to increase scalability by limiting the number of paths obtained to the minimum possible (two-disjoint paths), which is enough for backup applications but not for other purposes. This paper presents an algorithm that solves these drawbacks by discovering multiple disjoint paths among multiple nodes in an efficient way, while keeping bounded the computational cost and ensuring scalability. The proposed algorithm has been validated thoroughly by performing a theoretical analysis, bolstered afterwards by an exhaustive experimental evaluation. The collected results are promising, our algorithm reduces the time spent to obtain the disjoint paths regarding its competitors between one and three orders of magnitude, at the cost of a slight decrease in the number of paths discovered., Comunidad de Madrid, Junta de Comunidades de Castilla-La Mancha

This paper proposes an electromyogram (EMG) pattern classification method based on a mixture of variance distribution models. A variance distribution model is a stochastic model of raw surface EMG signals in which the EMG variance is taken as a random variable, allowing the representation of uncertainty in the variance. In this paper, we extend the variance distribution model to the multidimensional case and enhance its flexibility for multichannel and processed EMG signals. The enhanced model enables the accurate classification of EMG patterns while considering the uncertainty in the EMG variance. The robustness and applicability of the proposed method are demonstrated through a simulation experiment using artificially generated data and EMG classification experiments using two real datasets.

In Brain-Computer Interface (BCI) Research,Electroencephalography (EEG) has obtained great attention for biomedical applications. In BCI system, feature representation and classification are important tasks as the accuracy of classification highly depends on these stages. In this paper, we propose a model in which Common Spatial Pattern (CSP) is used to discriminate inter-class data using co-variance maximization and Fast Fourier Transform Energy Map (FFTEM) is used for feature selection and mapping of 1D data into 2D data (energy maps). Convolutional Neural Network is used for classification of multi-class Motor Imagery (MI) signals. Further, this paper investigates near-optimal parameter selection for feature mapping, frequency bands selection, and temporal segmentation. It is shown that our proposed method outperformed the reported methods by achieving 0.61 mean kappa value.

In this paper, we present our GPU implementation of the quotient filter, a compact data structure designed to implement approximate membership queries. The quotient filter is similar to the more well-known Bloom filter; however, in addition to set insertion and membership queries, the quotient filter also supports deletions and merging filters without requiring rehashing of the data set. Furthermore, the quotient filter can be extended to include counters without increasing the memory footprint. This paper describes our GPU implementation of two types of quotient filters: the standard quotient filter and the rank-and-select-based quotient filter. We describe the parallelization of all filter operations, including a comparison of the four different methods we devised for parallelizing quotient filter construction. In solving this problem, we found that we needed an operation similar to a parallel scan, but for non-associative operators. One outcome of this work is a variety of methods for computing parallel scan-type operations on a non-associative operator. For membership queries, we achieve a throughput of up to 1.13 billion items/second for the rank-and-select-based quotient filter: a speedup of 3x over the BloomGPU filter. Our fastest filter build method achieves a speedup of 2.1--3.1x over BloomGPU, with a peak throughput of 621 million items/second, and a rate of 516 million items/second for a 70% full filter. However, we find that our filters do not perform incremental updates as fast as the BloomGPU filter. For a batch of 2 million items, we perform incremental inserts at a rate of 81 million items/second -- a 2.5x slowdown compared to BloomGPU's throughput of 201 million items/second. The quotient filter's memory footprint is comparable to that of a Bloom filter.

Current studies on modulation and detection schemes in molecular communication mainly focus on the scenarios with static transmitters and receivers. However, mobile molecular communication is needed in many envisioned applications, such as target tracking and drug delivery. Until now, investigations about mobile molecular communication have been limited. In this paper, a static transmitter and a mobile bacterium-based receiver performing random walk are considered. In this mobile scenario, the channel impulse response changes due to the dynamic change of the distance between the transmitter and the receiver. Detection schemes based on fixed distance fail in signal detection in such a scenario. Furthermore, the intersymbol interference (ISI) effect becomes more complex due to the dynamic character of the signal which makes the estimation and mitigation of the ISI even more difficult. In this paper, an adaptive ISI mitigation method and two adaptive detection schemes are proposed for this mobile scenario. In the proposed scheme, adaptive ISI mitigation, estimation of dynamic distance and the corresponding impulse response reconstruction are performed in each symbol interval. Based on the dynamic channel impulse response in each interval, two adaptive detection schemes, concentration-based adaptive threshold detection (CATD) and peak-time-based adaptive detection (PAD), are proposed for signal detection. Simulations demonstrate that, the ISI effect is significantly reduced and the adaptive detection schemes are reliable and robust for mobile molecular communication.

Restricted Boltzmann Machine (RBM) is the building block of Deep Belief Nets and other deep learning tools. Fast learning and prediction are both essential for practical usage of RBM-based machine learning techniques. This paper proposes Lean Contrastive Divergence (LCD), a modified Contrastive Divergence (CD) algorithm, to accelerate RBM learning and prediction without changing the results. LCD avoids most of the required computations with two optimization techniques. The first is called bounds-based filtering, which, through triangle inequality, replaces expensive calculations of many vector dot products with fast bounds calculations. The second is delta product, which effectively detects and avoids many repeated calculations in the core operation of RBM, Gibbs Sampling. The optimizations are applicable to both the standard contrastive divergence learning algorithm and its variations. In addition, this paper presents how to implement these optimizations effectively on massively parallel processors. Results show that the optimizations can produce several-fold (up to 3X for training and 5.3X for prediction) speedups.

The ability of Machine Learning (ML) algorithms to learn and work with incomplete knowledge has motivated many system manufacturers to include such algorithms in their products. However, some of these systems can be described as Safety-Critical Systems (SCS) since their failure may cause injury or even death to humans. Therefore, the performance of ML algorithms with respect to the safety requirements of such systems must be evaluated before they are used in their operational environment. Although there exist several measures that can be used for evaluating the performance of ML algorithms, most of these measures focus mainly on some properties of interest in the domains where they were developed. For example, Recall, Precision and F-Factor are, usually, used in Information Retrieval (IR) domain, and they mainly focus on correct predictions with less emphasis on incorrect predictions, which are very important in SCS. Accordingly, such measures need to be tuned to fit the needs for evaluating the safe performance of ML algorithms. This position paper presents the authors’ view on the inadequacy of existing measures, and it proposes a new set of measures to be used for the evaluation of the safe performance of ML algorithms.

Nowadays Cloud computing permeates almost every domain in Information and Communications Technology (ICT) and, increasingly, most of the action is shifting from large, dominant players toward independent, heterogeneous, private/hybrid deployments, in line with an ever wider range of business models and stakeholders. The rapid growth in the numbers and diversity of small and medium Cloud providers is bringing new challenges in the as-a-Services space. Indeed, significant hurdles for smaller Cloud service providers in being competitive with the incumbent market leaders induce some innovative players to “federate” deployments in order to pool a larger, virtually limitless, set of resources across the federation, and stand to gain in terms of economies of scale and resource usage efficiency. Several are the challenges that need to be addressed in building and managing a federated environment, that may go under the “Security”, “Interoperability”, “Versatility”, “Automatic Selection” and “Scalability” labels. The aim of this paper is to present a survey about the approaches and challenges belonging to the “Automatic Selection” category. This work provides a literature review of different approaches adopted in the “Automatic and Optimal Cloud Service Provider Selection”, also covering “Federated and Multi-Cloud” environments.

Nowadays, we are surrounded by devices collecting and transmitting private information. Currently, the two main mathematical problems that guarantee security on the Internet are the Integer Factorization Problem and the Discrete Logarithm Problem. However, Shor’s quantum algorithm can easily solve both problems. Therefore, research into cryptographic algorithms that run in classical computers and are resistant to quantum computers is extremely necessary. This area is known as post-quantum cryptography and usually studies asymmetric cryptography. By means of asymptotic analysis, the purpose of this paper is to provide an evaluation of security and its performance for the types of cryptographic systems considered safe against quantum attacks in the second-round NIST Post-Quantum Standardization Process, namely isogeny cryptosystems based on supersingular elliptic curves, error correction code-based encryption system, and lattice-based ring learning with errors. We performed a security comparison of Key Agreements protocols based on these three post-quantum cryptographic primitives and compared them with Discrete Logarithm Problem and Integer Factorization Problem. The comparison of security and its performance is presented by security level, the former by complexity analyses to achieve theoretical minimum key sizes, and the latter by simulation to assess a practical performance comparison. In the complexity analysis, as we increase the security level and then the size of the cryptographic keys increases, techniques based on isogeny outperform all other post-quantum algorithms in relation to key sizes at practical security level. In the performance comparison, the results show that the code-based protocol presents the best results among the others.

Typical methods for solving reinforcement learning problems iterate two steps, policy evaluation and policy improvement. This paper proposes algorithms for the policy evaluation to improve learning efficiency. The proposed algorithms are based on the Krylov Subspace Method (KSM), which is a nonstationary iterative method. The algorithms based on KSM are tens to hundreds times more efficient than existing algorithms based on the stationary iterative methods. Algorithms based on KSM are far more efficient than they have been generally expected. This paper clarifies what makes algorithms based on KSM makes more efficient with numerical examples and theoretical discussions.

International audience; Process mining techniques aim at analyzing and monitoring processes through event data. Formal models like Petri nets serve as an effective representation of the processes. A central question in the field is to assess the conformance of a process model with respect to the real process executions. The notion of anti-alignment, which represents a model run that is as distant as possible to the process executions, has been demonstrated to be crucial to measure precision of models. However, the only known algorithm for computing anti-alignments has a high complexity, which prevents it from being applied on real-life problem instances. In this paper we propose a novel algorithm for computing anti-alignments, based on the well-known graphbased A * scheme. By introducing a discount factor in the edit distance used for the search of anti-alignments, we obtain the first efficient algorithm to approximate them. We show how this approximation is quite accurate in practice, by comparing it with the optimal results for small instances where the optimal algorithm can also compute anti-alignments. Finally, we compare the obtained precision metric with respect to the state-of-the-art metrics in the literature for real-life examples.

Statistical analysis using large-scale brain magnetic resonance (MR) image databases has examined that brain tissues have age-related morphological changes. The age of a subject can be estimated from the brain MR image by evaluating morphological changes with healthy aging. This paper proposes an age estimation method using local features of T1-weighted MR images. The brain local features are defined by volumes of brain tissues parcellated into 1,024 local regions defined by the automated anatomical labeling atlas. This paper also proposes the effective local feature selection method to improve the accuracy of age estimation. We evaluate the accuracy of the proposed method using 1,099 T1-weighted images from a Japanese MR image database. We also analyze effectiveness of each local region for age estimation and discuss its medical implication.

Support Vector Machine (SVM) is a common classifier used for efficient classification with high accuracy. SVM shows high accuracy for classifying melanoma (skin cancer) clinical images within computer-aided diagnosis systems used by skin cancer specialists to detect melanoma early and save lives. We aim to develop a medical low-cost handheld device that runs a real-time embedded SVM- based diagnosis system for use in primary care for early detection of melanoma. In this paper, an optimized SVM classifier is implemented onto a recent FPGA platform using the latest design methodology to be embedded into the proposed device for realizing online efficient melanoma detection on a single system on chip/device. The hardware implementation results demonstrate a high classification accuracy of 97.9% and a significant acceleration factor of 26 from equivalent software implementation on an embedded processor, with 34% of resources utilization and 2 watts for power consumption. Consequently, the implemented system meets crucial embedded systems constraints of high performance and low cost, resources utilization and power consumption, while achieving high classification accuracy., Conference paper, 5 pages, 4 figures, 1 tables

This paper considers the auditory attention detection (AAD) paradigm, where the goal is to determine which of two simultaneous speakers a person is attending to. The paradigm relies on recordings of the listener's brain activity, e.g., from electroencephalography (EEG). To perform AAD, decoded EEG signals are typically correlated with the temporal envelopes of the speech signals of the separate speakers. In this paper, we study how the inclusion of various degrees of auditory modelling in this speech envelope extraction process affects the AAD performance, where the best performance is found for an auditory-inspired linear filter bank followed by power law compression. These two modelling stages are computationally cheap, which is important for implementation in wearable devices, such as future neuro-steered auditory prostheses. We also introduce a more natural way to combine recordings (over trials and subjects) to train the decoder, which reduces the dependence of the algorithm on regularization parameters. Finally, we investigate the simultaneous design of the EEG decoder and the audio subband envelope recombination weights vector using either a norm-constrained least squares or a canonical correlation analysis, but conclude that this increases computational complexity without improving AAD performance. ispartof: IEEE Transactions on Neural Systems and Rehabilitation Engineering vol:25 issue:2 pages:1-11 ispartof: location:United States status: published

Tensors, or multidimensional arrays, are data structures that can naturally represent visual data of multiple dimensions. Inherently able to efficiently capture structured, latent semantic spaces and high-order interactions, tensors have a long history of applications in a wide span of computer vision problems. With the advent of the deep learning paradigm shift in computer vision, tensors have become even more fundamental. Indeed, essential ingredients in modern deep learning architectures, such as convolutions and attention mechanisms, can readily be considered as tensor mappings. In effect, tensor methods are increasingly finding significant applications in deep learning, including the design of memory and compute efficient network architectures, improving robustness to random noise and adversarial attacks, and aiding the theoretical understanding of deep networks. This article provides an in-depth and practical review of tensors and tensor methods in the context of representation learning and deep learning, with a particular focus on visual data analysis and computer vision applications. Concretely, besides fundamental work in tensor-based visual data analysis methods, we focus on recent developments that have brought on a gradual increase of tensor methods, especially in deep learning architectures, and their implications in computer vision applications. To further enable the newcomer to grasp such concepts quickly, we provide companion Python notebooks, covering key aspects of the paper and implementing them, step-by-step with TensorLy., Comment: Proceedings of the IEEE (2021)

Phenotypic heritability of complex traits and diseases is seldom explained by individual genetic variants identified in genome-wide association studies (GWAS). Many methods have been developed to select a subset of variant loci, which are associated with or predictive of the phenotype. Selecting SNPs that are close on a biological network such as SNP-SNP networks have been proven successful in finding biologically interpretable and predictive SNPs. However, we argue that the closeness constraint favors selecting redundant features that affect similar biological processes and therefore does not necessarily yield better predictive performance. An approach, which awards diversity of the selected SNPs and affected functional processes, would boost the predictive power without compromising biological interpretability. In this paper, we propose a novel method called SPADIS that selects a set of loci such that diverse regions in the underlying SNP-SNP network are covered. Instead of enforcing selections based on closeness in the network, SPADIS favors the selection of remotely located SNPs in order to account for the complementary additive effects of SNPs that are associated with the phenotype. This is achieved by maximizing a submodular set function with a greedy algorithm that ensures a constant factor (1 - 1/e) approximation to the optimal solution. We compare SPADIS to the state-of-the-art method SConES, on a dataset of Arabidopsis Thaliana genotype and continuous flowering time phenotypes. SPADIS has better regression performance in 12 out of 17 phenotypes on average, it identifies more candidate genes and runs faster. We also investigate the use of Hi-C data to construct SNP-SNP network in the context of SNP selection problem for the first time, which yields slight improvements in regression performance. SPADIS is available at http://ciceklab.cs.bilkent.edu.tr/spadis.

Pain sensation is essential for survival, since it draws attention to physical threat to the body. Pain assessment is usually done through self-reports. However, self-assessment of pain is not available in the case of noncommunicative patients, and therefore, observer reports should be relied upon. Observer reports of pain could be prone to errors due to subjective biases of observers. Moreover, continuous monitoring by humans is impractical. Therefore, automatic pain detection technology could be deployed to assist human caregivers and complement their service, thereby improving the quality of pain management, especially for noncommunicative patients. Facial expressions are a reliable indicator of pain, and are used in all observer-based pain assessment tools. Following the advancements in automatic facial expression analysis, computer vision researchers have tried to use this technology for developing approaches for automatically detecting pain from facial expressions. This paper surveys the literature published in this field over the past decade, categorizes it, and identifies future research directions. The survey covers the pain datasets used in the reviewed literature, the learning tasks targeted by the approaches, the features extracted from images and image sequences to represent pain-related information, and finally, the machine learning methods used.

A novel sensor placement method for better removal of motion artifacts (MA) from photoplethysmography (PPG) signal using Fast Independent Component Analysis (ICA) is proposed in this paper. The method enhances the determination of pulse transit time (PTT) of PPG signals. The design makes use of double reflectance mode based PPG probes, which are placed complementary to each other and on the two sides of a single finger. Furthermore, a novel indicator denoted as Separating Factor is proposed as well. It helps evaluating the performance of ICA with different sensors configuration. This paper then addresses the effectiveness of FastICA in motion artifacts reduction by using the novel method and normal sensors placement method to capture PPG. Results indicate that better independent source separation can be achieved and morphology of PPG signal is perfectly restored when using the method proposed in this paper.

The design of communication systems based on the transmission of information through neurons is envisioned as a key technology for the pervasive interconnection of future wearable and implantable devices. While previous literature has mainly focused on modeling propagation of electrochemical spikes carrying natural information through the nervous system, in recent work the authors of this paper proposed the so-called subthreshold electrical stimulation as a viable technique to propagate artificial information through neurons. This technique promises to limit the interference with natural communication processes, and it can be successfully approximated with linear models. In this paper, a novel model is proposed to account for the subthreshold stimuli propagation from the dendritic tree to the soma of a neuron. A computational approach is detailed to obtain this model for a given realistic 3D dendritic tree with an arbitrary morphology. Numerical results from the model are obtained over a stimulation signal bandwidth of 1KHz, and compared with the results of a simulation through the NEURON software.

Impulse radio ultra-wide band (IR-UWB) radar has recently emerged as a promising candidate for non-contact monitoring of respiration and heart rate. Different studies have reported various radar based algorithms for estimation of these physiological parameters. The radar can be placed under a subject's mattress as he lays stationary on his back or it can be attached to the ceiling directly above the subject's bed. However, advertent or inadvertent movement on part of the subject and different postures can affect the radar returned signal and also the accuracy of the estimated parameters from it. The detection and analysis of these postural changes can not only lead to improvement in estimation algorithms but also towards prevention of bed sores and ulcers in patients who require periodic posture changes. In this paper, we present an algorithm that detects and quantifies different types of motion events using an under-the-mattress IR-UWB radar. The algorithm also indicates a change in posture after a macro-movement event. Based on the findings of this paper, we anticipate that IR-UWB radar can be used for extracting posture related information in non-clinical enviroments for patients who are bed-ridden.

This paper presents a deep-pipelined FPGA implementation of real-time ellipse estimation for eye tracking. The system is constructed by the Starburst algorithm on a stream-oriented architecture and the RANSAC algorithm without any external memories. In particular, the paper presents comparative results between three different hypothesis generators for the RANSAC algorithm based on Cramer's rule, Gauss-Jordan elimination and LU decomposition. Comparison criteria include resource usage, throughput and energy consumption. The result shows that the three implementations have different characteristics and the optimal algorithm needs to be chosen depending on the amount of resources on FPGAs and required performance., 22nd International Conference on Field Programmable Logic and Applications, FPL 2012;Oslo;29 August 2012through31 August 2012, Field Programmable Logic and Applications (FPL), 2012 22nd International Conference on, pp.458-463; 2012

This paper investigates how to train a recurrent neural network (RNN) using the Levenberg–Marquardt (LM) algorithm as well as how to implement optimal control of a grid-connected converter (GCC) using an RNN. To successfully and efficiently train an RNN using the LM algorithm, a new forward accumulation through time (FATT) algorithm is proposed to calculate the Jacobian matrix required by the LM algorithm. This paper explores how to incorporate FATT into the LM algorithm. The results show that the combination of the LM and FATT algorithms trains RNNs better than the conventional backpropagation through time algorithm. This paper presents an analytical study on the optimal control of GCCs, including theoretically ideal optimal and suboptimal controllers. To overcome the inapplicability of the optimal GCC controller under practical conditions, a new RNN controller with an improved input structure is proposed to approximate the ideal optimal controller. The performance of an ideal optimal controller and a well-trained RNN controller was compared in close to real-life power converter switching environments, demonstrating that the proposed RNN controller can achieve close to ideal optimal control performance even under low sampling rate conditions. The excellent performance of the proposed RNN controller under challenging and distorted system conditions further indicates the feasibility of using an RNN to approximate optimal control in practical applications.

Estimation of neural models based on observed spike timing faces challenges as the amount of recorded units increases, especially when identifying detailed model features. Given that neural regions are generally sparsely connected, input selection is a critical step in model estimation but oftentimes computationally and theoretically challenging. In this paper, we detail an efficient methodology for estimating a sparse, nonlinear dynamical multiple-input, single-output model (MISO) applicable to large-scale (n > 50) single-unit recorded activity. The main contribution of this paper is the complete implementation of a principled group-lasso and local coordinate descent (LCD) algorithm into a generalized Volterra model (GVM) framework to achieve efficient sparse model estimation. Input selection is achieved with group-lasso by simultaneously selecting groups of parameters that are associated with each input. LCD yields efficient computation as the amount of inputs and parameters increase. We investigate and validate the performance of this estimation procedure with the application to a 64 input simulated model.

This paper addresses the problem of automatic detection of tumoral frames in endoscopic capsule videos by using features directly extracted from the color space. We show that tumor can be appropriately discriminated from normal tissue by using only color information histogram measures from the Lab color space and that light saturated regions are usually classified as tumoral regions when color based discriminative procedures are used. These regions are correctly classified if lightening is discarded becoming the tissue classifier based only on the color differences a and b of the Lab color space. While current state of the art systems for small bowel tumor detection usually rely on the processing of the whole frame regarding features extraction this paper proposes the use of fully automatic segmentation in order to select regions likely to contain tumoral tissue. Classification is performed by using Support Vector Machine (SVM) and Multilayer Perceptron (MLP) by using features from color channels a and b of the Lab color space. The proposed algorithm outperforms in more than 5% a series of other algorithms based on features obtained from the higher frequency components selected from Wavelets and Curvelets transforms while saving important computational resources. In a matter of fact the proposed algorithm is more than 25 times faster than algorithms requiring wavelet/curvelet and co-occurrence computations.

This paper presents QRS complex detection algorithm based on dual slope technique, which is suitable for wearable electrocardiogram (ECG) applications. For cardiac patients of different arrhythmias, ECG signals are needed to be monitored over an extensive period of time. Thus, the wearable heart monitoring system needs computationally efficient QRS detection technique with good accuracy. In this paper, a method of QRS detection based on two slopes on both sides of an R peak is presented which is computationally efficient. Based on the slopes, first, a variable measuring steepness is developed, then by introducing an adjustable R-R interval based window and adaptive thresholding techniques, depending on the number of peaks detected in such window, R peaks are detected. The algorithm was evaluated against MIT/BIH arrhythmia database and achieved 99.16% detection rate with sensitivity of 0.9935 and positive predictivity of 0.9981. The method was compared with two widely used R peaks detection algorithms.

This paper explores hardware specialization of low­ power processors to improve performance and energy efficiency. Our main contribution is an automated framework that analyzes instruction sequences of applications within a domain at the loop body level and identifies exactly and partially-matching sequences across applications that can become custom instructions. Our framework transforms sequences to a new code abstraction, a Merging Diagram, that improves similarity identification, clusters alike groups of potential custom instructions to effectively reduce the search space, and selects merged custom instructions to efficiently exploit the available customizable area. For a set of 11 media applications, our fast framework generates instructions that significantly improve the energy-delay product and speed­ up, achieving more than double the savings as compared to a technique analyzing sequences within basic blocks. This paper shows that partially-matched custom instructions, which do not significantly increase design time, are crucial to achieving higher energy efficiency at limited hardware areas.

In this paper, a non-contact respiration detection scheme based on Doppler radar-depth camera sensor fusion has been proposed. A continuous-wave (CW) Doppler radar sensor and a depth camera are used to measure the respiratory motion separately. Then the Bayesian sensor fusion algorithm is used to estimate the cycle-to-cycle breathing rate. The experiments prove that the proposed fusion scheme can provide an accurate breathing rate estimation than using a single sensor. In particular, the proposed scheme can give a reasonable estimation even under the influence of body movement.

Magnetic resonance fingerprinting is a recent quantitative MRI technique that simultaneously acquires multiple tissue parameter maps (e.g., T(1), T(2), and spin density) in a single imaging experiment. In our early work, we demonstrated that the low-rank/subspace reconstruction significantly improves the accuracy of tissue parameter maps over the conventional MR fingerprinting reconstruction that utilizes simple pattern matching. In this paper, we generalize the low-rank/subspace reconstruction by introducing a multilinear low-dimensional image model (i.e., a low-rank tensor model). With this model, we further estimate the subspace associated with magnetization evolutions to simplify the image reconstruction problem. The proposed formulation results in a nonconvex optimization problem which we solve by an alternating minimization algorithm. We evaluate the performance of the proposed method with numerical experiments, and demonstrate that the proposed method improves the conventional reconstruction method and the state-of-the-art low-rank reconstruction method.