Flamingo-Optimization-Based Deep Convolutional Neural Network for IoT-Based Arrhythmia Classification.

Cardiac arrhythmia is a deadly disease that threatens the lives of millions of people, which shows the need for earlier detection and classification. An abnormal signal in the heart causing arrhythmia can be detected at an earlier stage when the health data from the patient are monitored using IoT technology. Arrhythmias may suddenly lead to death and the classification of arrhythmias is considered a complicated process. In this research, an effective classification model for the classification of heart disease is developed using flamingo optimization. Initially, the ECG signal from the heart is collected and then it is subjected to the preprocessing stage; to detect and control the electrical activity of the heart, the electrocardiogram (ECG) is used. The input signals collected using IoT nodes are collectively presented in the base station for the classification using flamingo-optimization-based deep convolutional networks, which effectively predict the disease. With the aid of communication technologies and the contribution of IoT, medical professionals can easily monitor the health condition of patients. The performance is analyzed in terms of accuracy, sensitivity, and specificity.

An accurate generation of image captions for blind people using extended convolutional atom neural network.

Recently, the progress on image understanding and AIC (Automatic Image Captioning) has attracted lots of researchers to make use of AI (Artificial Intelligence) models to assist the blind people. AIC integrates the principle of both computer vision and NLP (Natural Language Processing) to generate automatic language descriptions in relation to the image observed. This work presents a new assistive technology based on deep learning which helps the blind people to distinguish the food items in online grocery shopping. The proposed AIC model involves the following steps such as Data Collection, Non-captioned image selection, Extraction of appearance, texture features and Generation of automatic image captions. Initially, the data is collected from two public sources and the selection of non-captioned images are done using the ARO (Adaptive Rain Optimization). Next, the appearance feature is extracted using SDM (Spatial Derivative and Multi-scale) approach and WPLBP (Weighted Patch Local Binary Pattern) is used in the extraction of texture features. Finally, the captions are automatically generated using ECANN (Extended Convolutional Atom Neural Network). ECANN model combines the CNN (Convolutional Neural Network) and LSTM (Long Short-Term Memory) architectures to perform the caption reusable system to select the most accurate caption. The loss in the ECANN architecture is minimized using AAS (Adaptive Atom Search) Optimization algorithm. The implementation tool used is PYTHON and the dataset used for the analysis are Grocery datasets (Freiburg Groceries and Grocery Store Dataset). The proposed ECANN model acquired accuracy (99.46%) on Grocery Store Dataset and (99.32%) accuracy on Freiburg Groceries dataset. Thus, the performance of the proposed ECANN model is compared with other existing models to verify the supremacy of the proposed work over the other existing works.

Detection and diagnosis of COVID-19 infection in lungs images using deep learning techniques.

World's science and technologies have been challenged by the COVID-19 pandemic. Each and every community across the globe are trying to find a real-time novel method for accurate treatment and cure of COVID-19 infected patients. The most important lead to take from this pandemic is to detect the infected patients as soon as possible and provide them an accurate treatment. At present, the worldwide methodology to detect COVID-19 is reverse transcription-polymerase chain reaction (RT-PCR). This technique is costly and time taking. For this reason, the implementation of a novel method is required. This paper includes the use of deep learning analysis to develop a system for identifying COVID-19 patients. Proposed technique is based on convolution neural network (CNN) and deep neural network (DNN). This paper proposes two models, first is designing DNN on the basis of fractal feature of the images and second is designing CNN using lungs x-ray images. To find the infected area (tissues) of the lungs image using CNN architecture, segmentation process has been used. Developed CNN architecture gave results of classification with accuracy equal to 94.6% and sensitivity equal to 90.5% which is much better than the proposed DNN method, which gave accuracy 84.11% and sensitivity 84.7%. The outcome of the presented model shows 94.6% accuracy in detecting infected regions. Using this method the growth of the infected regions can be monitored and controlled. The designed model can also be used in post-COVID-19 analysis.

Dynamic data processing system for sports training system using internet of things.

BACKGROUND: The Internet of Things (IoT) has recently become a prevalent technological culture in the sports training system. Although numerous technologies have grown in the sports training system domain, IoT plays a substantial role in its optimized health data processing framework for athletes during workouts. OBJECTIVE: In this paper, a Dynamic data processing system (DDPS) has been suggested with IoT assistance to explore the conventional design architecture for sports training tracking. METHOD: To track and estimate sportspersons physical activity in day-to-day living, a new paradigm has been combined with wearable IoT devices for efficient data processing during physical workouts. Uninterrupted observation and review of different sportspersons condition and operations by DDPS helps to assess the sensed data to analyze the sportspersons health condition. Additionally, Deep Neural Network (DNN) has been presented to extract important sports activity features. RESULTS: The numerical results show that the suggested DDPS method enhances the accuracy of 94.3%, an efficiency ratio of 98.2, less delay of 24.6%, error range 28.8%, and energy utilization of 31.2% compared to other existing methods.

Taylor and Gradient Descent-Based Actor Critic Neural Network for the Classification of Privacy Preserved Medical Data.

Classification of the privacy preserved medical data is the domain of the researchers as it stirs the importance behind hiding the sensitive data from the third-party authenticator. Ensuring the privacy of the medical records and using the disease prediction mechanisms played a remarkable role in peoples' lives such that the earlier detection of the diseases is required for earlier diagnosis. Accordingly, this article proposes a method, named Taylor gradient descent (TGD)-based actor critic neural network (ACNN), which concentrates on performing the medical data classification. Initially, the privacy of the medical data is ensured by using the key matrix developed based on the privacy utility coefficient matrix using the chronological-Whale optimization algorithm. The privacy protected data are subjected to classification by using ACNN that performs the optimal classification using the proposed TGD algorithm. The proposed TGD algorithm is the integration of Taylor series in the gradient descent algorithm that updates the optimal weight of ACNN based on the weights in the previous iterations. The analysis using the Cleveland, Switzerland, and Hungarian dataset proves that the proposed classification strategy obtains an accuracy of 0.9252, a sensitivity of 0.8419, and a specificity of 0.8387, respectively.

Handwashing, sanitation and family planning practices are the strongest underlying determinants of child stunting in rural indigenous communities of Jharkhand and Odisha, Eastern India: a cross-sectional study.

The World Health Organisation has called for global action to reduce child stunting by 40% by 2025. One third of the world's stunted children live in India, and children belonging to rural indigenous communities are the worst affected. We sought to identify the strongest determinants of stunting among indigenous children in rural Jharkhand and Odisha, India, to highlight key areas for intervention. We analysed data from 1227 children aged 6-23.99 months and their mothers, collected in 2010 from 18 clusters of villages with a high proportion of people from indigenous groups in three districts. We measured height and weight of mothers and children, and captured data on various basic, underlying and immediate determinants of undernutrition. We used Generalised Estimating Equations to identify individual determinants associated with children's height-for-age z-score (HAZ; p < 0.10); we included these in a multivariable model to identify the strongest HAZ determinants using backwards stepwise methods. In the adjusted model, the strongest protective factors for linear growth included cooking outdoors rather than indoors (HAZ +0.66), birth spacing ≥24 months (HAZ +0.40), and handwashing with a cleansing agent (HAZ +0.32). The strongest risk factors were later birth order (HAZ -0.38) and repeated diarrhoeal infection (HAZ -0.23). Our results suggest multiple risk factors for linear growth faltering in indigenous communities in Jharkhand and Odisha. Interventions that could improve children's growth include reducing exposure to indoor air pollution, increasing access to family planning, reducing diarrhoeal infections, improving handwashing practices, increasing access to income and strengthening health and sanitation infrastructure.

Effect of participatory women's groups facilitated by Accredited Social Health Activists on birth outcomes in rural eastern India: a cluster-randomised controlled trial.

BACKGROUND: A quarter of the world's neonatal deaths and 15% of maternal deaths happen in India. Few community-based strategies to improve maternal and newborn health have been tested through the country's government-approved Accredited Social Health Activists (ASHAs). We aimed to test the effect of participatory women's groups facilitated by ASHAs on birth outcomes, including neonatal mortality. METHODS: In this cluster-randomised controlled trial of a community intervention to improve maternal and newborn health, we randomly assigned (1:1) geographical clusters in rural Jharkhand and Odisha, eastern India to intervention (participatory women's groups) or control (no women's groups). Study participants were women of reproductive age (15-49 years) who gave birth between Sept 1, 2009, and Dec 31, 2012. In the intervention group, ASHAs supported women's groups through a participatory learning and action meeting cycle. Groups discussed and prioritised maternal and newborn health problems, identified strategies to address them, implemented the strategies, and assessed their progress. We identified births, stillbirths, and neonatal deaths, and interviewed mothers 6 weeks after delivery. The primary outcome was neonatal mortality over a 2 year follow up. Analyses were by intention to treat. This trial is registered with ISRCTN, number ISRCTN31567106. FINDINGS: Between September, 2009, and December, 2012, we randomly assigned 30 clusters (estimated population 156 519) to intervention (15 clusters, estimated population n=82 702) or control (15 clusters, n=73 817). During the follow-up period (Jan 1, 2011, to Dec 31, 2012), we identified 3700 births in the intervention group and 3519 in the control group. One intervention cluster was lost to follow up. The neonatal mortality rate during this period was 30 per 1000 livebirths in the intervention group and 44 per 1000 livebirths in the control group (odds ratio [OR] 0.69, 95% CI 0·53-0·89). INTERPRETATION: ASHAs can successfully reduce neonatal mortality through participatory meetings with women's groups. This is a scalable community-based approach to improving neonatal survival in rural, underserved areas of India. FUNDING: Big Lottery Fund (UK).

Improved neonatal survival after participatory learning and action with women's groups: a prospective study in rural eastern India.

OBJECTIVE: To determine whether a women's group intervention involving participatory learning and action has a sustainable and replicable effect on neonatal survival in rural, eastern India. METHODS: From 2004 to 2011, births and neonatal deaths in 36 geographical clusters in Jharkhand and Odisha were monitored. Between 2005 and 2008, these clusters were part of a randomized controlled trial of how women's group meetings involving participatory learning and action influence maternal and neonatal health. Between 2008 and 2011, groups in the original intervention clusters (zone 1) continued to meet to discuss post-neonatal issues and new groups in the original control clusters (zone 2) met to discuss neonatal health. Logistic regression was used to examine neonatal mortality rates after 2008 in the two zones. FINDINGS: Data on 41,191 births were analysed. In zone 1, the intervention's effect was sustained: the cluster-mean neonatal mortality rate was 34.2 per 1000 live births (95% confidence interval, CI: 28.3-40.0) between 2008 and 2011, compared with 41.3 per 1000 live births (95% CI: 35.4-47.1) between 2005 and 2008. The effect of the intervention was replicated in zone 2: the cluster-mean neonatal mortality rate decreased from 61.8 to 40.5 per 1000 live births between two periods: 2006-2008 and 2009-2011 (odds ratio: 0.69, 95% CI: 0.57-0.83). Hygiene during delivery, thermal care of the neonate and exclusive breastfeeding were important factors. CONCLUSION: The effect of participatory women's groups on neonatal survival in rural India, where neonatal mortality is high, was sustainable and replicable.

Predictors of maternal psychological distress in rural India: a cross-sectional community-based study.

BACKGROUND: Maternal common mental disorders are prevalent in low-resource settings and have far-reaching consequences for maternal and child health. We assessed the prevalence and predictors of psychological distress as a proxy for common mental disorders among mothers in rural Jharkhand and Orissa, eastern India, where over 40% of the population live below the poverty line and access to reproductive and mental health services is low. METHOD: We screened 5801 mothers around 6 weeks after delivery using the Kessler-10 item scale, and identified predictors of distress using multiple hierarchical logistic regression. RESULTS: 11.5% (95% CI: 10.7-12.3) of mothers had symptoms of distress (K10 score >15). High maternal age, low asset ownership, health problems in the antepartum, delivery or postpartum periods, caesarean section, an unwanted pregnancy for the mother, small perceived infant size and a stillbirth or neonatal death were all independently associated with an increased risk of distress. The loss of an infant or an unwanted pregnancy increased the risk of distress considerably (AORs: 7.06 95% CI: 5.51-9.04 and 1.49, 95% CI: 1.12-1.97, respectively). LIMITATIONS: We did not collect data on antepartum depression, domestic violence or a mother's past birth history, and were therefore unable to examine the importance of these factors as predictors of psychological distress. CONCLUSIONS: Mothers living in underserved areas of India who experience infant loss, an unwanted pregnancy, health problems in the perinatal and postpartum periods and socio-economic disadvantage are at increased risk of distress and require access to reproductive healthcare with integrated mental health interventions.

Community mobilisation with women's groups facilitated by Accredited Social Health Activists (ASHAs) to improve maternal and newborn health in underserved areas of Jharkhand and Orissa: study protocol for a cluster-randomised controlled trial.

BACKGROUND: Around a quarter of the world's neonatal and maternal deaths occur in India. Morbidity and mortality are highest in rural areas and among the poorest wealth quintiles. Few interventions to improve maternal and newborn health outcomes with government-mandated community health workers have been rigorously evaluated at scale in this setting.The study aims to assess the impact of a community mobilisation intervention with women's groups facilitated by ASHAs to improve maternal and newborn health outcomes among rural tribal communities of Jharkhand and Orissa. METHODS/DESIGN: The study is a cluster-randomised controlled trial and will be implemented in five districts, three in Jharkhand and two in Orissa. The unit of randomisation is a rural cluster of approximately 5000 population. We identified villages within rural, tribal areas of five districts, approached them for participation in the study and enrolled them into 30 clusters, with approximately 10 ASHAs per cluster. Within each district, 6 clusters were randomly allocated to receive the community intervention or to the control group, resulting in 15 intervention and 15 control clusters. Randomisation was carried out in the presence of local stakeholders who selected the cluster numbers and allocated them to intervention or control using a pre-generated random number sequence. The intervention is a participatory learning and action cycle where ASHAs support community women's groups through a four-phase process in which they identify and prioritise local maternal and newborn health problems, implement strategies to address these and evaluate the result. The cycle is designed to fit with the ASHAs' mandate to mobilise communities for health and to complement their other tasks, including increasing institutional delivery rates and providing home visits to mothers and newborns. The trial's primary endpoint is neonatal mortality during 24 months of intervention. Additional endpoints include home care practices and health care-seeking in the antenatal, delivery and postnatal period. The impact of the intervention will be measured through a prospective surveillance system implemented by the project team, through which mothers will be interviewed around six weeks after delivery. Cost data and qualitative data are collected for cost-effectiveness and process evaluations. STUDY REGISTRATION: ISRCTN: ISRCTN31567106.

Intracluster correlation coefficients and coefficients of variation for perinatal outcomes from five cluster-randomised controlled trials in low and middle-income countries: results and methodological implications.

BACKGROUND: Public health interventions are increasingly evaluated using cluster-randomised trials in which groups rather than individuals are allocated randomly to treatment and control arms. Outcomes for individuals within the same cluster are often more correlated than outcomes for individuals in different clusters. This needs to be taken into account in sample size estimations for planned trials, but most estimates of intracluster correlation for perinatal health outcomes come from hospital-based studies and may therefore not reflect outcomes in the community. In this study we report estimates for perinatal health outcomes from community-based trials to help researchers plan future evaluations. METHODS: We estimated the intracluster correlation and the coefficient of variation for a range of outcomes using data from five community-based cluster randomised controlled trials in three low-income countries: India, Bangladesh and Malawi. We also performed a simulation exercise to investigate the impact of cluster size and number of clusters on the reliability of estimates of the coefficient of variation for rare outcomes. RESULTS: Estimates of intracluster correlation for mortality outcomes were lower than those for process outcomes, with narrower confidence intervals throughout for trials with larger numbers of clusters. Estimates of intracluster correlation for maternal mortality were particularly variable with large confidence intervals. Stratified randomisation had the effect of reducing estimates of intracluster correlation. The simulation exercise showed that estimates of intracluster correlation are much less reliable for rare outcomes such as maternal mortality. The size of the cluster had a greater impact than the number of clusters on the reliability of estimates for rare outcomes. CONCLUSIONS: The breadth of intracluster correlation estimates reported here in terms of outcomes and contexts will help researchers plan future community-based public health interventions around maternal and newborn health. Our study confirms previous work finding that estimates of intracluster correlation are associated with the prevalence of the outcome of interest, the nature of the outcome of interest (mortality or behavioural) and the size and number of clusters. Estimates of intracluster correlation for maternal mortality need to be treated with caution and a range of estimates should be used in planning future trials.

Explaining the impact of a women's group led community mobilisation intervention on maternal and newborn health outcomes: the Ekjut trial process evaluation.

BACKGROUND: Few large and rigorous evaluations of participatory interventions systematically describe their context and implementation, or attempt to explain the mechanisms behind their impact. This study reports process evaluation data from the Ekjut cluster-randomised controlled trial of a participatory learning and action cycle with women's groups to improve maternal and newborn health outcomes in Jharkhand and Orissa, eastern India (2005-2008). The study demonstrated a 45% reduction in neonatal mortality in the last two years of the intervention, largely driven by improvements in safe practices for home deliveries. METHODS: A participatory learning and action cycle with 244 women's groups was implemented in 18 intervention clusters covering an estimated population of 114 141. We describe the context, content, and implementation of this intervention, identify potential mechanisms behind its impact, and report challenges experienced in the field. Methods included a review of intervention documents, qualitative structured discussions with group members and non-group members, meeting observations, as well as descriptive statistical analysis of data on meeting attendance, activities, and characteristics of group attendees. RESULTS: Six broad, interrelated factors influenced the intervention's impact: (1) acceptability; (2) a participatory approach to the development of knowledge, skills and 'critical consciousness'; (3) community involvement beyond the groups; (4) a focus on marginalized communities; (5) the active recruitment of newly pregnant women into groups; (6) high population coverage. We hypothesize that these factors were responsible for the increase in safe delivery and care practices that led to the reduction in neonatal mortality demonstrated in the Ekjut trial. CONCLUSIONS: Participatory interventions with community groups can influence maternal and child health outcomes if key intervention characteristics are preserved and tailored to local contexts. Scaling-up such interventions requires (1) a detailed understanding of the way in which context affects the acceptability and delivery of the intervention; (2) planned but flexible replication of key content and implementation features; (3) strong support for participatory methods from implementing agencies.

Effect of a participatory intervention with women's groups on birth outcomes and maternal depression in Jharkhand and Orissa, India: a cluster-randomised controlled trial.

BACKGROUND: Community mobilisation through participatory women's groups might improve birth outcomes in poor rural communities. We therefore assessed this approach in a largely tribal and rural population in three districts in eastern India. METHODS: From 36 clusters in Jharkhand and Orissa, with an estimated population of 228 186, we assigned 18 clusters to intervention or control using stratified randomisation. Women were eligible to participate if they were aged 15-49 years, residing in the project area, and had given birth during the study. In intervention clusters, a facilitator convened 13 groups every month to support participatory action and learning for women, and facilitated the development and implementation of strategies to address maternal and newborn health problems. The primary outcomes were reductions in neonatal mortality rate (NMR) and maternal depression scores. Analysis was by intention to treat. This trial is registered as an International Standard Randomised Controlled Trial, number ISRCTN21817853. FINDINGS: After baseline surveillance of 4692 births, we monitored outcomes for 19 030 births during 3 years (2005-08). NMRs per 1000 were 55.6, 37.1, and 36.3 during the first, second, and third years, respectively, in intervention clusters, and 53.4, 59.6, and 64.3, respectively, in control clusters. NMR was 32% lower in intervention clusters adjusted for clustering, stratification, and baseline differences (odds ratio 0.68, 95% CI 0.59-0.78) during the 3 years, and 45% lower in years 2 and 3 (0.55, 0.46-0.66). Although we did not note a significant effect on maternal depression overall, reduction in moderate depression was 57% in year 3 (0.43, 0.23-0.80). INTERPRETATION: This intervention could be used with or as a potential alternative to health-worker-led interventions, and presents new opportunities for policy makers to improve maternal and newborn health outcomes in poor populations. FUNDING: Health Foundation, UK Department for International Development, Wellcome Trust, and the Big Lottery Fund (UK).