Statistical and cross-sectional analysis of factors effecting the prevalence of pressure ulcer in hospital admitted stroke patients.

Every year, millions of people around the world are disabled by stroke, it is well recognized that complications aftera stroke extend hospital stays and pressure ulcers, a stroke consequence, which can be prevented by educating the caregiver. The primary focus of this research is not only to investigate the prevalence of pressure ulcers (PU) among stroke patients, but this study also introduced a variety of factors which influence the formation of PU, such as restricted mobility, gender, duration of stroke, hypertension, diabetes, hygiene, type of mattress, malnutrition, awareness, etc. In addition, this research provides a comparative and statistical analysis, a cause of the catastrophic disabilities influenced by a variety of factors. Moreover, the proposed research also provides a room for the pertinent treatment of stroke patient to curtail the formation of pressure ulcer. In this research, a total of 120 stroke patients were initially included to monitor the frequency of pressure ulcers at incipient stage. Out of the total patients, the number of patients with ischemic stroke were 78.5 % while 8.3 % were of haemorrhagic type. In the results, the demographic characteristics and the factors which influence the formation of PU of the patients were examined with their cross-sectional impact on each other through comparative and statistical analysis. It was discovered that among all the stroke patients, 8.3 % were found with a PUs and the most frequent localization was sacrum and no new PU was observed for the participants under the observation.

Transcription factor combinations that define human astrocyte identity encode significant variation of maturity and function.

Increasing evidence indicates that cellular identity can be reduced to the distinct gene regulatory networks controlled by transcription factors (TFs). However, redundancy exists in these states as different combinations of TFs can induce broadly similar cell types. We previously demonstrated that by overcoming gene silencing, it is possible to deterministically reprogram human pluripotent stem cells directly into cell types of various lineages. In the present study we leverage the consistency and precision of our approach to explore four different TF combinations encoding astrocyte identity, based on previously published reports. Analysis of the resulting induced astrocytes (iAs) demonstrated that all four cassettes generate cells with the typical morphology of in vitro astrocytes, which expressed astrocyte-specific markers. The transcriptional profiles of all four iAs clustered tightly together and displayed similarities with mature human astrocytes, although maturity levels differed between cells. Importantly, we found that the TF cassettes induced iAs with distinct differences with regards to their cytokine response and calcium signaling. In vivo transplantation of selected iAs into immunocompromised rat brains demonstrated long term stability and integration. In conclusion, all four TF combinations were able to induce stable astrocyte-like cells that were morphologically similar but showed subtle differences with respect to their transcriptome. These subtle differences translated into distinct differences with regards to cell function, that could be related to maturation state and/or regional identity of the resulting cells. This insight opens an opportunity to precision-engineer cells to meet functional requirements, for example, in the context of therapeutic cell transplantation.

A Comprehensive Review on Machine Learning in Healthcare Industry: Classification, Restrictions, Opportunities and Challenges.

Recently, various sophisticated methods, including machine learning and artificial intelligence, have been employed to examine health-related data. Medical professionals are acquiring enhanced diagnostic and treatment abilities by utilizing machine learning applications in the healthcare domain. Medical data have been used by many researchers to detect diseases and identify patterns. In the current literature, there are very few studies that address machine learning algorithms to improve healthcare data accuracy and efficiency. We examined the effectiveness of machine learning algorithms in improving time series healthcare metrics for heart rate data transmission (accuracy and efficiency). In this paper, we reviewed several machine learning algorithms in healthcare applications. After a comprehensive overview and investigation of supervised and unsupervised machine learning algorithms, we also demonstrated time series tasks based on past values (along with reviewing their feasibility for both small and large datasets).

Nano-Enabled Antivirals for Overcoming Antibody Escaped Mutations Based SARS-CoV-2 Waves.

This review discusses receptor-binding domain (RBD) mutations related to the emergence of various SARS-CoV-2 variants, which have been highlighted as a major cause of repetitive clinical waves of COVID-19. Our perusal of the literature reveals that most variants were able to escape neutralizing antibodies developed after immunization or natural exposure, pointing to the need for a sustainable technological solution to overcome this crisis. This review, therefore, focuses on nanotechnology and the development of antiviral nanomaterials with physical antagonistic features of viral replication checkpoints as such a solution. Our detailed discussion of SARS-CoV-2 replication and pathogenesis highlights four distinct checkpoints, the S protein (ACE2 receptor coupling), the RBD motif (ACE2 receptor coupling), ACE2 coupling, and the S protein cleavage site, as targets for the development of nano-enabled solutions that, for example, prevent viral attachment and fusion with the host cell by either blocking viral RBD/spike proteins or cellular ACE2 receptors. As proof of this concept, we highlight applications of several nanomaterials, such as metal and metal oxide nanoparticles, carbon-based nanoparticles, carbon nanotubes, fullerene, carbon dots, quantum dots, polymeric nanoparticles, lipid-based, polymer-based, lipid-polymer hybrid-based, surface-modified nanoparticles that have already been employed to control viral infections. These nanoparticles were developed to inhibit receptor-mediated host-virus attachments and cell fusion, the uncoating of the virus, viral gene expression, protein synthesis, the assembly of progeny viral particles, and the release of the virion. Moreover, nanomaterials have been used as antiviral drug carriers and vaccines, and nano-enabled sensors have already been shown to enable fast, sensitive, and label-free real-time diagnosis of viral infections. Nano-biosensors could, therefore, also be useful in the remote testing and tracking of patients, while nanocarriers probed with target tissue could facilitate the targeted delivery of antiviral drugs to infected cells, tissues, organs, or systems while avoiding unwanted exposure of non-target tissues. Antiviral nanoparticles can also be applied to sanitizers, clothing, facemasks, and other personal protective equipment to minimize horizontal spread. We believe that the nanotechnology-enabled solutions described in this review will enable us to control repeated SAR-CoV-2 waves caused by antibody escape mutations.

Detection of blaTEM, blaCTX-M, blaCMY, and blaSHV Genes Among Extended-Spectrum Beta-Lactamase-Producing Escherichia coli Isolated from Migratory Birds Travelling to Bangladesh.

Extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli cause severe health hazards. Migratory birds are reservoirs and transmitters of many pathogens including ESBL-producing E. coli. To examine migratory birds as potential carriers of ESBL-producing E. coli and E. coli-carrying antibiotic resistance genes, 55 PCR-positive E. coli isolates were screened using the disk diffusion method, double-disk synergy test, and further polymerase chain reaction (PCR) tests. Genes encoding resistance to tetracycline [tetA, 100% (35/35); tetB, 31.43% (11/35)], fluoroquinolone [qnrA, 35.71% (10/28); qnrB, 25% (7/28)], and streptomycin [aadA1, 90.24% (37/41)] were detected in the isolated E. coli. Of the 55 E. coli isolates, 21 (38.18%) were ESBL producers, and all of them were multidrug resistant. All the ESBL-producing E. coli isolates harbored at least two or more beta-lactamase genes, of which blaTEM, blaCMY, blaCTX-M, and blaSHV were detected in 95.24%, 90.48%, 85.71%, and 42.86% of isolates, respectively. All the beta-lactamase genes were present in four of the ESBL-producing E. coli isolates. Furthermore, 95.24% of ESBL-producing E. coli isolates were positive for one or more antibiotic resistance genes. To the best of our knowledge, this is the first study to detect E. coli-carrying antibiotic resistance genes including beta-lactamase blaCMY and blaSHV originating from migratory birds in Bangladesh. These results suggest that migratory birds are potential carriers of ESBL-producing E. coli along with other clinically important antibiotic resistance genes which may have detrimental impacts on human health.

Unusual Driver Behavior Detection in Videos Using Deep Learning Models.

Anomalous driving behavior detection is becoming more popular since it is vital in ensuring the safety of drivers and passengers in vehicles. Road accidents happen for various reasons, including health, mental stress, and fatigue. It is critical to monitor abnormal driving behaviors in real time to improve driving safety, raise driver awareness of their driving patterns, and minimize future road accidents. Many symptoms appear to show this condition in the driver, such as facial expressions or abnormal actions. The abnormal activity was among the most common causes of road accidents, accounting for nearly 20% of all accidents, according to international data on accident causes. To avoid serious consequences, abnormal driving behaviors must be identified and avoided. As it is difficult to monitor anyone continuously, automated detection of this condition is more effective and quicker. To increase drivers' recognition of their driving behaviors and prevent potential accidents, a precise monitoring approach that detects abnormal driving behaviors and identifies abnormal driving behaviors is required. The most common activities performed by the driver while driving is drinking, eating, smoking, and calling. These types of driver activities are considered in this work, along with normal driving. This study proposed deep learning-based detection models for recognizing abnormal driver actions. This system is trained and tested using a newly created dataset, including five classes. The main classes include Driver-smoking, Driver-eating, Driver-drinking, Driver-calling, and Driver-normal. For the analysis of results, pre-trained and fine-tuned CNN models are considered. The proposed CNN-based model and pre-trained models ResNet101, VGG-16, VGG-19, and Inception-v3 are used. The results are compared by using the performance measures. The results are obtained 89%, 93%, 93%, 94% for pre-trained models and 95% by using the proposed CNN-based model. Our analysis and results revealed that our proposed CNN base model performed well and could effectively classify the driver's abnormal behavior.

Deep Learning Models for Classification of Dental Diseases Using Orthopantomography X-ray OPG Images.

The teeth are the most challenging material to work with in the human body. Existing methods for detecting teeth problems are characterised by low efficiency, the complexity of the experiential operation, and a higher level of user intervention. Older oral disease detection approaches were manual, time-consuming, and required a dentist to examine and evaluate the disease. To address these concerns, we propose a novel approach for detecting and classifying the four most common teeth problems: cavities, root canals, dental crowns, and broken-down root canals, based on the deep learning model. In this study, we apply the YOLOv3 deep learning model to develop an automated tool capable of diagnosing and classifying dental abnormalities, such as dental panoramic X-ray images (OPG). Due to the lack of dental disease datasets, we created the Dental X-rays dataset to detect and classify these diseases. The size of datasets used after augmentation was 1200 images. The dataset comprises dental panoramic images with dental disorders such as cavities, root canals, BDR, dental crowns, and so on. The dataset was divided into 70% training and 30% testing images. The trained model YOLOv3 was evaluated on test images after training. The experiments demonstrated that the proposed model achieved 99.33% accuracy and performed better than the existing state-of-the-art models in terms of accuracy and universality if we used our datasets on other models.

Prevalence and prediction of pressure ulcers in admitted stroke patients in a tertiary care hospital.

Pressure ulcer (PU) is a localized injury to the skin or underlying tissues usually over a bony prominence, which results due to pressure or pressure in combination with shear. It is an expensive health care problem that have deterring impact on the length of hospitalization and cause extra nursing care time. Moreover, PUs negatively impacts patients' health related quality of life. High PUs prevalence figures were found in specialized hospital units such as intensive care unit (ICU), orthopedics, surgery, and also in stroke patients in medical units. The major purpose of this study is to assess the frequency of pressure ulcers in stroke patients at Ayub teaching hospital. The methodology used for carrying out the research was cross-sectional study conducted during months of September, October, and November 2020. Questionnaire was used to collect the data and well-informed written consent was taken from the patients. A total of 120 stroke patients were initially included with the intention to study the frequency of PUs among them. Different age groups were taken but majority (48.3%) belonged to the age group 31-60 years. Maximum patients were hypertensive (65%), while few of them were diabetic (35%). From the results of proposed work, it is found that out of 120 stroke patients, 75.8% presented with ischemic stroke while 24.2% presented with hemorrhagic stroke. 8.3% that is 10 out of 120 stroke patients developed pressure ulcers of grade 1 (1.7%), grade 2 (1.7%), grade 3 (2.5%), and grade 4 (2.5%) mostly in the sacral region (6.7%) and also on ankle (0.8%), and shoulder (0.8%) respectively. Patients in the study group had unsatisfactory hygiene (6.7%) were malnourished (11.7%) and were not using preventive mattresses (79.2%). Those at the risk of developing pressure ulcers were not being repositioned (6.7%) and did not had awareness (10%). Prevention and treatment used in ward is 100%. Conclusively, the frequency of pressure ulcers in stroke patients was determined to be 8.3% and the most frequent localization was sacrum. The PU care in this hospital is appropriate but still could be improved further by improving risk assessment, prevention specially use of air mattress and patient education regarding PUs. The main objective of the study is to identify the frequency of PUs in stroke patients and to highlight various factors that would avoid PUs development.

A Novel Feature Extraction and Fault Detection Technique for the Intelligent Fault Identification of Water Pump Bearings.

The reliable and cost-effective condition monitoring of the bearings installed in water pumps is a real challenge in the industry. This paper presents a novel strong feature selection and extraction algorithm (SFSEA) to extract fault-related features from the instantaneous power spectrum (IPS). The three features extracted from the IPS using the SFSEA are fed to an extreme gradient boosting (XBG) classifier to reliably detect and classify the minor bearing faults. The experiments performed on a lab-scale test setup demonstrated classification accuracy up to 100%, which is better than the previously reported fault classification accuracies and indicates the effectiveness of the proposed method.

Natural Disasters Intensity Analysis and Classification Based on Multispectral Images Using Multi-Layered Deep Convolutional Neural Network.

Natural disasters not only disturb the human ecological system but also destroy the properties and critical infrastructures of human societies and even lead to permanent change in the ecosystem. Disaster can be caused by naturally occurring events such as earthquakes, cyclones, floods, and wildfires. Many deep learning techniques have been applied by various researchers to detect and classify natural disasters to overcome losses in ecosystems, but detection of natural disasters still faces issues due to the complex and imbalanced structures of images. To tackle this problem, we propose a multilayered deep convolutional neural network. The proposed model works in two blocks: Block-I convolutional neural network (B-I CNN), for detection and occurrence of disasters, and Block-II convolutional neural network (B-II CNN), for classification of natural disaster intensity types with different filters and parameters. The model is tested on 4428 natural images and performance is calculated and expressed as different statistical values: sensitivity (SE), 97.54%; specificity (SP), 98.22%; accuracy rate (AR), 99.92%; precision (PRE), 97.79%; and F1-score (F1), 97.97%. The overall accuracy for the whole model is 99.92%, which is competitive and comparable with state-of-the-art algorithms.

Low-density granulocytes activate T cells and demonstrate a non-suppressive role in systemic lupus erythematosus.

OBJECTIVES: The presence of proinflammatory low-density granulocytes (LDG) has been demonstrated in autoimmune and infectious diseases. Recently, regulatory neutrophilic polymorphonuclear myeloid-derived suppressor cells (PMN-MDSC) were identified in systemic lupus erythematosus (SLE). Because LDG and PMN-MDSC share a similar phenotype with contrasting functional effects, we explored these cells in a cohort of patients with SLE. METHODS: LDG and normal-density granulocytes (NDG) were isolated from fresh blood of healthy donors (HD) and patients with SLE. Associations between LDG and clinical manifestations were analysed. Multicolor flow cytometry and confocal imaging were performed to immunophenotype the cells. The ability of LDG and NDG to suppress T cell function and induce cytokine production was quantified. RESULTS: LDG prevalence was elevated in SLE versus HD, associated with the interferon (IFN) 21-gene signature and disease activity. Also, the LDG-to-lymphocyte ratio associated better with SLE disease activity index than neutrophil-to-lymphocyte ratio. SLE LDG exhibited significantly heightened surface expression of various activation markers and also of lectin-like oxidised low-density lipoprotein receptor-1, previously described to be associated with PMN-MDSC. Supernatants from SLE LDG did not restrict HD CD4+ T cell proliferation in an arginase-dependent manner, suggesting LDG are not immunosuppressive. SLE LDG supernatants induced proinflammatory cytokine production (IFN gamma, tumour necrosis factor alpha and lymphotoxin alpha) from CD4+ T cells. CONCLUSIONS: Based on our results, SLE LDG display an activated phenotype, exert proinflammatory effects on T cells and do not exhibit MDSC function. These results support the concept that LDG represent a distinct proinflammatory subset in SLE with pathogenic potential, at least in part, through their ability to activate type 1 helper responses.

Robustness of Deep Learning models in electrocardiogram noise detection and classification.

BACKGROUND AND OBJECTIVE: Automatic electrocardiogram (ECG) signal analysis for heart disease detection has gained significant attention due to busy lifestyles. However, ECG signals are susceptible to noise, which adversely affects the performance of ECG signal analysers. Traditional blind filtering methods use predefined noise frequency and filter order, but they alter ECG biomarkers. Several Deep Learning-based ECG noise detection and classification methods exist, but no study compares recurrent neural network (RNN) and convolutional neural network (CNN) architectures and their complexity. METHODS: This paper introduces a knowledge-based ECG filtering system using Deep Learning to classify ECG noise types and compare popular computer vision model architectures in a practical Internet of Medical Things (IoMT) framework. Experimental results demonstrate that the CNN-based ECG noise classifier outperforms the RNN-based model in terms of performance and training time. RESULTS: The study shows that AlexNet, visual geometry group (VGG), and residual network (ResNet) achieved over 70% accuracy, specificity, sensitivity, and F1 score across six datasets. VGG and ResNet performances were comparable, but VGG was more complex than ResNet, with only a 4.57% less F1 score. CONCLUSIONS: This paper introduces a Deep Learning (DL) based ECG noise classifier for a knowledge-driven ECG filtering system, offering selective filtering to reduce signal distortion. Evaluation of various CNN and RNN-based models reveals VGG and Resnet outperform. Further, the VGG model is superior in terms of performance. But Resnet performs comparably to VGG with less model complexity.

Study of Kinetics and the Working Mechanism of Silica-Coated Amino-Functionalized CoFe2O4 Ferrite Nanoparticles to Treat Wastewater for Heavy Metals.

This study used a simple coprecipitation method to produce pristine, silica-coated, and amino-functionalized CoFe2O4 nanoadsorbents. Amino-functionalization was done to increase the active surface area and metal ion removal efficiency. Both pristine and functionalized adsorbents were employed to recover Pb(II), Zn(II), and Cu(II) ions from wastewater. The adsorption tests were performed by varying the initial concentration of metal ions and contact time at a fixed pH of 6.5. Atomic adsorption spectroscopy was utilized to detect the proportion of metals removed from water. Additionally, the pseudo-first-order, pseudo-second-order, Freundlich, and Langmuir models were employed to compute the kinetic and isothermic data from metal ion adsorption onto the adsorbents. The amino-functionalized adsorbent showed adsorption capacities of 277.008, 254.453, and 258.398 mg/g for Cu(II), Pb(II), and Zn(II) ions, respectively. According to the adsorption results, the Langmuir isotherm and the pseudo-second-order model best suit the data. The best fitting of the pseudo-second-order model with the data indicates that coordinative interactions between amino groups and metal ions are responsible for chemisorption. The metal ions bind with -NH2 groups on the adsorbent surface through chelate bonds. Chelate bonds are extremely strong and stable, indicating the effectiveness of the CoFe2O4@SiO2-NH2 adsorbent in adsorbing heavy-metal ions. The tested adsorbent exhibited good performance, batter stability, and good reusable values around 77, 81, and 76% for Cu(II), Pb(II), and Zn(II) ions, respectively, after five adsorption cycles.

Nanofiltration PVP Membrane with Gradient Cross-Linking for Efficient Polycyclic Aromatic Hydrocarbons Removal.

Polycyclic aromatic hydrocarbons (PAHs) are persistent organic pollutants that may contaminate various water sources and pose serious dangers to human health and the environment. Due to their capacity for size-based separation, nanofiltration membranes have become efficient instruments for PAH removal. However, issues such as membrane fouling and ineffective rejection still exist. To improve PAH rejection while reducing fouling problems, this work created a new gradient cross-linking poly(vinylpyrrolidone) (PVP) nanofiltration membrane. The gradient cross-linking technique enhanced the rejection performance and antifouling characteristics of the membrane. The results demonstrated that the highest membrane flow was achieved at a 0.15% SDS-PVP membrane. There is a trade-off between membrane flux and salt rejection since salt rejection increases with SDS owing to the growth of big pores. The membrane flux was reduced for the 0.25% SDS-PVP membrane owing to poor SDS dispersion. The prepared membrane showed enhanced removal efficiencies for the removal of the PAH compounds. The PVP membrane has the potential to be used in several water treatment applications, improving water quality, and preserving the environment.

Murine FLT3 ligand-derived dendritic cell-mediated early immune responses are critical to controlling cell-free human T cell leukemia virus type 1 infection.

Human T cell leukemia virus type 1 (HTLV-1) is associated with two immunologically distinct diseases: HTLV-1-associated myelopathy/tropical spastic paraparesis and adult T cell leukemia. We observed previously that depletion of dendritic cells (DCs) in CD11c-diphtheria toxin receptor transgenic mice followed by infection with cell-free virus led to greater proviral and Tax mRNA loads and diminished cellular immune response compared with mice infected with cell-associated virus. To understand the significance of these in vivo results and explore the host-pathogen interaction between DCs and cell-free HTLV-1, we used FLT3 ligand-cultured mouse bone marrow-derived DCs (FL-DCs) and chimeric HTLV-1. Phenotypically, the FL-DCs upregulated expression of surface markers (CD80, CD86, and MHC class II) on infection; however, the level of MHC class I remained unchanged. We performed kinetic studies to understand viral entry, proviral integration, and expression of the viral protein Tax. Multiplex cytokine profiling revealed production of an array of proinflammatory cytokines and type 1 IFN (IFN-α) by FL-DCs treated with virus. Virus-matured FL-DCs stimulated proliferation of autologous CD3(+) T cells as shown by intracellular nuclear Ki67 staining and produced IFN-γ when cultured with infected FL-DCs. Gene expression studies using type 1 IFN-specific and DC-specific arrays revealed upregulation of IFN-stimulated genes, most cytokines, and transcription factors, but a distinct downregulation of many chemokines. Overall, these results highlight the critical early responses generated by FL-DCs on challenge with cell-free chimeric HTLV-1.

E-health: potential benefits and challenges in providing and accessing sexual health services.

BACKGROUND: E-health has become a burgeoning field in which health professionals and health consumers create and seek information. E-health refers to internet-based health care and information delivery and seeks to improve health service locally, regionally and worldwide. E-sexual health presents new opportunities to provide online sexual health services irrespective of gender, age, sexual orientation and location. DISCUSSION: The paper used the dimensions of the RE-AIM model (reach, efficacy, adoption, implementation and maintenance) as a guiding principle to discuss potentials of E-health in providing and accessing sexual health services. There are important issues in relation to utilising and providing online sexual health services. For healthcare providers, e-health can act as an opportunity to enhance their clients' sexual health care by facilitating communication with full privacy and confidentiality, reducing administrative costs and improving efficiency and flexibility as well as market sexual health services and products. Sexual health is one of the common health topics which both younger and older people explore on the internet and they increasingly prefer sexual health education to be interactive, non-discriminate and anonymous. This commentary presents and discusses the benefits of e-sexual health and provides recommendations towards addressing some of the emerging challenges. FUTURE DIRECTIONS: The provision of sexual health services can be enhanced through E-health technology. Doing this can empower consumers to engage with information technology to enhance their sexual health knowledge and quality of life and address some of the stigma associated with diversity in sexualities and sexual health experiences. In addition, e-sexual health may better support and enhance the relationship between consumers and their health care providers across different locations. However, a systematic and focused approach to research and the application of findings in policy and practice is required to ensure that E-health benefits all population groups and the information is current and clinically valid and effective, including preventative approaches for various client groups with diverse needs.

Nonlinear Features from Multi-Modal Signals for Continuous Stress Monitoring.

Continuous monitoring of stress in individuals during their daily activities has become an inevitable need in present times. Unattended stress is a silent killer and may lead to fatal physical and mental disorders if left unidentified. Stress identification based on individual judgement often leads to under-diagnosis and delayed treatment possibilities. EEG-based stress monitoring is quite popular in this context, but impractical to use for continuous remote monitoring.Continuous remote monitoring of stress using signals acquired from everyday wearables like smart watches is the best alternative here. Non-EEG data such as heart rate and ectodermal activity can also act as indicators of physiological stress. In this work, we have explored the possibility of using nonlinear features from non-EEG data such as (a) heart rate, (b) ectodermal activity, (c) body temperature (d) SpO2 and (e) acceleration in detecting four different types of neurological states; namely (1) Relaxed state, (2) State of Physical stress, (3) State of Cognitive stress and (4) State of Emotional stress. Physiological data of 20 healthy adults have been used from the noneeg database of PhysioNet.Results: We used two machine learning models; a linear logistic regression and a nonlinear random forest to detect (a) stress from relaxed state and (4) the four different neurological states. We trained the models using linear and nonlinear features separately. For the 2-class and 4-class problems, using nonlinear features increased the accuracy of the models. Moreover, it is also proved in this study that by using nonlinear features, we can avoid the use of complex machine learning models.

Impact of synthetic noise signature and physiologic ECG signal on designing ML-based ECG noise detection framework.

Automatic signal analysis using artificial intelligence is getting popular in digital healthcare, such as ECG rhythm analysis, where ECG signals are collected from traditional ECG machines or wearable ECG sensors. However, the risk of using an automated system for ECG analysis when noise is present can lead to incorrect diagnosis or treatment decisions. A noise detector is crucial to minimise the risk of incorrect diagnosis. Machine learning (ML) models are used in ECG noise detection before clinical decision-making systems to mitigate false alarms. However, it is essential to prove the generalisation capability of the ML model in different situations. ML models performance is 50% lesser when the model is trained with synthetic and tested with physiologic ECG datasets compared to trained and tested with physiologic ECG datasets. This suggests that the ML model must be trained with physiologic ECG datasets rather than synthetic ones or add more various types of noise in synthetic ECG datasets that can mimic physiologic ECG.Clinical relevance- ML model trained with synthetic noisy ECG can increase the 50% misclassification rate in ECG noise detection compared to training with physiologic ECG datasets. The wrong classification of noise-free and noisy ECG will lead to misdiagnosis regarding the patient's condition, which could be a cause of death.

Measurement of stress-induced sympathetic nervous activity using multi-wavelength PPG.

The onset of stress triggers sympathetic arousal (SA), which causes detectable changes to physiological parameters such as heart rate, blood pressure, dilation of the pupils and sweat release. The objective quantification of SA has tremendous potential to prevent and manage psychological disorders. Photoplethysmography (PPG), a non-invasive method to measure skin blood flow changes, has been used to estimate SA indirectly. However, the impact of various wavelengths of the PPG signal has not been investigated for estimating SA. In this study, we explore the feasibility of using various statistical and nonlinear features derived from peak-to-peak (AC) values of PPG signals of different wavelengths (green, blue, infrared and red) to estimate stress-induced changes in SA and compare their performances. The impact of two physical stressors: and Hand Grip are studied on 32 healthy individuals. Linear (Mean, s.d.) and nonlinear (Katz, Petrosian, Higuchi, SampEn, TotalSampEn) features are extracted from the PPG signal's AC amplitudes to identify the onset, continuation and recovery phases of those stressors. The results show that the nonlinear features are the most promising in detecting stress-induced sympathetic activity. TotalSampEn feature was capable of detecting stress-induced changes in SA for all wavelengths, whereas other features (Petrosian, AvgSampEn) are significant (AUC ≥ 0.8) only for IR and Red wavelengths. The outcomes of this study can be used to make device design decisions as well as develop stress detection algorithms.

Response Surface Methodology for the Synthesis and Characterization of Bio-Oil Extracted from Biomass Waste and Upgradation Using the Rice Husk Ash Catalyst.

Rice husk ash (RHA), a low-cost biomaterial, was utilized to form bio-oil from pyrolysis in a batch-stirred reactor, followed by its upgradation using the RHA catalyst. In the present study, the effect of temperature (ranging from 400 to 480 °C) on bio-oil production produced from RHA was studied to obtain the maximum bio-oil yield. Response surface methodology (RSM) was applied to investigate the effect of operational parameters (temperature, heating rate, and particle size) on the bio-oil yield. The results showed that a maximum bio-oil output of 20.33% was obtained at 480 °C temperature, 80 °C/min heating rate, and 200 µm particle size. Temperature and heating rate positively impact the bio-oil yield, while particle size has little effect. The overall R2 value of 0.9614 for the proposed model proved in good agreement with the experimental data. The physical properties of raw bio-oil were determined, and 1030 kg/m3 density, 12 MJ/kg calorific value, 1.40 cSt viscosity, 3 pH, and 72 mg KOH/g acid value were obtained, respectively. To enhance the characteristics of the bio-oil, upgradation was performed using the RHA catalyst through the esterification process. The upgraded bio-oil stemmed from a density of 0.98 g/cm3, an acid value of 58 mg of KOH/g, a calorific value of 16 MJ/kg, and a viscosity 10.5 cSt, respectively. The physical properties, GC-MS and FTIR, showed an improvement in the bio-oil characterization. The findings of this study indicate that RHA can be used as an alternative source for bio-oil production to create a more sustainable and cleaner environment.