Super-Resolution Reconstruction of Terahertz Images Based on Residual Generative Adversarial Network with Enhanced Attention.

Terahertz (THz) waves are widely used in the field of non-destructive testing (NDT). However, terahertz images have issues with limited spatial resolution and fuzzy features because of the constraints of the imaging equipment and imaging algorithms. To solve these problems, we propose a residual generative adversarial network based on enhanced attention (EA), which aims to pay more attention to the reconstruction of textures and details while not influencing the image outlines. Our method successfully recovers detailed texture information from low-resolution images, as demonstrated by experiments on the benchmark datasets Set5 and Set14. To use the network to improve the resolution of terahertz images, we create an image degradation algorithm and a database of terahertz degradation images. Finally, the real reconstruction of terahertz images confirms the effectiveness of our method.

Investigating the binding mechanism of topiramate with bovine serum albumin using spectroscopic and computational methods.

Various spectroscopic techniques involving fluorescence spectroscopy, circular dichroism (CD), and computational approaches were used to elucidate the molecular aspects of interaction between the antiepileptic drug topiramate and the multifunctional transport protein bovine serum albumin (BSA) under physiological conditions. Topiramate quenched BSA fluorescence in a static quenching mode, according to the Stern-Volmer quenching constant (Ksv ) data derived from fluorescence spectroscopy for the topiramate-BSA complex. The binding constant was also used to calculate the binding affinity for the topiramate-BSA interaction. Fluorescence and circular dichroism experiments demonstrate that the protein's tertiary structure is affected by the microenvironmental alterations generated by topiramate binding to BSA. To establish the exact binding site, interacting residues, and interaction forces involved in the binding of topiramate to BSA, molecular modeling and simulation approaches were used. According to the Molecular Mechanics Poisson-Boltzmann Surface Area (MMPBSA) calculations, the average binding energy between topiramate and BSA is -421.05 kJ/mol. Topiramate was discovered to have substantial interactions with BSA, changing the structural dynamic and Gibbs free energy landscape patterns.

Computational analysis of PTP-1B site-directed mutations and their structural binding to potential inhibitors.

Protein tyrosine phosphatase-1B (PTP-1B) is a well-known therapeutic target for diabetes and obesity as it suppresses insulin and leptin signaling. PTP-1B deletion or pharmacological suppression boosted glucose homeostasis and insulin signaling without altering hepatic fat storage. Inhibitors of PTP-1B may be useful in the treatment of type 2 diabetes, and shikonin, a naturally occurring naphthoquinone dye pigment, is reported to inhibit PTP-1B and possess antidiabetic properties. Since the cell contains a large number of phosphatases, PTP-1B inhibitors must be effective and selective. To explore more about the mechanism underlying the inhibitor's efficacy and selectivity, we investigated its top four pharmacophores and used site-directed mutagenesis to insert amino acid mutations into PTP-1B as an extension of our previous study where we identified 4 pharmacophores of shikonin. The study aimed to examine the site-directed mutations like R24Y, S215E, and S216C influence the binding of shikonin pharmacophores, which act as selective inhibitors of PTP-1B. To achieve this purpose, docking and molecular dynamics simulations of wild-type (WT) and mutant PTP-1B with antidiabetic compounds were undertaken. The simulation results revealed that site-directed mutations can change the hydrogen bond and hydrophobic interactions between shikonin pharmacophores and many residues in PTP-1B's active site, influencing the drug's binding affinity. These findings could aid researchers in better understanding PTP-1B inhibitors' selective binding mechanism and pave the path for the creation of effective PTP-1B inhibitors.

[Automatic detection model of hypertrophic cardiomyopathy based on deep convolutional neural network].

The diagnosis of hypertrophic cardiomyopathy (HCM) is of great significance for the early risk classification of sudden cardiac death and the screening of family genetic diseases. This research proposed a HCM automatic detection method based on convolution neural network (CNN) model, using single-lead electrocardiogram (ECG) signal as the research object. Firstly, the R-wave peak locations of single-lead ECG signal were determined, followed by the ECG signal segmentation and resample in units of heart beats, then a CNN model was built to automatically extract the deep features in the ECG signal and perform automatic classification and HCM detection. The experimental data is derived from 108 ECG records extracted from three public databases provided by PhysioNet, the database established in this research consists of 14,459 heartbeats, and each heartbeat contains 128 sampling points. The results revealed that the optimized CNN model could effectively detect HCM, the accuracy, sensitivity and specificity were 95.98%, 98.03% and 95.79% respectively. In this research, the deep learning method was introduced for the analysis of single-lead ECG of HCM patients, which could not only overcome the technical limitations of conventional detection methods based on multi-lead ECG, but also has important application value for assisting doctor in fast and convenient large-scale HCM preliminary screening.

[Virtual screening of active ingredients of traditional Chinese medicine in treating COVID-19 based on molecular docking and molecular dynamic simulation].

We aim to screen out the active components that may have therapeutic effect on coronavirus disease 2019 (COVID-19) from the severe and critical cases' prescriptions in the "Coronavirus Disease 2019 Diagnosis and Treatment Plan (Trial Ninth Edition)" issued by the National Health Commission of the People's Republic of China and explain its mechanism through the interactions with proteins. The ETCM database and SwissADME database were used to screen the active components contained in 25 traditional Chinese medicines in 3 prescriptions, and the PDB database was used to obtain the crystal structures of 4 proteins of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Molecular docking was performed using Autodock Vina and molecular dynamics simulations were performed using GROMACS. Binding energy results showed that 44 active ingredients including xambioona, gancaonin L, cynaroside, and baicalin showed good binding affinity with multiple targets of SARS-CoV-2, while molecular dynamics simulations analysis showed that xambioona bound more tightly to the nucleocapsid protein of SARS-CoV-2 and exerted a potent inhibitory effect. Modern technical methods are used to study the active components of traditional Chinese medicine and show that xambioona is an effective inhibitor of SARS-CoV-2 nucleocapsid protein, which provides a theoretical basis for the development of new anti-SARS-CoV-2 drugs and their treatment methods.

[Automatic epileptic seizure detection algorithm based on dual density dual tree complex wavelet transform].

It is very important for epilepsy treatment to distinguish epileptic seizure and non-seizure. In this study, an automatic seizure detection algorithm based on dual density dual tree complex wavelet transform (DD-DT CWT) for intracranial electroencephalogram (iEEG) was proposed. The experimental data were collected from 15 719 competition data set up by the National Institutes of Health (NINDS) in Kaggle. The processed database consisted of 55 023 seizure epochs and 501 990 non-seizure epochs. Each epoch was 1 second long and contained 174 sampling points. Firstly, the signal was resampled. Then, DD-DT CWT was used for EEG signal processing. Four kinds of features include wavelet entropy, variance, energy and mean value were extracted from the signal. Finally, these features were sent to least squares-support vector machine (LS-SVM) for learning and classification. The appropriate decomposition level was selected by comparing the experimental results under different wavelet decomposition levels. The experimental results showed that the features selected in this paper were different between seizure and non-seizure. Among the eight patients, the average accuracy of three-level decomposition classification was 91.98%, the sensitivity was 90.15%, and the specificity was 93.81%. The work of this paper shows that our algorithm has excellent performance in the two classification of EEG signals of epileptic patients, and can detect the seizure period automatically and efficiently.

Identifying novel sphingosine kinase 1 inhibitors as therapeutics against breast cancer.

Sphingosine kinase 1 (SphK1) is a promising therapeutic target against several diseases including mammary cancer. The aim of present work is to identify a potent lead compound against breast cancer using ligand-based virtual screening, molecular docking, MD simulations, and the MMPBSA calculations. The LBVS in molecular and virtual libraries yielded 20,800 hits, which were reduced to 621 by several parameters of drug-likeness, lead-likeness, and PAINS. Furthermore, 55 compounds were selected by ADMET descriptors carried forward for molecular interaction studies with SphK1. The binding energy (ΔG) of three screened compounds namely ZINC06823429 (-11.36 kcal/mol), ZINC95421501 (-11.29 kcal/mol), and ZINC95421070 (-11.26 kcal/mol) exhibited stronger than standard drug PF-543 (-9.9 kcal/mol). Finally, it was observed that the ZINC06823429 binds tightly to catalytic site of SphK1 and remain stable during MD simulations. This study provides a significant understanding of SphK1 inhibitors that can be used in the development of potential therapeutics against breast cancer.

A Semi-supervised Algorithm for Atrial Fibrillation Attack Prediction Using Convolution Auto-encoder of Time Series Signal.

During the initial stages, atrial fibrillation (AF) typically presents as paroxysmal atrial fibrillation (PAF), which may further progress into persistent atrial fibrillation, leading to high-risk diseases such as ischemic stroke and heart failure. Given that the current machine learning algorithms used for predicting AF involve time-consuming and labor-intensive processes of feature extraction and labeling electrocardiogram data, this study proposes a novel two-stage semi-supervised AF attack prediction algorithm. The first stage is designed as unsupervised learning based on convolutional autoencoder (CAE) network when inputting RR interval time series signal, while the second stage is designed as supervised learning using a Long Short-Term Memory (LSTM) model. A training set consisting of 20 segments of PAF and 20 normal heart rates was used to evaluate the performance of the CAE-LSTM combination model. The results showed that the average accuracy and root mean square error of ten-fold cross-validation were 93.56% and 0.004, respectively, with an F1 parameter of 0.9345. In summary, the preliminary results suggest that the combination of unsupervised CAE model and supervised LSTM model can reduce the dimensionality of the input data while using a small amount of labeled data as input for subsequent classification. Furthermore, the proposed algorithm can be used for predicting atrial fibrillation when the sample size is limited.Clinical Relevance- Compared with common supervised methods, our proposed method only requires a small number of tagged ECG signals, which can reduce the workload of clinicians to complete the task of atrial fibrillation attack prediction.

The Molecular Basis of the Effect of Temperature on the Structure and Function of SARS-CoV-2 Spike Protein.

The remarkable rise of the current COVID-19 pandemic to every part of the globe has raised key concerns for the current public healthcare system. The spike (S) protein of SARS-CoV-2 shows an important part in the cell membrane fusion and receptor recognition. It is a key target for vaccine production. Several researchers studied the nature of this protein under various environmental conditions. In this work, we applied molecular modeling and extensive molecular dynamics simulation approaches at 0°C (273.15 K), 20°C (293.15 K), 40°C (313.15 K), and 60°C (333.15 K) to study the detailed conformational alterations in the SARS-CoV-2 S protein. Our aim is to understand the influence of temperatures on the structure, function, and dynamics of the S protein of SARS-CoV-2. The structural deviations, and atomic and residual fluctuations were least at low (0°C) and high (60°C) temperature. Even the internal residues of the SARS-CoV-2 S protein are not accessible to solvent at high temperature. Furthermore, there was no unfolding of SARS-CoV-2 spike S reported at higher temperature. The most stable conformations of the SARS-CoV-2 S protein were reported at 20°C, but the free energy minimum region of the SARS-CoV-2 S protein was sharper at 40°C than other temperatures. Our findings revealed that higher temperatures have little or no influence on the stability and folding of the SARS-CoV-2 S protein.

Addition of Mercury Causes Quenching of NIR Fluorescence Emission Spectra of a Photoactivatable PAiRFP1 Protein.

BACKGROUND: Biliverdin (BV) containing far-red light photoactivatable near-infrared fluorescent protein (NIR-FP) named PAiRFP1 has been developed by directed molecular evolution from one bathy bacteriophytochrome of Agrobacterium tumefaciens C58 called Agp2 or AtBphP2. Usually, the fluorescence intensity of the NIR emission spectra of PAiRFP1 tends to increase upon repeated excitation by far-red light. OBJECTIVE: This study aimed at exploring the role of PAiRFP1 and its mutants, such as V386A, V480A, and Y498H, as NIR biosensors for the detection of Hg2+ ions in the buffer solutions. METHODS: In this study, we used PCR-based site-directed reverse mutagenesis, fluorescence spectroscopy, and molecular modeling approaches on PAiRFP1 and its mutants. RESULTS: It was found that PAiRFP1 and its mutants experienced strong quenching of NIR fluorescence emission spectra upon the addition of different concentrations (0-3µM) of mercuric chloride (HgCl2). CONCLUSION: We hypothesized that PAiRFP1 and its variants have some potential to be used as NIR biosensors for the in vitro detection of Hg2+ ions in biological media. Moreover, we also hypothesized that PAiRFP1 would be the best tool to use as a NIR biosensor to detect Hg2+ ions in living organisms because of its higher signal-to-noise (SNR) ratio than other infra-red fluorescent proteins.

In silico prediction of natural compounds as potential multi-target inhibitors of structural proteins of SARS-CoV-2.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a colossal loss to human health and lives and has deeply impacted socio-economic growth. Remarkable efforts have been made by the scientific community in containing the virus by successful development of vaccines and diagnostic kits. Initiatives towards drug repurposing and discovery have also been undertaken. In this study, we compiled the known natural anti-viral compounds using text mining of the literature and examined them against four major structural proteins of SARS-CoV-2, namely, spike (S) protein, nucleocapsid (N) protein, membrane (M) protein and envelope (E) protein. Following computational approaches, we identified fangchinoline and versicolactone C as the compounds to exhibit strong binding to the target proteins and causing structural deformation of three structural proteins (N, S and M). We recommend the inhibitory effects of these compounds from our study should be experimentally validated against SARS-CoV-2.Communicated by Ramaswamy H. Sarma.

Effect of polyol osmolytes on the structure-function integrity and aggregation propensity of catalase: A comprehensive study based on spectroscopic and molecular dynamic simulation measurements.

Owing to the ability of catalase to function under oxidative stress vis-à-vis its industrial importance, the structure-function integrity of the enzyme is of prime concern. In the present study, polyols (glycerol, sorbitol, sucrose, xylitol), were evaluated for their ability to modulate structure, activity and aggregation of catalase using in vitro and in silico approaches. All polyols were found to increase catalase activity by decreasing Km and increasing Vmax resulting in enhanced catalytic efficiency (kcat/Km) of the enzyme. Glycerol was found to be the most efficient polyol with a kcat/Km increase from 4.38 × 104 mM-1 S-1 (control) to 5.8 × 105 mM-1 S-1. Correlatively with this, enhanced secondary structure with reduced hydrophobic exposure was observed in all polyols. Furthermore, increased stability, with an increase in melting temperature by 15.2 °C, and almost no aggregation was observed in glycerol. Overall, ability to regulate structure-function integrity and aggregation propensity was highest for glycerol and lowest for xylitol. Simulation studies were performed involving structural dynamics measurement, principal component analysis and free energy landscape analysis. Altogether, all polyols were stabilizing in nature and glycerol, in particular, has potential to efficiently prevent not only the aggregation of the antioxidant defense system but might also serve as a stability aid during industrial processing of catalase.

Recent Advances in Stretchable and Wearable Capacitive Electrophysiological Sensors for Long-Term Health Monitoring.

Over the past several years, wearable electrophysiological sensors with stretchability have received significant research attention because of their capability to continuously monitor electrophysiological signals from the human body with minimal body motion artifacts, long-term tracking, and comfort for real-time health monitoring. Among the four different sensors, i.e., piezoresistive, piezoelectric, iontronic, and capacitive, capacitive sensors are the most advantageous owing to their reusability, high durability, device sterilization ability, and minimum leakage currents between the electrode and the body to reduce the health risk arising from any short circuit. This review focuses on the development of wearable, flexible capacitive sensors for monitoring electrophysiological conditions, including the electrode materials and configuration, the sensing mechanisms, and the fabrication strategies. In addition, several design strategies of flexible/stretchable electrodes, body-to-electrode signal transduction, and measurements have been critically evaluated. We have also highlighted the gaps and opportunities needed for enhancing the suitability and practical applicability of wearable capacitive sensors. Finally, the potential applications, research challenges, and future research directions on stretchable and wearable capacitive sensors are outlined in this review.

Wearable Flexible Electronics Based Cardiac Electrode for Researcher Mental Stress Detection System Using Machine Learning Models on Single Lead Electrocardiogram Signal.

In the modern world, wearable smart devices are continuously used to monitor people's health. This study aims to develop an automatic mental stress detection system for researchers based on Electrocardiogram (ECG) signals from smart T-shirts using machine learning classifiers. We used 20 subjects, including 10 from mental stress (after twelve hours of continuous work in the laboratory) and 10 from normal (after completing the sleep or without any work). We also applied three scoring techniques: Chalder Fatigue Scale (CFS), Specific Fatigue Scale (SFS), Depression, Anxiety, and Stress Scale (DASS), to confirm the mental stress. The total duration of ECG recording was 1800 min, including 1200 min during mental stress and 600 min during normal. We calculated two types of features, such as demographic and extracted by ECG signal. In addition, we used Decision Tree (DT), Naive Bayes (NB), Random Forest (RF), and Logistic Regression (LR) to classify the intra-subject (mental stress and normal) and inter-subject classification. The DT leave-one-out model has better performance in terms of recall (93.30%), specificity (96.70%), precision (94.40%), accuracy (93.30%), and F1 (93.50%) in the intra-subject classification. Additionally, The classification accuracy of the system in classifying inter-subjects is 94.10% when using a DT classifier. However, our findings suggest that the wearable smart T-shirt based on the DT classifier may be used in big data applications and health monitoring. Mental stress can lead to mitochondrial dysfunction, oxidative stress, blood pressure, cardiovascular disease, and various health problems. Therefore, real-time ECG signals help assess cardiovascular and related risk factors in the initial stage based on machine learning techniques.

Role of Oxidative Stress and Inflammation in Insomnia Sleep Disorder and Cardiovascular Diseases: Herbal Antioxidants and Anti-inflammatory Coupled with Insomnia Detection using Machine Learning.

Insomnia is well-known as trouble in sleeping and enormously influences human life due to the shortage of sleep. Reactive Oxygen Species (ROS) accrue in neurons during the waking state, and sleep has a defensive role against oxidative damage and dissipates ROS in the brain. In contrast, insomnia is the source of inequity between ROS generation and removal by an endogenous antioxidant defense system. The relationship between insomnia, depression, and anxiety disorders damages the cardiovascular systems' immune mechanisms and functions. Traditionally, polysomnography is used in the diagnosis of insomnia. This technique is complex, with a long time overhead. In this work, we have proposed a novel machine learning-based automatic detection system using the R-R intervals extracted from a single-lead electrocardiograph (ECG). Additionally, we aimed to explore the role of oxidative stress and inflammation in sleeping disorders and cardiovascular diseases, antioxidants' effects, and the psychopharmacological effect of herbal medicine. This work has been carried out in steps, which include collecting the ECG signal for normal and insomnia subjects, analyzing the signal, and finally, automatic classification. We used two approaches, including subjects (normal and insomnia), two sleep stages, i.e., wake and rapid eye movement, and three Machine Learning (ML)-based classifiers to complete the classification. A total number of 3000 ECG segments were collected from 18 subjects. Furthermore, using the theranostics approach, the role of mitochondrial dysfunction causing oxidative stress and inflammatory response in insomnia and cardiovascular diseases was explored. The data from various databases on the mechanism of action of different herbal medicines in insomnia and cardiovascular diseases with antioxidant and antidepressant activities were also retrieved. Random Forest (RF) classifier has shown the highest accuracy (subjects: 87.10% and sleep stage: 88.30%) compared to the Decision Tree (DT) and Support Vector Machine (SVM). The results revealed that the suggested method could perform well in classifying the subjects and sleep stages. Additionally, a random forest machine learning-based classifier could be helpful in the clinical discovery of sleep complications, including insomnia. The evidence retrieved from the databases showed that herbal medicine contains numerous phytochemical bioactives and has multimodal cellular mechanisms of action, viz., antioxidant, anti-inflammatory, vasorelaxant, detoxifier, antidepressant, anxiolytic, and cell-rejuvenator properties. Other herbal medicines have a GABA-A receptor agonist effect. Hence, we recommend that the theranostics approach has potential and can be adopted for future research to improve the quality of life of humans.

An End-to-End Cardiac Arrhythmia Recognition Method with an Effective DenseNet Model on Imbalanced Datasets Using ECG Signal.

Electrocardiography (ECG) is a well-known noninvasive technique in medical science that provides information about the heart's rhythm and current conditions. Automatic ECG arrhythmia diagnosis relieves doctors' workload and improves diagnosis effectiveness and efficiency. This study proposes an automatic end-to-end 2D CNN (two-dimensional convolution neural networks) deep learning method with an effective DenseNet model for addressing arrhythmias recognition. To begin, the proposed model is trained and evaluated on the 97720 and 141404 beat images extracted from the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia and St. Petersburg Institute of Cardiological Technics (INCART) datasets (both are imbalanced class datasets) using a stratified 5-fold evaluation strategy. The data is classified into four groups: N (normal), V (ventricular ectopic), S (supraventricular ectopic), and F (fusion), based on the Association for the Advancement of Medical Instrumentation® (AAMI). The experimental results show that the proposed model outperforms state-of-the-art models for recognizing arrhythmias, with the accuracy of 99.80% and 99.63%, precision of 98.34% and 98.94%, and F 1-score of 98.91% and 98.91% on the MIT-BIH arrhythmia and INCART datasets, respectively. Using a transfer learning mechanism, the proposed model is also evaluated with only five individuals of supraventricular MIT-BIH arrhythmia and five individuals of European ST-T datasets (both of which are also class imbalanced) and achieved satisfactory results. So, the proposed model is more generalized and could be a prosperous solution for arrhythmias recognition from class imbalance datasets in real-life applications.

Withania somnifera (L.) Dunal (Ashwagandha) for the possible therapeutics and clinical management of SARS-CoV-2 infection: Plant-based drug discovery and targeted therapy.

Coronavirus disease 2019 (COVID-19) pandemic has killed huge populations throughout the world and acts as a high-risk factor for elderly and young immune-suppressed patients. There is a critical need to build up secure, reliable, and efficient drugs against to the infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus. Bioactive compounds of Ashwagandha [Withania somnifera (L.) Dunal] may implicate as herbal medicine for the management and treatment of patients infected by SARS-CoV-2 infection. The aim of the current work is to update the knowledge of SARS-CoV-2 infection and information about the implication of various compounds of medicinal plant Withania somnifera with minimum side effects on the patients' organs. The herbal medicine Withania somnifera has an excellent antiviral activity that could be implicated in the management and treatment of flu and flu-like diseases connected with SARS-CoV-2. The analysis was performed by systematically re-evaluating the published articles related to the infection of SARS-CoV-2 and the herbal medicine Withania somnifera. In the current review, we have provided the important information and data of various bioactive compounds of Withania somnifera such as Withanoside V, Withanone, Somniferine, and some other compounds, which can possibly help in the management and treatment of SARS-CoV-2 infection. Withania somnifera has proved its potential for maintaining immune homeostasis of the body, inflammation regulation, pro-inflammatory cytokines suppression, protection of multiple organs, anti-viral, anti-stress, and anti-hypertensive properties. Withanoside V has the potential to inhibit the main proteases (Mpro) of SARS-CoV-2. At present, synthetic adjuvant vaccines are used against COVID-19. Available information showed the antiviral activity in Withanoside V of Withania somnifera, which may explore as herbal medicine against to SARS-CoV-2 infection after standardization of parameters of drug development and formulation in near future.

An Automatic Premature Ventricular Contraction Recognition System Based on Imbalanced Dataset and Pre-Trained Residual Network Using Transfer Learning on ECG Signal.

The development of automatic monitoring and diagnosis systems for cardiac patients over the internet has been facilitated by recent advancements in wearable sensor devices from electrocardiographs (ECGs), which need the use of patient-specific approaches. Premature ventricular contraction (PVC) is a common chronic cardiovascular disease that can cause conditions that are potentially fatal. Therefore, for the diagnosis of likely heart failure, precise PVC detection from ECGs is crucial. In the clinical settings, cardiologists typically employ long-term ECGs as a tool to identify PVCs, where a cardiologist must put in a lot of time and effort to appropriately assess the long-term ECGs which is time consuming and cumbersome. By addressing these issues, we have investigated a deep learning method with a pre-trained deep residual network, ResNet-18, to identify PVCs automatically using transfer learning mechanism. Herein, features are extracted by the inner layers of the network automatically compared to hand-crafted feature extraction methods. Transfer learning mechanism handles the difficulties of required large volume of training data for a deep model. The pre-trained model is evaluated on the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) Arrhythmia and Institute of Cardiological Technics (INCART) datasets. First, we used the Pan-Tompkins algorithm to segment 44,103 normal and 6423 PVC beats, as well as 106,239 normal and 9987 PVC beats from the MIT-BIH Arrhythmia and IN-CART datasets, respectively. The pre-trained model employed the segmented beats as input after being converted into 2D (two-dimensional) images. The method is optimized with the using of weighted random samples, on-the-fly augmentation, Adam optimizer, and call back feature. The results from the proposed method demonstrate the satisfactory findings without the using of any complex pre-processing and feature extraction technique as well as design complexity of model. Using LOSOCV (leave one subject out cross-validation), the received accuracies on MIT-BIH and INCART are 99.93% and 99.77%, respectively, suppressing the state-of-the-art methods for PVC recognition on unseen data. This demonstrates the efficacy and generalizability of the proposed method on the imbalanced datasets. Due to the absence of device-specific (patient-specific) information at the evaluating stage on the target datasets in this study, the method might be used as a general approach to handle the situations in which ECG signals are obtained from different patients utilizing a variety of smart sensor devices.

A Novel Smart Belt for Anxiety Detection, Classification, and Reduction Using IIoMT on Students' Cardiac Signal and MSY.

The prevalence of anxiety among university students is increasing, resulting in the negative impact on their academic and social (behavioral and emotional) development. In order for students to have competitive academic performance, the cognitive function should be strengthened by detecting and handling anxiety. Over a period of 6 weeks, this study examined how to detect anxiety and how Mano Shakti Yoga (MSY) helps reduce anxiety. Relying on cardiac signals, this study follows an integrated detection-estimation-reduction framework for anxiety using the Intelligent Internet of Medical Things (IIoMT) and MSY. IIoMT is the integration of Internet of Medical Things (wearable smart belt) and machine learning algorithms (Decision Tree (DT), Random Forest (RF), and AdaBoost (AB)). Sixty-six eligible students were selected as experiencing anxiety detected based on the results of self-rating anxiety scale (SAS) questionnaire and a smart belt. Then, the students were divided randomly into two groups: experimental and control. The experimental group followed an MSY intervention for one hour twice a week, while the control group followed their own daily routine. Machine learning algorithms are used to analyze the data obtained from the smart belt. MSY is an alternative improvement for the immune system that helps reduce anxiety. All the results illustrate that the experimental group reduced anxiety with a significant (p < 0.05) difference in group × time interaction compared to the control group. The intelligent techniques achieved maximum accuracy of 80% on using RF algorithm. Thus, students can practice MSY and concentrate on their objectives by improving their intelligence, attention, and memory.

An Effective and Lightweight Deep Electrocardiography Arrhythmia Recognition Model Using Novel Special and Native Structural Regularization Techniques on Cardiac Signal.

Recently, cardiac arrhythmia recognition from electrocardiography (ECG) with deep learning approaches is becoming popular in clinical diagnosis systems due to its good prognosis findings, where expert data preprocessing and feature engineering are not usually required. But a lightweight and effective deep model is highly demanded to face the challenges of deploying the model in real-life applications and diagnosis accurately. In this work, two effective and lightweight deep learning models named Deep-SR and Deep-NSR are proposed to recognize ECG beats, which are based on two-dimensional convolution neural networks (2D CNNs) while using different structural regularizations. First, 97720 ECG beats extracted from all records of a benchmark MIT-BIH arrhythmia dataset have been transformed into 2D RGB (red, green, and blue) images that act as the inputs to the proposed 2D CNN models. Then, the optimization of the proposed models is performed through the proper initialization of model layers, on-the-fly augmentation, regularization techniques, Adam optimizer, and weighted random sampler. Finally, the performance of the proposed models is evaluated by a stratified 5-fold cross-validation strategy along with callback features. The obtained overall accuracy of recognizing normal beat and three arrhythmias (V-ventricular ectopic, S-supraventricular ectopic, and F-fusion) based on the Association for the Advancement of Medical Instrumentation (AAMI) is 99.93%, and 99.96% for the proposed Deep-SR model and Deep-NSR model, which demonstrate that the effectiveness of the proposed models has surpassed the state-of-the-art models and also expresses the higher model generalization. The received results with model size suggest that the proposed CNN models especially Deep-NSR could be more useful in wearable devices such as medical vests, bracelets for long-term monitoring of cardiac conditions, and in telemedicine to accurate diagnose the arrhythmia from ECG automatically. As a result, medical costs of patients and work pressure on physicians in medicals and clinics would be reduced effectively.