Face Mask detection using Mask R-CNN to control the spread of Covid-19

Business executives are developing cutting-edge digital solutions as the virus outbreak spreads. A face mask detection system is one of them, and it can be used to spot people wearing them. Face mask identification software and applications have already been released by a few businesses, and others have promised to do the same for the service. The proposed work examines face mask detection accuracy using CNN networks. Mask wear is now required in many developed and developing countries worldwide when leaving the house or entering public spaces. It will be difficult to maintain touchless access control in buildings while recognising faces wearing masks on any surveillance systems. Masks covering faces has made face detection algorithms and performance difficult. The proposed work detect face mask labeled no mask or mask with detection accuracy. The work train the system to click images of a face and provide labeled data. The work is classified using Convolution Neural Network (CNN), a Deep learning technique, to classify the input image with the help of the classification algorithm MobileNetV2. The trained system shows whether a person in the video frame is wearing a mask or not. © 2023 IEEE.

Development and validation of a situational anxiety scale screening assessment among adults with type 2 diabetes during COVID-19 at a tertiary centre in Chennai, India

Aim: The aim of this study was to develop and validate the situational anxiety scale (SAS) during COVID-19 among adults with type 2 diabetes attending a tertiary diabetes center in Southern India. Materials and Methods: A total of 100 individuals aged from 18 to 65 years with type 2 diabetes attending a tertiary care diabetes center completed a structured SAS at two visits. The first visit (visit 1) survey was conducted in April 2021 and the second visit (visit 2) survey was conducted in March 2022. The SAS was administered to all 100 individuals. The State Trait Anxiety Inventory Scale (STAI-S) consisting of 20 questions was administered to the same 100 individuals in addition to the SAS during Visit 2. Results: The SAS showed good internal consistency for visit 1 (alpha = 0.855) and visit 2 (alpha = 0.795). Exploratory factor analysis showed four factors and explained 69% of variance. The four factors identified were as follows: (1) fear, (2) desire for COVID-free state, (3) lack of interest and energy, and (4) financial worries. A weak positive correlation was observed between SAS visit 2 and STAI-S, and it was statistically significant (r = 0.223;P = 0.026). Conclusion: The SAS is a valid and reliable tool for measuring situational anxiety during pandemics and post-COVID anxiety levels, which can help in the development of a holistic approach.

Evaluation of Twitter Sentiment towards COVID-19

The novel (COVID-19) pandemic is the most talked subject in web-based media stages in 2020. Individuals are utilizing web-based media, for example, Twitter to offer their viewpoint and offer data on various issues identified with the COVID-19 in this stay-At-home request. Here we have explored the conclusion and feeling of people groups in the U.S.A regarding the matter of resuming. We pick the online media stage Twitter for our examination and study the Tweets to find the nostalgic point of view, enthusiastic viewpoint, and setting off words towards the returning. During this COVID-19 pandemic, analysts have made some examination on different online media dataset with respect to lockdown and stay at home. Be that as it may, in our examination, we are especially in trigged to dissect public estimation on returning. Our significant finding is that when all states resorted to lockdown in March, individuals showed predominant feeling of dread, yet as resuming begins individuals have less dread. While this might be valid, because of this resuming stage day by day sure case surprising contrasted with the lockdown circumstance. Generally speaking, individuals have a more positive assumption towards the circumstance of resuming. © 2022 IEEE.

Advocacy and Adoption of Technology and Disparities in India

The number of people with diabetes globally, is rising at an alarming rate. South Asia is one of the hot spots of the diabetes epidemic. In India alone, there are over 74 million people with diabetes today. Unfortunately, 70% of the doctors in India practice in urban areas while 70% of India's population lives in rural areas. This mismatch between the availability of health care professionals and the rapid spread of diabetes in rural areas, provides an opportunity to use technology to deliver the diabetes care to remote rural areas. The first part of this presentation will talk about a model of successful delivery of diabetes health care in rural India. The Chunampet Rural Diabetes Program was carried out in a group of 42 villages in Kancheepuram District in Tamilnadu. Using a Mobile van, a population of 27,014 individuals (86.5% of the adult population) were screened for diabetes. All those detected with diabetes were offered a follow up care at a rural diabetes centre which was set up during the project. The results were very impressive and led to good improvement in A1c levels using low cost generic drugs. The second use of technology was during the COVID - 19 pandemic and the lock down which was enforced in India and many other countries. Thankfully, Telemedicine was also legalized in India at that time. Using technology, a system was created whereby the doctor and the patient stayed at home but blood tests were arranged at home for the patient.With the results, teleconsultation was done by doctors using the Electronic Medical Records which were made available on their mobile phones. Thus, despite the lockdown, patients managed to get their tests and diabetes consultations done remotely. The third use of technology is through our network of diabetes clinics across India. Even at centres where there was no ophthalmologist, retinal photographs were obtained using a lowcost retinal camera and were uploaded for centralized diabetic retinopathy grading unit where the images were read by trained retina specialists. The eye reports were sent back to the peripheral clinics in real time. Over one year period, 25,316 individuals with diabetes could have their eyes screened for diabetic retinopathy. Only 11.4 % needed referral to an ophthalmologist for further management. Finally, the use of mobile Apps has revolutionized diabetes treatment. Recently, we have developed three diabetes related tools. 'DIA' - an AI powered chatbot to assist people through automated digital conversations, 'DIALA' - a patientfriendly mobile app and 'DIANA' - a healthcare application for precision diabetes care. The details of these three tools are briefly described below : DIA : The Conversational AI Virtual Assistant 'DIA' can interact in English with its unique conversational AI technology and intuitive interface, it has proved to be a useful solution for patients, providing complex dialogues, with quick response time and offers comprehensive solutions for patients with diabetes. DIA's uses range from scheduling appointments and reminders for visits, lab tests and teleconsultation, to addressing enquiries on available medicines, treatments, and facilities.During an emergency, health crisis or in pandemic situations, it connects with caregivers and patients to take proper action as per the seriousness of their conditions. Further, it shares notifications, updates patient engagement and special offers. In addition to this, DIA can assist patients through reminders on their medicine refill via WhatsApp or SMS notifications and even facilitate purchase and tracking of medicine orders. DIALA : 'DIALA' is a DIAbetes Lifestyle Assistant Mobile Application. This app helps deliver superior and positive patient outcomes with weight tracking, step counts, diet plan adjustment, prescription refilling, availing reports of tests done, glucose monitoring data, scheduling appointments and sends reminders. It can help to monitor one's health and manage diabetes effectively. It is currently available in Android. DIANA : An advanced machine learning tool DIANA (DIAbetes Novel subgroup Assessment) is used to classify individuals with newly detected type 2 diabetes into specific subgroups such as insulin deficient or insulin resistance forms. This tool also gives the estimates of the risk for developing diabetes complications like eye or kidney disease. This machine learning approach has been developed based on published real world clinical data and will help the clinician offer individualized care for people with diabetes. In conclusion, judicious use of technology can help to bridge the socioeconomic and geographical challenges in delivering diabetes health care in developing countries.

Robotic Arm controlled using IoT application

Work-from-home policies have been the standard since the worldwide pandemic breakout, and this has spurred the fast development of applications in the area of IoT for remotely monitoring and managing applications. This has encouraged us to design and develop a remotely controlled robotic arm that can be used in applications where the engaging human hazardous environment (such as quarantined rooms of COVID affected patients) is dangerous. This has led to the development of a B-rover called a robotic arm, which the technicians remotely control to reduce the direct contact between the technician and the hazardous environment. It has various applications, such as a health monitoring system for monitoring the patient's health conditions, sample collection from the patients and the capability of the Robotic Arm to deliver medications to the COVID affected patients without engaging humans. It is proposed to design a 3DOF(degrees of freedom) robotic arm with stepper motor which is controlled through Wi-Fi using the BlynkIoT App with widgets like Joystick and Sliders. This will pick and drop the objects from one place to another. The results show that the designed robotic arm shows a 3% variation from the simulated and actual results when the slider is adjusted. © 2022

Epidemiological investigation of the COVID-19 outbreak in Vellore district in South India using Geographic Information Surveillance (GIS)

Background. Geographical Information Surveillance (GIS) is an advanced digital technology tool that maps location-based data and helps in epidemiological modeling. During pandemics a rapid analysis of patterns of spread can help put in place strategies for containment and infection control. We applied GIS to analyze patterns of spread and hotspots of COVID-19 infected cases in Vellore district in Tamil Nadu, South India. Methods. Laboratory-confirmed COVID-19 patients from Vellore district and neighboring taluks from March 2020 to June 2021 were geo coded (based on addresses) and spatial maps generated. These were then layered as points on the base map to illustrate the distribution of all COVID-19 cases. Time trends exploring urban-rural burden with age-sex distribution of COVID-19 cases and other variables were correlated with outcomes of death, symptoms and complications. Map of undivided Vellore district showing rural and urban settlements. Results. A total of 45,401 cases of COVID-19 were detected between 28 March 2020 to 31 June 2021 with 20730 cases during the first wave (28 March 2020 to 31 March 2021) and 24671 cases during the second wave (1 April 2021 to 30 June 2021). The overall incidence rates of COVID-19 across the study region was 462.8 per 100,000 and 588.6 per 100,000 population during the first and second waves respectively. Pattern of spread revealed epicentres in densely populated urban areas with radial spread, sparing rural areas, Heat maps also confirmed higher densities at these epicentres, however, the second wave had more peri-urban and rural area involvement. Case fatality rate was 1.89% and 1.6% during the first and second waves and increased with advancing age, i.e., 7.38% were aged more than 60 years in the first wave and 5.02% in the second wave. Incidence was higher in men, 2.40%, and 1.76% as compared to women who had 1.16% and 1.38% in the first and second waves respectively. Overall, case fatality rates were the highest among those who had >2 comorbidities (9.52%). Subdistrict level incidence of COVID-19 during the first and second waves. Epidemic curve of the COVID-19 pandemic during the first and the second waves. Conclusion. Modern surveillance systems like GIS can accurately predict the trends of the outbreak and pattern of spread during future respiratory pandemics. Employing this in real time can help design risk mitigation strategies improving health care access and monitoring with prevention of spread to rural areas.

A Study on the Health and Socioeconomic Impact of COVID-19 Pandemic and Barriers to Self-management of Diabetes during the Lockdown among Rural Residents of South India

Aim: To study the health and socioeconomic impact of the COVID-19 pandemic and to assess the barriers to self-management of diabetes during the lockdown, in rural South India. Materials and Methods: Details of demographic, social, economic, migration and health status were collected using a structured questionnaire from participants aged & GE;18 years belonging to the 21 villages of Chengalpattu and Kancheepuram districts of Tamil Nadu state in south India as part of the Telemedicine pRoject for screENing Diabetes and its complications in rural Tamil Nadu (TREND) study. From the 11,249 TREND participants, a random list of 25% (n = 2812) was system-generated using random numbers and 2812 participants were contacted for the study, of whom 2511 individuals participated. Telephonic interviews were conducted during the lockdown from June to August 2020. Further, qualitative interviews(Focus group discussions) were conducted among 27 individuals with diabetes between September and December 2020. Data were analyzed using thematic analysis. Results: The mean age of the study population was 43 & PLUSMN;14 years and 50.4% were women. Diabetes was present in 14.7%, hypertension in 31.9%, generalized and abdominal obesity in 33.3% and 46.5% respectively. When the lockdown was implemented in March 2020, 37% had migrated from urban to rural areas. Lack of daily wage jobs (68%), price of essential commodities (41.7%), social distancing/curfew (34.8%), mental fatigue/depression (14.7%), and loss of job (7.1%) were some reasons stated for their adverse social and financial circumstances. People with diabetes stated that they had to avoid or cut down their regular hospital visits due to travel restrictions. Many of the patients took the same medications for almost a year. Conclusion: Unemployment, poor mental health, and reduced household income were the most significant negative impacts faced by rural residents during the lockdown due to COVID-19. People with diabetes experienced disruptions in diabetes management due to the pandemic.

Atypical Presentation of Antenatal Eclampsia.

Most women who develop eclampsia have preceding preeclampsia (proteinuria and hypertension). This is especially true for otherwise healthy nulliparous women. However, recently, there has been a paradigm shift in this philosophy. There is mounting evidence that preeclampsia can develop even in the absence of proteinuria and hypertension and that eclampsia itself may be the initial manifestation of hypertensive disorder during pregnancy. We report a rare case of a 24-year-old primigravida at 30 weeks of gestation who presented with new-onset generalised tonic-clonic seizures without prior hypertension or proteinuria in her antenatal records. A thorough workup revealed this presentation to be the initial feature of atypical eclampsia. She was managed appropriately and discharged with an excellent outcome. This experience highlights some of the difficulties in managing a case of atypical eclampsia, namely, erratic onset and an unpredictable course, all of which interfere with timely diagnosis and treatment and contribute to maternal and fetal morbidity and mortality.

Automating the Body Temperature Scanning System and Predictive Analysis Using Machine Learning Models

In December 2019, the COVID-19 broke out. From that point, the situation has become much dire. The number of cases kept spiking and a cure is still unknown for COVID-19. For this reason, we must be more cautious and take all possible precautions. We know a few things about this disease. Fever happens to be one of the early symptoms of COVID-19. Hence, we do thermal scanning in public places. Our paper proposes a way to make this process more efficient. We can scan body temperature using various sensors and store it in the cloud. Doing so, it gives us more flexibility to monitor the data and predict if someone might suffer from fever in the future. In our analysis, we have found that among the different machine learning algorithms, moving averages smoothing was able to predict the data better. Now, in order to run this machine learning model automatically, we used AWS. Also, due to GUI, it is much easier to use the system. Overall, the main purpose of our work is to collect daily thermal scan reports and use that data for our benefit. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

EXPERIENCE IN INDIA

The number of people with diabetes globally, is rising at an alarming rate. South Asia is one of the hot spots of the diabetes epidemic. In India alone, there are over 74 million people with diabetes today. Unfortunately, 70% of the doctors in India practice in urban areas while 70% of India's population lives in rural areas. This mismatch between the availability of health care professionals and the rapid spread of diabetes in rural areas, provides an opportunity to use technology to deliver the diabetes care to remote rural areas. The first part of this presentation will talk about a model of successful delivery of diabetes health care in rural India. The Chunampet Rural Diabetes Program was carried out in a group of 42 villages in Kancheepuram District in Tamilnadu. Using a Mobile van, a population of 27,014 individuals (86.5% of the adult population) were screened for diabetes. All those detected with diabetes were offered a follow up care at a rural diabetes centre which was set up during the project. The results were very impressive and led to good improvement in A1c levels using low cost generic drugs. The second use of technology was during the COVID ± 19 pandemic and the lock down which was enforced in India. Thankfully, Telemedicine was also legalized in India at that time. Using technology, a system was created whereby the doctor and the patient stayed at home but blood tests were arranged at home for the patient. With the results, teleconsultation was done by doctors using the Electronic Medical Records which were made available on their mobile phones. Thus, despite the lockdown, patients managed to get their tests and diabetes consultations done remotely. The third use of technology which will be presented is through our network of diabetes clinics across India. Even at centres where there was no ophthalmologist, retinal photographs were obtained using a lowcost retinal camera and were uploaded for centralized diabetic retinopathy grading unit where the images were read by trained retina specialists. The eye reports were sent back to the peripheral clinics in real time. Over one year period, 25,316 individuals with diabetes could have their eyes screened for diabetic retinopathy. Only 11.4 % needed referral to an ophthalmologist for further management. In conclusion, judicious use of technology can help to bridge the socioeconomic and geographical challenges in delivering diabetes health care in developing countries.

Detection of COVID-19 from Chest X-ray Images: A Deep Learning Approach

In the current COVID-19 pandemic situation, there is an urgent need to properly diagnose whether people are infected by COVID-19 or not. Fast and accurate methods should be there to improve the efficiency of the health care system so that the infected people can be given priority treatment. Deep learning methods are largely incorporated in many medical fields, especially in medical diagnosis. Deep learning techniques could find patterns that can be attributed to various diseases. The main challenge in applying deep learning techniques in the medical field is the lack of availability of quality labelled data. Machine learning approaches, such as One-shot learning methods, are becoming increasingly popular in the medical community because they perform better with limited data. We compared several deep learning models for COVID image classification in this paper. State-of-the-art architectures like ResNet, EfficientNet were compared. We also present a technique that combines the triplet loss and cross-entropy loss functions. This technique enables the model to learn weights in such a way that it attempts to cluster different classes during classification. It will increase the model's interpretability and group together similar data. The dataset we used was made available as part of the Chest XR COVID 19 detection challenge. EfficientNet B7 model got the best result on the test set with 95.67% accuracy. Using Siamese Network, we were able to embed the images into a lower dimension in such a way that they can be clustered into different groups. Classification based on this embedded space obtained an accuracy of 93.76% in the test set. © 2021 IEEE.

Impact of COVID-19 on noncommunicable diseases (NCDs)

Since December 2019, a novel coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), pandemic has affected more than 18.6 million people worldwide. Male gender, older age, obesity, and comorbid noncommunicable diseases (NCDs) like diabetes, hypertension, cardiovascular disease (CVD), chronic respiration illnesses, and cancer have higher risk and fatal outcome of COVID-19. India has a huge burden of NCDs and their associated risk factors which could act in harmony with COVID-19 to produce severe and fatal outcome. Till date the specific treatment options for COVID-19 are elusive and as NCDs are reported as the main causative risk factors for COVID-19 which can worsen the outcome, the focus should be made on continuing and improving the healthcare facilities related to the prevention, management, and control of NCDs. The management of NCDs in the context of SARS-CoV-2 infection are quite challenging. The restrictive measures imposed by governments all over the world such as complete or partial lockdown, travel restrictions, and physical distancing to contain the spread of SARS-CoV-2 infection have affected the people with NCDs by limiting their access to healthcare facilities, physical activity access to healthy food, and even to medicines and essential supplies. These factors increase the risk of developing obesity, diabetes, and CVDs. This article reviews the burden of NCDs in India, the cross-connection between NCDs and COVID-19, disruptions of healthcare services for NCDs, and proposes research priorities during COVID-19 for effective management and control of NCDs.

COVID-19 and tobacco cessation: lessons from India.

OBJECTIVES: The Government of India prohibited the sale of tobacco products during the COVID-19 lockdown to prevent the spread of the SARS-CoV-2 virus. This study assessed the tobacco cessation behaviour and its predictors among adult tobacco users during the initial COVID-19 lockdown period in India. METHODS: A cross-sectional study was conducted with 801 adult tobacco users (both smoking and smokeless tobacco) in two urban metropolitan cities of India over a 2-month period (July to August 2020). The study assessed complete tobacco cessation and quit attempts during the lockdown period. Logistic and negative binomial regression models were used to study the correlates of tobacco cessation and quit attempts, respectively. RESULTS: In total, 90 (11.3%) tobacco users reported that they had quit using tobacco after the COVID-19 lockdown period. Overall, a median of two quit attempts (interquartile range 0-6) was made by tobacco users. Participants with good knowledge on the harmful effects of tobacco use and COVID-19 were significantly more likely to quit tobacco use (odds ratio [OR] 2.2; 95% confidence interval [CI] 1.2-4.0) and reported more quit attempts (incidence risk ratio 5.7; 95% CI 2.8-11.8) compared to those with poor knowledge. Participants who had access to tobacco products were less likely to quit tobacco use compared to those who had no access (OR 0.3; 95% CI 0.2-0.5]. CONCLUSIONS: Access restrictions and correct knowledge on the harmful effects of tobacco use and COVID-19 can play an important role in creating a conducive environment for tobacco cessation among users.

COVID-19 associated mucormycosis: A Descriptive Multisite Study from India.

BACKGROUND AND AIMS: Mucormycosis is an invasive fungal infection and carries a significant morbidity and mortality. A number of cases of mucormycosis have been reported in association with COVID-19. In this study, a consortium of clinicians from various parts of India studied clinical profile of COVID-19 associated mucormycosis (CAM) and this analysis is presented here. METHODS: Investigators from multiple sites in India were involved in this study. Clinical details included the treatment and severity of COVID-19, associated morbidities, as well as the diagnosis, treatment and prognosis of mucormycosis. These data were collected using google spreadsheet at one centre. Descriptive analysis was done. RESULTS: There were 115 patients with CAM. Importantly, all patients had received corticosteroids. Diabetes was present in 85.2% of patients and 13.9% of patients had newly detected diabetes. The most common site of involvement was rhino-orbital. Mortality occurred in 25 (21.7%) patients. On logistic regression analysis, CT scan-based score for severity of lung involvement was associated with mortality. CONCLUSION: Universal administration of corticosteroids in our patients is notable. A large majority of patients had diabetes, while mortality was seen in ∼1/5th of patients, lower as compared to recently published data.

Survey of diabetologists and obstetricians' practice patterns related to care for gestational diabetes mellitus during the COVID-19 pandemic in India

Aim: There are limited data on the management of gestational diabetes mellitus (GDM) during the coronavirus disease 2019 (COVID-19) pandemic. This survey was carried out in India to understand the practice patterns of diabetologists and obstetricians (OBs) during the pandemic. Materials and Methods: An online questionnaire was designed, and the link to the survey was shared with doctors through email. Questions were related to the diagnosis and management of GDM both before and during the COVID-19 pandemic. Results: A total of 117 diabetologists and 90 OBs from different parts of India participated in the survey. During the COVID-19 pandemic, diabetologists carried out higher random glucose and HbA1c tests and lower numbers of oral glucose tolerance tests (OGTTs), but differences compared with before COVID-19 were nonsignificant. The OBs reported doing a significantly lower number of OGTTs (85.6% vs. 95.6%, P = 0.02) and significantly more HbA1c tests (16.7% vs. 5.6%, P = 0.03) and self-monitoring of blood glucose (59.4% vs. 37.1%, P < 0.0001) during the pandemic, than earlier. Although 97% of all the doctors surveyed reported using some form of telemedicine, several challenges were identified. Conclusion: The COVID-19 pandemic has resulted in changes in the management of women with GDM. The use of digital technologies could help improve the care of women with GDM during such pandemics.