Binary Output Research Articles

An Extended Version of the Proportional Adaptive Algorithm Based on Kernel Methods for Channel Identification with Binary Measurements

In recent years, kernel methods have provided an important alternative solution, as they offer a simple way of expanding linear algorithms to cover the non-linear mode as well. In this paper, we propose a novel recursive kernel approach allowing to identify the finite impulse response (FIR) in non-linear systems, with binary value output observations. This approach employs a kernel function to perform implicit data mapping. The transformation is performed by changing the basis of the data In a high-dimensional feature space in which the relations between the different variables become linearized. To assess the performance of the proposed approach, we have compared it with two other algorithms, such as proportionate normalized least-meansquare (PNLMS) and improved PNLMS (IPNLMS). For this purpose, we used three measurable frequency-selective fading radio channels, known as the broadband radio access Network (BRAN C, BRAN D, and BRAN E), which are standardized by the European Telecommunications Standards Institute (ETSI), and one theoretical frequency selective channel, known as the Macchi’s channel. Simulation results show that the proposed algorithm offers better results, even in high noise environments, and generates a lower mean square error (MSE) compared with PNLMS and IPNLMS.

Treatment Discontinuation Prediction in Patients With Diabetes Using a Ranking Model: Machine Learning Model Development

Background Treatment discontinuation (TD) is one of the major prognostic issues in diabetes care, and several models have been proposed to predict a missed appointment that may lead to TD in patients with diabetes by using binary classification models for the early detection of TD and for providing intervention support for patients. However, as binary classification models output the probability of a missed appointment occurring within a predetermined period, they are limited in their ability to estimate the magnitude of TD risk in patients with inconsistent intervals between appointments, making it difficult to prioritize patients for whom intervention support should be provided. Objective This study aimed to develop a machine-learned prediction model that can output a TD risk score defined by the length of time until TD and prioritize patients for intervention according to their TD risk. Methods This model included patients with diagnostic codes indicative of diabetes at the University of Tokyo Hospital between September 3, 2012, and May 17, 2014. The model was internally validated with patients from the same hospital from May 18, 2014, to January 29, 2016. The data used in this study included 7551 patients who visited the hospital after January 1, 2004, and had diagnostic codes indicative of diabetes. In particular, data that were recorded in the electronic medical records between September 3, 2012, and January 29, 2016, were used. The main outcome was the TD of a patient, which was defined as missing a scheduled clinical appointment and having no hospital visits within 3 times the average number of days between the visits of the patient and within 60 days. The TD risk score was calculated by using the parameters derived from the machine-learned ranking model. The prediction capacity was evaluated by using test data with the C-index for the performance of ranking patients, area under the receiver operating characteristic curve, and area under the precision-recall curve for discrimination, in addition to a calibration plot. Results The means (95% confidence limits) of the C-index, area under the receiver operating characteristic curve, and area under the precision-recall curve for the TD risk score were 0.749 (0.655, 0.823), 0.758 (0.649, 0.857), and 0.713 (0.554, 0.841), respectively. The observed and predicted probabilities were correlated with the calibration plots. Conclusions A TD risk score was developed for patients with diabetes by combining a machine-learned method with electronic medical records. The score calculation can be integrated into medical records to identify patients at high risk of TD, which would be useful in supporting diabetes care and preventing TD.

An application of CART algorithms for detection of an association between VDR polymorphisms and reduced bone density in individuals with type 2 diabetes: a population-based cross-sectional study

Introduction: An important part of preventing major common diseases is identifying genetic factors that contribute to their occurrence. For the first time in our knowledge, we investigated the association between five polymorphisms of vitamin D receptor (VDR) gene (ApaI, BsmI, FokI, EcoRV, and TaqI) and low bone density/osteopenia/osteoporosis in individuals with type 2 diabetes using classification and regression tree (CART) algorithms. Methods: Data from 158 participants with T2D were used to develop the CART analysis. The binary output variable was "bone state" with low or normal values. Age and BMI (continuous variables), vitamin D deficiency (yes/no), and gender (binary variables), as well as the studied polymorphism of the VDR gene (categorical variables) all played a role in the explanatory model. A 5-fold cross-validation process was used for model validation. Results: Participants were divided into three groups: men, women, and both sexes. In all groups, age was the major factor predicting the low state in the final obtained tree model. The second most significant predictor in each model was BMI in both sexes (accuracy:75.30% ± 2.80%, AUC: 0.740 ± 0.064), EcoRV polymorphism in women (accuracy: 80.79% ± 6.58%, AUC:0.785 ± 0.063), and TaqI polymorphism in men (accuracy: 76.36% ± 3.05%, AUC:0.706 ± 0.125). Conclusion: Model validation of the final tree models demonstrated that the use of CART algorithms could be an acceptable technique for risk factors of osteoporosis among individuals with T2D. Our recommendation is to conduct more population-based studies. We hope this study will serve as a basis for future research.

An Application of Surrogate and Resampling for the Optimization of Success Probability from Binary-Response Type Simulation

Since traditional derivative-based optimization for noisy simulation shows bad performance, evolutionary algorithms are considered as substitutes. Especially in case when outputs are binary, more simulation trials are needed to get near-optimal solution since the outputs are discrete and have high and heterogeneous variance. In this paper, we propose a genetic algorithm called SARAGA which adopts dynamic resampling and fitness approximation using surrogate. SARAGA reduces unnecessary numbers of expensive simulations to estimate success probabilities estimated from binary simulation outputs. SARAGA allocates number of samples to each solution dynamically and sometimes approximates the fitness without additional expensive experiments. Experimental results show that this novel approach is effective and proper hyper parameter choice of surrogate and resampling can improve the performance of algorithm.

A Universal Multichannel Platform for Assembling Enzyme‐Based Boolean Logic Gates

Concatenated enzyme-based Boolean logic gates activated with 5 chemical input signals were analyzed with a smartphone photo camera. Simultaneous detection of 32 input combinations was conveniently performed using enzyme-modified fiberglass sensing spots generating fluorescence with different intensities for the 0 and 1 binary outputs. The developed technology offers an easy readout method for multi-channel logic systems.

Multi-Wavelength Quantum Key Distribution Emulation with Physical Unclonable Function

This work details the theory and implementation of a multi-wavelength quantum key distribution (QKD) emulation system with a physical unclonable function (PUF). Multi-wavelength QKD can eliminate the need to share a subsection of the final key for eavesdropper detection and allow for ternary and quaternary data transmission. The inclusion of the PUF adds an additional layer of security. We provide preliminary error analysis of our emulation system. To support this work, we introduce a bitwise transform operator that enables binary output of the PUF to satisfy the ternary and quaternary input requirements of the QKD system.

An Improved POD Model for Fast Semi-Quantitative Analysis of Carbendazim in Fruit by Surface Enhanced Raman Spectroscopy.

The current detection method of carbendazim suffers from the disadvantages of complicated preprocessing and long cycle time. In order to solve the problem of rapid quantitative screening of finite contaminants, this article proposed a qualitative method based on characteristic peaks and a semi-quantitative method based on threshold to detect carbendazim in apple, and finally the method is evaluated by a validation system based on binary output. The results showed that the detection limit for carbendazim was 0.5 mg/kg, and the detection probability was 100% when the concentration was no less than 1 mg/kg. The semi-quantitative analysis method had a false positive rate of 0% and 5% at 0.5 mg/kg and 2.5 mg/kg, respectively. The results of method evaluation showed that when the added concentration was greater than 2.5 mg/kg, the qualitative detection method was consistent with the reference method. When the concentration was no less than 5 mg/kg, the semi-quantitative method is consistent between different labs. The semi-quantitative method proposed in this study can achieve the screening of finite contaminants in blind samples and simplify the test validation process through the detection probability model, which can meet the needs of rapid on-site detection and has a good application prospect.

Статистичне оцінювання якості освітньої програми “Економічна аналітика та статистика”, рівня задоволеності навчанням та факторів, що їх визначають

In today’s changing world, training of specialists in any sphere requires a constant active search on the part of educational service providers for the development of relevant competencies of students of higher education, to ensure the correspondence of educational training to the modern needs of practice. That is why any educational program needs not only an internal but also an external audit aimed at its improvement. Understanding the factors that determine the quality of the program by chance is the basis not only for building a sustainable and competitive system of training specialists, but also for its successful practical implementation. Such a system should be based on providing feedback to students. One of the forms of such communication is a survey of students and graduates regarding the quality of the educational program and the level of satisfaction with education to find the factors that determine them. To achieve this goal, an expert questionnaire was developed, which includes three blocks of questions regarding the completeness of the program itself, its implementation in the educational process and in practice. A list of questionnaire questions is proposed, which potentially outline the range of problems of student training and possible ways to solve them.
 To evaluate the quality of the educational program in different blocks of disciplines, its main advantages and disadvantages, comparative diagrams of factors that influenced the decision-making regarding admission were constructed. Considerable attention is paid to the comparative analysis of the distribution of grades depending on the educational level of training: bachelors, masters, graduates.
 The implementation of the program in the educational process is considered in detail, namely: the level of teaching disciplines, their methodical support, and the evaluation system. The observance of the rules of academic integrity by teachers and students studying in the educational program and the organization of the educational process are analyzed separately. Questions were also asked that were supposed to reveal the need to implement certain actions to become a good specialist.
 Special attention is paid to the retrospective evaluation of the educational program, which is a generalized measure of its quality, because it summarizes the respondents’ answers to the important question: “If you had the information, you now have about studying in this program at the university before entering, would you choose it?”. For this purpose, the authors proposed an approach to the analysis of the level of satisfaction with education using binary output models. The change in the probability of the outcome variables when the predictors change, and the reliability of the models were also evaluated. The capabilities of binary output models in confirming analytical assumptions are shown.
 Proposals have been put forward regarding the possibilities of improving students’ attitudes towards the educational program and eliminating existing shortcomings.

Hyperspectral imaging for green pepper segmentation using a complex-valued neural network

Background:Green pepper segmentation is an intractable issue due to its similar color with large amounts of green leaves as background. For the above reason, many existing works incorporate hyperspectral imaging as a powerful tool with richer information than RGB components. However, these methods are all under real values. Objective:This paper mainly investigates the possibility of leveraging a complex-valued neural network(cNN) for hyperspectral classification problem, to get better results for green pepper automatic picking in agriculture. Method:Firstly, we obtain several hyperspectral images in the greenhouse and make pixel-wise datasets as hyperspectral input for the following classification network. Secondly and most importantly, we make a novel attempt by leveraging a small cNN as classification module for binary output, which may be better suited for wave propagation problems. Results:Experimental results have showcased that cNN can generate illumination-robust and clearer outputs with less noise. As for the quantitative results, the outcomes of cNN outperform those from a real-valued one under the same configuration by 3.48 percents in terms of accuracy.

Relaxation LIF: A gradient-based spiking neuron for direct training deep spiking neural networks

Spiking neural networks (SNNs) is a promising learning model due to its computational efficiency for discrete spike events. However, because of the binary output of spiking neurons, the standard backpropagation (BP) method is not suitable for deep SNN training. In this paper, we design a gradient-based spiking neuron named Relaxation Leaky Integrate-and-Fire (RLIF) neuron to enable the direct training of deep SNNs. Specifically, we develop a Re-weighted Straight-Through-Estimator (Re-STE) to model the RLIF neuron and SNNs. The Re-STE decomposes the output of the famous LIF neuron and generates a spike by iteratively re-weighting the membrane potential (MP). Then, the RLIF neuron is built, and the re-weighted MP provides a surrogate gradient for BP training. We theoretically prove that the Re-STE is an approximation of the STE technique with logarithmic activation. Such property guarantees that the designed RLIF neuron mitigates the mismatching gradient problem within limits. Finally, we design the RLIF-based VGG-like and ResNet-like SNNs for classification tasks. Experimental results show that the RLIF-based SNNs perform well on the popular benchmark datasets (MNIST, CIFAR10, CIFAR100, N-MNIST, CIFAR10-DVS, and DVS128-GESTURE) and achieve significant efficiencies in most cases.

Radial Peripapillary Capillary Plexus Perfusion and Endothelial Dysfunction in Early Post-SARS-CoV-2 Infection.

Background: Endothelial cells damage and thromboinflammation are considered key elements in the generation of organ impairment in patients with COVID-19 disease. The endothelial function is evaluated by measuring flow-mediated dilation (FMD). We aimed to analyze the association between FMD impairment and retinal vascular parameters in early post-COVID-19 patients. 00118-00199Tomography (OCT), OCT Angiography (OCTA) and slit lamp examination were performed. FMD ≤ 7% was considered as pathological. Our primary outcome was to assess potential differences in the radial peripapillary capillary plexus flow index (RPCP-FI) and RPCP density (RPCP-D) values between post-COVID-19 patients with and without FMD impairment. The associations of other retinal vascular parameters with FMD impairment were assessed as secondary endpoints. Results: FMD impairment was detected in 31 patients (37.8%). RPCP-FI (p = 0.047), age (p = 0.048) and prevalence of diabetes (p = 0.046) significantly differed in patients with FMD ≤ 7% in regression analysis. RPCP-FI was linearly correlated with FMD values (R = 0.244, p =0.027). SCT was found to be lower in patients with impaired FMD (p = 0.004), although this difference was only a trend in binary logistic regression output (p = 0.07). Conclusions: Early post-COVID-19 patients showed a higher prevalence of FMD impairment compared to the general population. Age, diabetes and RPCP-FI were independently correlated with the presence of endothelial impairment in the early post-infective period.

A Deep Learning Approach for the Detection of COVID-19 from Chest X-Ray images using Convolutional Neural Networks

The COVID-19 (coronavirus) is an ongoing pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The virus was first identified in mid-December 2019 in the Hubei province of Wuhan, China and by now has spread throughout the planet with more than 75.5 million confirmed cases and more than 1.67 million deaths. With limited number of COVID-19 test kits available in medical facilities, it is important to develop and implement an automatic detection system as an alternative diagnosis option for COVID-19 detection that can used on a commercial scale. Chest X-ray is the first imaging technique that plays an important role in the diagnosis of COVID-19 disease. Computer vision and deep learning techniques can help in determining COVID-19 virus with Chest X-ray Images. Due to the high availability of large-scale annotated image datasets, great success has been achieved using convolutional neural network for image analysis and classification. In this research, we have proposed a deep convolutional neural network trained on five open access datasets with binary output: Normal and Covid. The performance of the model is compared with four pre-trained convolutional neural network- based models (COVID-Net, ResNet18, ResNet and MobileNet-V2) and it has been seen that the proposed model provides better accuracy on the validation set as compared to the other four pre-trained models. This research work provides promising results which can be further improvise and implement on a commercial scale.

ASLNet: An Encoder-Decoder Architecture for Audio Splicing Detection and Localization

Audio splicing means inserting an audio segment into another audio, which presents a great challenge to audio forensics. In this paper, a novel audio splicing detection and localization method based on an encoder-decoder architecture (ASLNet) is proposed. Firstly, an audio clip is divided into several small audio segments according to the size of the smallest localization region L slr , and the acoustic feature matrix and corresponding binary ground truth mask are created from each audio segment. Then, we concatenate acoustic feature matrices from all segments of an audio clip into an acoustic feature matrix and send it to a fully convolutional network (FCN) based encoder-decoder architecture which consists of a series of convolutional, pooling and transposed convolutional layers to get a binary output mask. Next, the binary output mask is divided into small segments according to the L slr , and the ratio ρ of the number of elements equal to one to the number of all elements in a small segment is calculated. Finally, we compare ρ with the predetermined threshold T to determine whether the corresponding audio segment is spliced. We evaluate the effectiveness of the proposed ASLNet on four datasets produced from publicly available speech corpus. Extensive experiments show that the best detection accuracy of ASLNet for the intradatabase and cross-database evaluation can achieve 0.9965 and 0.9740 receptively, which outperforms the state-of-the-art method.

Differentiable Content Addressable Memory with Memristors

AbstractMemristors, Flash, and related nonvolatile analog device technologies offer in‐memory computing structures operating in the analog domain, such as accelerating linear matrix operations in array structures. These take advantage of analog tunability and large dynamic range. At the other side, content addressable memories (CAM) are fast digital lookup tables which effectively perform nonlinear Boolean logic and return a digital match/mismatch value. Recently, nonvolatile analog CAMs have been presented merging analog storage and analog search operations with digital match/mismatch output. However, CAM blocks cannot easily be inserted within a larger adaptive system due to the challenges of training and learning with binary outputs. Here, a missing link between analog crossbar arrays and CAMs, namely a differentiable content addressable memory (dCAM), is presented. Utilizing nonvolatile memories that act as a “soft” memory with analog outputs, dCAM enables learning and fine‐tuning of the memory operation and performance. Four applications are quantitatively evaluated to highlight the capabilities: improved data pattern storage, improved robustness to noise and variability, reduced energy and latency performance, and an application to solving Boolean satisfiability optimization problems. The use of dCAM is envisioned as a core building block of fully differentiable computing systems employing multiple types of analog compute operations and memories.

Specification of leaf dorsiventrality via a prepatterned binary readout of a uniform auxin input.

Developmental boundaries play an important role in coordinating the growth and patterning of lateral organs. In plants, specification of dorsiventrality is critical to leaf morphogenesis. Despite its central importance, the mechanism by which leaf primordia acquire adaxial versus abaxial cell fates to establish dorsiventrality remains a topic of much debate. Here, by combining time-lapse confocal imaging, cell lineage tracing and molecular genetic analyses, we demonstrate that a stable boundary between adaxial and abaxial cell fates is specified several plastochrons before primordium emergence when high auxin levels accumulate on a meristem prepattern formed by the AS2 and KAN1 transcription factors. This occurrence triggers a transient induction of ARF3 and an auxin transcriptional response in AS2-marked progenitors that distinguishes adaxial from abaxial identity. As the primordium emerges, dynamic shifts in auxin distribution and auxin-related gene expression gradually resolve this initial polarity into the stable regulatory network known to maintain adaxial-abaxial polarity within the developing organ. Our data show that spatial information from an AS2-KAN1 meristem prepattern governs the conversion of a uniform auxin input into an ARF-dependent binary auxin response output to specify adaxial-abaxial polarity. Auxin thus serves as a single morphogenic signal that orchestrates distinct, spatially separated responses to coordinate the positioning and emergence of a new organ with its patterning.

Federal SNN Distillation: A Low-Communication-Cost Federated Learning Framework for Spiking Neural Networks

In recent years, research on the federated spiking neural network (SNN) framework has attracted increasing attention in the area of on-chip learning for embedded devices, because of its advantages of low power consumption and privacy security. Most of the existing federated SNN frameworks are based on the classical federated learning framework -- Federated Average (FedAvg) framework, where internal communication is achieved by exchanging network parameters or gradients. However, although these frameworks take a series of methods to reduce the communication cost, the communication of frameworks still increases with the scale of the backbone network. To solve the problem, we propose a new federated SNN framework, Federal SNN distillation (FSD), whose communication is independent of the scale of the network. Through the idea of knowledge distillation, FSD replaces the network parameters or gradients with the output spikes of SNN, which greatly reduces the communication while ensuring the effect. In addition, we propose a lossless compression algorithm to further compress the binary output spikes of SNN. The proposed framework FSD is compared with the existing FedAvg frameworks on MNIST, Fashion MNIST and CIFAR10 datasets. The experiment results demonstrate that FSD communication is decreased by 1-2 orders of magnitude when reaching the same accuracy.

An Incremental-ΔΣ ADC With 106-dB DR for Reconfigurable Class-D Audio Amplifiers

This paper presents a hybrid ADC, by combining an incremental converter with a delta-sigma modulator (ΣM) to achieve a high-dynamic response, intended for audio applications. The circuit uses a zero + first-order incremental converter with a 4-b quantizer and a third-order ΣM. Combining the two binary outputs generates a 576-kHz 6-bit binary flow for directly driving a class-D amplifier. A 4-bit flash used in the incremental running at 1/16 the clock frequency and a one-shift DEM technique compensate for the capacitive mismatch and cancel its effect for input signals lower than -35 dBFS. The proposed ADC, fabricated in a 0.18-m CMOS process, occupies an active area of 1.2 mm and consumes 7.2 mW with a 1.8-V supply. The circuit achieves an SNR/SNDR/DR of 96.8/94.45/106 dB for a 21-kHz signal bandwidth.

A new approach for writer verification based on segments of handwritten graphemes

Abstract Traditional literature considers complex biometric sources such as words, letters and signatures for writer verification/identification. In this work the use of small segments of the handwritten stroke for writer verification is proposed. A grapheme is defined as the concatenation of smaller segments or fragments. Two models of grapheme are developed based on the idea that the segments are parts of a circle with or without direction. The average of Gray Level of the Perpendicular Line to the Skeleton and Local Binary Pattern are adopted as descriptors. A database of 3,000 images of 50 writers, with 6 types of segments and with 10 samples per segment has been developed. A binary output support vector machine was applied as classifier. Thus, 50 classifiers were trained using 100 balanced data sets generated using subsampling of the majority class. Experiments are carried out with the proposed models, with an identity verification hit rate of 97 % on average. The positive results of experiments show that it is possible to achieve high precision in writer verification using segment of graphemes.

Programming cell-free biosensors with DNA strand displacement circuits

Cell-free biosensors are powerful platforms for monitoring human and environmental health. Here, we expand their capabilities by interfacing them with toehold-mediated strand displacement circuits, a dynamic DNA nanotechnology that enables molecular computation through programmable interactions between nucleic acid strands. We develop design rules for interfacing a small molecule sensing platform called ROSALIND with toehold-mediated strand displacement to construct hybrid RNA–DNA circuits that allow fine-tuning of reaction kinetics. We use these design rules to build 12 different circuits that implement a range of logic functions (NOT, OR, AND, IMPLY, NOR, NIMPLY, NAND). Finally, we demonstrate a circuit that acts like an analog-to-digital converter to create a series of binary outputs that encode the concentration range of the molecule being detected. We believe this work establishes a pathway to create ‘smart’ diagnostics that use molecular computations to enhance the speed and utility of biosensors.

A deep learning and heuristic methodology for predicting breakups in social network structures

AbstractLiterature have focused on studying the apparent and latent interactions within social graphs as an n‐ary operation, which yields binary outputs comprising positives (friends, likes, etc.) and negatives (foes, dislikes, etc.). Inasmuch as interactions constitute the bedrock of any given social network (SN) structure; there exist scenarios where an interaction, which was once considered a positive, transmutes into a negative as a result of one or more indicators which have affected the interaction quality. At present, this transmutation has to be manually executed by the affected actors in the SN. These manual transmutations can be quite inefficient, ineffective, and a mishap might have been incurred by the constituent actors and the SN structure prior to a resolution. Our problem statement aims at automatically flagging positive ties that should be considered for breakups or rifts (negative‐tie state), as they tend to pose potential threats to actors and the SN. Therefore, we have proposed ClasReg: a unique framework capable of breakup and link predictions.