Hybrid CNN-SVM Model for Face Mask Detector to Protect from COVID-19

The 2019 coronavirus outbreak (COVID-19) has had a huge impact on humanity. By May 2021, nearly, 172 million people worldwide were affected by the infectious spread of COVID-19. While the distribution of vaccines has already begun, mass distribution around the world has yet to take place. According to the World Health Organization (WHO), wearing a face mask can significantly reduce the spread of the COVID-19 virus. However, even improper wearing of face mask can prevent the purposes and lead to the spread of the virus. Under the influence of public health and the global economy, an effective Covid-19 pandemic strategy requires a lot of attention of humanity. To prevent the spread of such deadly virus, intelligent techniques are required. In the proposed work, an intelligent face mask detector framework is proposed based on deep learning concept which can classify the person who wear mask from those who are not wearing mask. In the proposed work, a hybrid model of convolution neural network with support vector machine is used for designing the mask detector. The performance of the proposed method is evaluated on real-world masked face recognition dataset (RMFD) and medical mask dataset (MDD). When implemented, it has been found that the proposed method can achieve high accuracy (99.11%). The excellent performance of the proposed model is very suitable for video surveillance equipment also. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

A novel DCNN-ELM hybrid framework for face mask detection

The Coronavirus disease (2019) has caused massive destruction of human lives and capital around the world. The latest variant Omicron is proved to be the most infectious of all its previous counterparts – Alpha, Beta and Delta. Various measures are identified, tested and implemented to minimize the attack on humans. Face masks are one of those measures that are shown to be very effective in containing the infection. However, it requires continuous monitoring for law enforcement. In the present manuscript, a detailed research investigation using different ablation studies is carried out to develop the framework for face mask recognition using pre-trained deep convolution neural networks (DCNN) models used in conjunction with a fast single layer feed-forward neural network (SLFNN) commonly known as Extreme Learning Machine (ELM) as classification technique. The ELM is well known for its real time data processing capabilities and has been successfully applied both for regression and classification problems of image processing and biomedical domain. It is for the first time that in this paper we have proposed the use of ELM as classifier for face mask detection. As a precursor to this, for feature selection, six pre-trained DCNNs such as Xception, Vgg16, Vgg19, ResNet50, ResNet 101 and ResNet152 are tested for this purpose. The best testing accuracy is obtained in case of ResNet152 transfer learning model used with ELM as the classifier. The performance evaluation through different ablation studies on testing accuracy explicitly proves that ResNet152 - ELM hybrid architecture is not only the best among the selected transfer learning models but also proves so when it is compared with several other classifiers used for the face mask detection operation. Through this investigation, novelty of the use of ResNet152 + ELM for face mask detection framework in real time domain is established. © 2022

Clinical Profile and Treatment Outcome of Severe Aplastic Anemia in Adults: An Experience from Tertiary Care Centre in North India

Introduction: Untreated/refractory severe aplastic anemia (SAA) is associated with very high mortality. Allogenic bone marrow transplantation or immunosuppressive therapy remains mainstay of treatment but these treatments are timely available to only a select subset of patients. Recently eltrombopag has been approved for treatment of SAA. Aims & Objectives: We aimed to describe clinical profile and treatment response in patients with SAA from a tertiary care centre. Material(s) and Method(s): A retrospective analysis of patients diagnosed with SAA over a period of 7 years from January 2015-December 2021 was performed. The details of demographic profile, laboratory features, treatment given and response were analyzed. Result(s): Ninety patients were diagnosed with SAA during this period out of which 18 patients went elsewhere for treatment. Seventy-two patients who received treatment in our hospital were included in the analysis. Sixty-two patients were SAA while 10 VSAA. PNH screening was done in 24 patients, out of which 17 (70%) had small clone. The details of treatment and response achieved is shown in Table 1. Eight patients (11.1%) received matched related donor allogenic hemopoietic cell transplant, out of which one had rejection followed by auto recovery while one died 6 months later due to covid 19 disease. Sixty-four patients received immunosuppressive therapy, forty-nine (76%) responded. Recurrence of SAA occurred in two patients who has achieved complete response to ATG therapy;one received second course of horse ATG + CSA + ETP and responded again. Conclusion(s): Timely diagnosis and appropriate treatment selection is of utmost importance to achieve optimal outcome in severe aplastic anemia. Eltrombopag has become an important addition not only in front line but also in relapsed refractory aplastic anemia. Patients lacking donor, or resources for ATG should be treated with cyclosporine and eltrombopag as early as possible. (Table Presented).

A Web-based Blood Bank System for Managing Records of Donors and Receipts

The Online Blood Donation Management System, the purpose of which is to act as a bridge between a person who needs blood, a patient, and a blood donor. The design of an automatic blood system has become an integral part for saving the human lives, who need the blood under different situations. Since, there are various drawbacks of the pre-existing system like privacy issues for the donors, which are getting reflected directly on the interface. Thus, we have designed a robust system that will create a connection between different hospitals, NGOs, and blood banks to help the patient in any difficult situation. Thus, HIPPA model provides a backbone for security breaches The interface designed will be easy-to-use and easy to access and will be a fast, efficient, and reliable way to get lifesaving blood, totally free of charge. Apart from this the visualization of the data is present along with the one extra COVID module, which will help covid and normal patients for plasma donation. The main aim of the paper is to reduce the complications of finding a blood donor during panic situations and provide a high level of security for the donors. © 2022 IEEE.

Hybrid CNN-SVM Model for Face Mask Detector to Protect from COVID-19

The 2019 coronavirus outbreak (COVID-19) has had a huge impact on humanity. By May 2021, nearly, 172 million people worldwide were affected by the infectious spread of COVID-19. While the distribution of vaccines has already begun, mass distribution around the world has yet to take place. According to the World Health Organization (WHO), wearing a face mask can significantly reduce the spread of the COVID-19 virus. However, even improper wearing of face mask can prevent the purposes and lead to the spread of the virus. Under the influence of public health and the global economy, an effective Covid-19 pandemic strategy requires a lot of attention of humanity. To prevent the spread of such deadly virus, intelligent techniques are required. In the proposed work, an intelligent face mask detector framework is proposed based on deep learning concept which can classify the person who wear mask from those who are not wearing mask. In the proposed work, a hybrid model of convolution neural network with support vector machine is used for designing the mask detector. The performance of the proposed method is evaluated on real-world masked face recognition dataset (RMFD) and medical mask dataset (MDD). When implemented, it has been found that the proposed method can achieve high accuracy (99.11%). The excellent performance of the proposed model is very suitable for video surveillance equipment also. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Depression Detection from Social Site using Machine Learning and Deep Learning

As COVID-19 crisis is settling down in countries, whether or not a person has been affected personally by the disease, he fights with issues such as anxiety, panic attacks, grief, low mood, and many other psychotic disorders. Mental fitness is one of the major strengths in the development of the individual. Development of social sites turns out to be one platform where the person feels free to vent out their thoughts and to easily interact with people. Extracting useful information from those posts is a part of sentimental analysis, which is the technique of machine learning that helps to know the mental condition of the individual. In this paper, various machine learning algorithms such as random forest, Naive Bayes, decision tree, multilayer perceptron, maximum entropy, KNN, gradient boosted decision tree, adaptive boosting, bagged logistic regression, tree ensemble model, Liblinear, convolutional neural network, and long short-term memory are applied on the dataset, and different mathematical scales such as accuracy, precision, recall, and F1 score concluded that bagged logistic regression has given the better accuracy results. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Concurrent Presentation of Acute Myeloid Leukemiawith COVID 19: A Diagnostic Challenge

Introduction: COVID-19 disease caused due to infection with thenovel coronavirus (SARS CoV-2) is causing havoc worldwide eversince its first appearance in Wuhan, China in December 2019.Although the awareness regarding the disease has increased manifoldsince its inception in December 2019, the treatment of patients withalready suppressed immune status owing to hematological malignancies and other comorbidities is a challenge. Patients withleukemias and COVID-19 show a higher fatality rate of 37%.Thetime-lapse between disease presentation and diagnosis is alsoincreased due to COVID-19 as many symptoms are overshadoweddue to the typical presentation of COVID-19.Aims &Objectives: The aim is to diagnose the patients with leukemias at the very onset of disease during this pandemic as theoutcome is very poor so that the management and follow-up of thepatients begin at the earliest.Materials &Methods: We are presenting case series of four patientsof COVID-19 with coexistent AML. The diagnosis was establishedwith the correlation of flow cytometry, bone marrow examination,and radiology.Interestingly all the patients diagnosed with AML were not havingany clinical abnormality before the COVID-19 infection so we feelthat this is a de-novo presentation of AML with COVID-19 and canonly speculate about the coronavirus per se causing the hematologicalmalignancy. However, more data is required to confirm this association, particularly the role of IL-6 and its direct effect in the causationof hematological malignancy.Result: One of the patients was in the intensive care unit for a weekhowever patient's condition deteriorated and he could not survive.Rest of them are under follow-up.Conclusions: Concurrent presentation of AML and COVID-19 ischallenging from a diagnostic and management point of view. Therehave been few case reports in patients with leukemia getting infectedwith COVID-19 but to the best of our knowledge, these are possiblythe first few cases of AML occurring in concurrence with COVID-19infection detected incidentally without any prior history of any significant disease. These cases highlight the association ofhematological malignancies with the current pandemic and also difficulty in their management.

Should COVID-19 patients >75 years be Ventilated? An Outcome Study.

BACKGROUND: Elderly patients with COVID-19 disease are at increased risk for adverse outcomes. Current data regarding disease characteristics and outcomes in this population are limited. AIM: To delineate the adverse factors associated with outcomes of COVID-19 patients ≥75 years of age. DESIGN: Retrospective cohort study. METHODS: Patients were classified into mild/moderate, severe/very severe and critical disease (intubated) based on oxygen requirements. The primary outcome was in-hospital mortality. RESULTS: A total of 355 patients aged ≥75 years hospitalized with COVID-19 between 19 March and 25 April 2020 were included.Mean age was 84.3 years. One-third of the patients developed critical disease. Mean length of stay was 7.10 days. Vasopressors were required in 27%, with the highest frequency in the critical disease group (74.1%). Overall mortality was 57.2%, with a significant difference between severity groups (mild/moderate disease: 17.4%, severe/very severe disease: 71.3%, critical disease: 94.9%, P < 0.001).Increased age, dementia, and severe/very severe and critical disease groups were independently associated with increased odds for mortality while diarrhea was associated with decreased odds for mortality (OR: 0.12, 95% CI: 0.02-0.60, P < 0.05). None of the cardiovascular comorbidities were significantly associated with mortality. CONCLUSION: Age and dementia are associated with increased odds for mortality in patients ≥75 years of age hospitalized with COVID-19. Those who require intubation have the greatest odds for mortality. Diarrhea as a presenting symptom was associated with lower odds for mortality.

Outcomes in patients with severe COVID-19 disease treated with tocilizumab: a case-controlled study.

BACKGROUND: COVID-19 is an ongoing threat to society. Patients who develop the most severe forms of the disease have high mortality. The interleukin-6 inhibitor tocilizumab has the potential to improve outcomes in these patients by preventing the development of cytokine release storm. AIMS: To evaluate the outcomes of patients with severe COVID-19 disease treated with the interleukin-6 inhibitor tocilizumab. METHODS: We conducted a retrospective, case-control, single-center study in patients with severe to critical COVID-19 disease treated with tocilizumab. Disease severity was defined based on the amount of oxygen supplementation required. The primary endpoint was the overall mortality. Secondary endpoints were mortality in non-intubated patients and mortality in intubated patients. RESULTS: A total of 193 patients were included in the study. Ninety-six patients received tocilizumab, while 97 served as the control group. The mean age was 60 years. Patients over 65 years represented 43% of the population. More patients in the tocilizumab group reported fever, cough and shortness of breath (83%, 80% and 96% vs. 73%, 69% and 71%, respectively). There was a non-statistically significant lower mortality in the treatment group (52% vs. 62.1%, P = 0.09). When excluding intubated patients, there was statistically significant lower mortality in patients treated with tocilizumab (6% vs. 27%, P = 0.024). Bacteremia was more common in the control group (24% vs. 13%, P = 0.43), while fungemia was similar for both (3% vs. 4%, P = 0.72). CONCLUSION: Our study showed a non-statistically significant lower mortality in patients with severe to critical COVID-19 disease who received tocilizumab. When intubated patients were excluded, the use of tocilizumab was associated with lower mortality.