Efficacy and safety of luspatercept versus epoetin alfa in erythropoiesis-stimulating agent-naive, transfusion-dependent, lower-risk myelodysplastic syndromes (COMMANDS): interim analysis of a phase 3, open-label, randomised controlled trial.

BACKGROUND: Erythropoiesis-stimulating agents (ESAs) are the standard-of-care treatment for anaemia in most patients with lower-risk myelodysplastic syndromes but responses are limited and transient. Luspatercept promotes late-stage erythroid maturation and has shown durable clinical efficacy in patients with lower-risk myelodysplastic syndromes. In this study, we report the results of a prespecified interim analysis of luspatercept versus epoetin alfa for the treatment of anaemia due to lower-risk myelodysplastic syndromes in the phase 3 COMMANDS trial. METHODS: The phase 3, open-label, randomised controlled COMMANDS trial is being conducted at 142 sites in 26 countries. Eligible patients were aged 18 years or older, had a diagnosis of myelodysplastic syndromes of very low risk, low risk, or intermediate risk (per the Revised International Prognostic Scoring System), were ESA-naive, and required red blood cell transfusions (2-6 packed red blood cell units per 8 weeks for ≥8 weeks immediately before randomisation). Integrated response technology was used to randomly assign patients (1:1, block size 4) to luspatercept or epoetin alfa, stratified by baseline red blood cell transfusion burden (<4 units per 8 weeks vs ≥4 units per 8 weeks), endogenous serum erythropoietin concentration (≤200 U/L vs >200 to <500 U/L), and ring sideroblast status (positive vs negative). Luspatercept was administered subcutaneously once every 3 weeks starting at 1·0 mg/kg body weight with possible titration up to 1·75 mg/kg. Epoetin alfa was administered subcutaneously once a week starting at 450 IU/kg body weight with possible titration up to 1050 IU/kg (maximum permitted total dose of 80 000 IU). The primary endpoint was red blood cell transfusion independence for at least 12 weeks with a concurrent mean haemoglobin increase of at least 1·5 g/dL (weeks 1-24), assessed in the intention-to-treat population. Safety was assessed in patients who received at least one dose of study treatment. The COMMANDS trial was registered with ClinicalTrials.gov, NCT03682536 (active, not recruiting). FINDINGS: Between Jan 2, 2019 and Aug 31, 2022, 356 patients were randomly assigned to receive luspatercept (178 patients) or epoetin alfa (178 patients), comprising 198 (56%) men and 158 (44%) women (median age 74 years [IQR 69-80]). The interim efficacy analysis was done for 301 patients (147 in the luspatercept group and 154 in the epoetin alfa group) who completed 24 weeks of treatment or discontinued earlier. 86 (59%) of 147 patients in the luspatercept group and 48 (31%) of 154 patients in the epoetin alfa group reached the primary endpoint (common risk difference on response rate 26·6; 95% CI 15·8-37·4; p<0·0001). Median treatment exposure was longer for patients receiving luspatercept (42 weeks [IQR 20-73]) versus epoetin alfa (27 weeks [19-55]). The most frequently reported grade 3 or 4 treatment-emergent adverse events with luspatercept (≥3% patients) were hypertension, anaemia, dyspnoea, neutropenia, thrombocytopenia, pneumonia, COVID-19, myelodysplastic syndromes, and syncope; and with epoetin alfa were anaemia, pneumonia, neutropenia, hypertension, iron overload, COVID-19 pneumonia, and myelodysplastic syndromes. The most common suspected treatment-related adverse events in the luspatercept group (≥3% patients, with the most common event occurring in 5% patients) were fatigue, asthenia, nausea, dyspnoea, hypertension, and headache; and none (≥3% patients) in the epoetin alfa group. One death after diagnosis of acute myeloid leukaemia was considered to be related to luspatercept treatment (44 days on treatment). INTERPRETATION: In this interim analysis, luspatercept improved the rate at which red blood cell transfusion independence and increased haemoglobin were achieved compared with epoetin alfa in ESA-naive patients with lower-risk myelodysplastic syndromes. Long-term follow-up and additional data will be needed to confirm these results and further refine findings in other subgroups of patients with lower-risk myelodysplastic syndromes, including non-mutated SF3B1 or ring sideroblast-negative subgroups. FUNDING: Celgene and Acceleron Pharma.

Lessons Learned from POCUS Instruction in Undergraduate Medicine During the COVID-19 Pandemic.

Point of care Ultrasound (POCUS) has been adopted into clinical practice across many fields of medicine. Undergraduate medical education programs have recognized the need to incorporate POCUS training into their curricula, traditionally done in small groups with in-person sessions. This method is resource intensive and requires sufficient equipment and expertise. These requirements are often cited as barriers for implementation. During the Coronavirus Disease 2019 (COVID-19) pandemic, POCUS education was required to adapt to physical distancing regulations, giving rise to novel teaching methods for POCUS. This article outlines the implementation of a POCUS teaching session before and during the pandemic. It describes how these innovations can scale POCUS teaching and overcome barriers moving forward. A flipped classroom model was implemented for all learners. Learners were given an introductory POCUS module before the scheduled in-person or virtual teaching session. Sixty-nine learners participated in conventional in-person teaching, while twenty-two learners participated in virtual teaching following the pandemic-related restrictions. Learners completed a written test before and following the teaching. In-person learners were assessed using an objective structured assessment of ultrasound skills (OSAUS) pre- and post-learning sessions. A follow-up survey was conducted three years after the teaching sessions were completed. Both in-person and virtual groups demonstrated statistically significant improvement in knowledge scores (p <0.0001). Both groups had similar post-test learning scores (74.2 ± 13.6% vs. 71.8 ± 14.5 %, respectively). On follow-up questionnaires, respondents indicate that they found our online and in-person modes of teaching helpful during their residency. POCUS education continues to face a variety of barriers, including limitations in infrastructure and expertise. This study describes an adapted POCUS teaching model that is scalable, uses minimal infrastructure and retains the interactivity of conventional small-group POCUS teaching. This program can serve as a blueprint for other institutions offering POCUS teaching, especially when conventional teaching methods are limited.

Islamophobic Discourses in India During COVID-19 Pandemic: A Case of Tablighi Jamaat

Islamophobia is a hot issue in the world nowadays. This study aimed to analyse the discourses which were produced against Muslims and Islam during the novel coronavirus outbreak in India, where the Muslim minority group Tablighi Jamaat was targeted and held responsible for spreading the virus. The study employed Critical Discourse Analysis to analyse selected tweets from Indian politicians, government dignitaries, and common Indian users to find out Islamophobic themes and ideological structures in their discourses. The researcher hypothesizes that Islamophobia is an integral feature of Indian political communication which is also obvious in their discursive and rhetorical devices. The researcher purposively selected 50 of the most controversial tweets by Indian politicians, government officials, and common Indian Twitter users against Muslims and Islam and carried out a critical discourse analysis through political discourse strategies by Wodak's (2014). The findings of the study reveal that Indians use discriminatory language against Muslims and Islam to create a difference of "us versus them” and by using anti-Islamic strategies and Islamophobic remarks against the Muslims of India and propagating a common belief in Indians that Muslims and Islam are the main culprit of the COVID-19 outbreak in India. © 2023 World Communication Association.

ASSESSMENT WITHOUT BORDERS: Modernising Placement Tests for Diverse Contexts

In this chapter, we propose an innovative, holistic, and inclusive approach to assessment practices with a focus on placement exams and the need triggered by COVID-19 for alignment, adaptability, and flexibility to ensure not only proper placement but long-term retention and success in language programmes. This chapter describes our approach and the cross-platform placement exam application that we built for this purpose. We discuss the importance of aligning assessment with our proficiency-based programme goals and department culture to ensure that all students and their backgrounds (traditional learners, heritage learners, among others) are considered when developing placement exams. We highlight the value of utilising existing resources to address gaps in the field of proficiency-based assessment for diverse learners. We also highlight the importance of innovation in assessment without borders especially considering the COVID-19 pandemic and its impact on forcing a shift in the delivery of assessment via computer-assisted technology. © 2023 selection and editorial matter, Karim Sadeghi;individual chapters, the contributors.

Challenges and opportunities in online education in Architecture: Lessons learned for Post-Pandemic education

This paper reports on the online architectural learning experience at a university in the Kingdom of Saudi Arabia. The study surveyed students and faculty members regarding their experience of learning and assessment and proposed recommendations for future online education. The survey supported the study hypothesis that students generally believe that they had a positive online learning experience during the pandemic. However, the principal challenge was the lack of a physical interactive environment in studio-based courses, which resulted in a sense of isolation. Thus, the study recommends that online education in architecture could be used as a supportive tool rather than an alternative. In this context, design studio courses, especially at the beginning of a study programme, should not be given completely online unless this was essential. The use of blended learning is a balanced strategy that uses interactive online technologies to enhance traditional face-to-face learning in design studio courses.

COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans.

BACKGROUND: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the "COVLIAS 2.0-cXAI" system using four kinds of class activation maps (CAM) models. METHODOLOGY: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. RESULTS: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. CONCLUSIONS: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.

Primary health care centers Physicians' satisfaction toward COVID-19 management in Asser Region, Saudi Arabia

Background. The presence of extensive workload and pressure associated with COVID19 has resulted in a lot of mental and physical trauma in primary health care physicians (PHCC) across the world. Aim of the study. The present study is aimed at understanding the level of dissatisfaction in PHCC physicians in the Asser region of Saudi Arabia. The various risk factors associated with dissatisfaction also have been analyzed in this study. Method. An analytical cross-sectional study was performed on various PHCCs who were working in the Ministry of Health in the Asser region. Results. The overall analysis showed that around 73% of the physicians who responded showed dissatisfaction. The major factors that were found to affect satisfaction based on the variables analyzed include physicians who are males, Saudi Nationals, training residents, and those who received recognition. These four variables had a p value of less than 0.0001 making it statistically significant. Conclusion. The study observed that there is dissatisfaction amongst PHCC physicians towards the management of pandemics. The factors identified should help officials in order to address these issues [ FROM AUTHOR] Copyright of Cardiometry is the property of Cardiometry and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)

Vascular Implications of COVID-19: Role of Radiological Imaging, Artificial Intelligence, and Tissue Characterization: A Special Report.

The SARS-CoV-2 virus has caused a pandemic, infecting nearly 80 million people worldwide, with mortality exceeding six million. The average survival span is just 14 days from the time the symptoms become aggressive. The present study delineates the deep-driven vascular damage in the pulmonary, renal, coronary, and carotid vessels due to SARS-CoV-2. This special report addresses an important gap in the literature in understanding (i) the pathophysiology of vascular damage and the role of medical imaging in the visualization of the damage caused by SARS-CoV-2, and (ii) further understanding the severity of COVID-19 using artificial intelligence (AI)-based tissue characterization (TC). PRISMA was used to select 296 studies for AI-based TC. Radiological imaging techniques such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound were selected for imaging of the vasculature infected by COVID-19. Four kinds of hypotheses are presented for showing the vascular damage in radiological images due to COVID-19. Three kinds of AI models, namely, machine learning, deep learning, and transfer learning, are used for TC. Further, the study presents recommendations for improving AI-based architectures for vascular studies. We conclude that the process of vascular damage due to COVID-19 has similarities across vessel types, even though it results in multi-organ dysfunction. Although the mortality rate is ~2% of those infected, the long-term effect of COVID-19 needs monitoring to avoid deaths. AI seems to be penetrating the health care industry at warp speed, and we expect to see an emerging role in patient care, reduce the mortality and morbidity rate.

A comprehensive genomic study, mutation screening, phylogenetic and statistical analysis of SARS-CoV-2 and its variant omicron among different countries.

BACKGROUND: With the rapid development of the genomic sequence data for the Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and its variants Delta (B.1.617.2) and Omicron (B.1.1.529), it is vital to successfully identify mutations within the genome. OBJECTIVE: The main objective of the study is to investigate the full-length genome mutation analysis of 157 SARS-CoV-2 and its variant Delta and Omicron isolates. This study also provides possible effects at the structural level to understand the role of mutations and new insights into the evolution of COVID-19 and evaluates the differential level analysis in viral genome sequence among different nations. We have also tried to offer a mutation snapshot for these differences that could help in vaccine formulation. This study utilizes a unique and efficient method of targeting the stable genes for the drug discovery approach. METHODS: Complete genome sequence information of SARS-CoV-2, Delta, and Omicron from online resources were used to predict structure domain identification, data mining, and screening; employing different bioinformatics tools. BioEdit software was used to perform their genomic alignments across countries and a phylogenetic tree as per the confidence of 500 bootstrapping values was constructed. Heterozygosity ratios were determined in-silico. A minimum spanning network (MSN) of selected populations was determined by Bruvo's distance role-based framework. RESULTS: Out of all 157 different strains of SARS-CoV-2 and its variants, and their complete genome sequences from different countries, Corona nucleoca and DUF5515 were observed to be the most conserved domains. All genomes obtained changes in comparison to the Wuhan-Hu-1 strain, mainly in the TRS region (CUAAAC or ACGAAC). We discovered 596 mutations in all genes, with the highest number (321) found in ORF1ab (QHD43415.1), or TRS site mutations found only in ORF7a (1) and ORF10 (2). The Omicron variant has 30 mutations in the Spike protein and has a higher alpha-helix shape (23.46%) than the Delta version (22.03%). T478 was also discovered to be a prevalent polymorphism in Delta and Omicron variations, as well as genomic gaps ranging from 45 to 65aa. All 157 sequences contained variations and conformed to Nei's Genetic distance. We discovered heterozygosity (Hs) 0.01, mean anticipated Hs 0.32, the genetic diversity index (GDI) 0.01943989, and GD within population 0.01266951. The Hedrick value was 0.52324978, the GD coefficient was 0.52324978, the average Hs was 0.01371452, and the GD coefficient was 0.52324978. Among other countries, Brazil has the highest standard error (SE) rate (1.398), whereas Japan has the highest ratio of Nei's gene diversity (0.01). CONCLUSIONS: The study's findings will assist in comprehending the shape and kind of complete genome, their streaming genomic sequences, and mutations in various additions of SARS-CoV-2, as well as its different variant strains like Omicron. These results will provide a scientific basis to design the vaccines and understand the genomic study of these viruses.

Health Literacy and Protective Measures Among Older Adults during COVID-19 Pandemic

Study’s purpose: to assess health literacy, protective measures among older adults and the association between them. Methodology: a descriptive study conducted in a rural area at Sharkia Governorate, Shobera el nakhla with a random sample composed of 300 older adults. Health literacy was measured by the European health literacy survey questionnaire (HLS-EU-Q) and protective measures were measured by a tool developed by the researchers. Association between health literacy and protective measures was identified through Pearson’s correlation coefficient. Major results: 44.3% and 24.3% of older adults had inadequate and problematic health literacy, respectively and 58.0% of them had unsatisfactory protective measures. Highly Statistically significant positive correlation was found between health literacy and protective measures p < 0.01. Clinical implications: Better health literacy means recognizing how to protect oneself and others through basic measures. We should invest in strategies to increase health literacy so that older adults’ practices improve and coronavirus transmission is reduced.

Ferric Oxide Colloid: Towards Green Nano-Fertilizer for Tomato Plant with Enhanced Vegetative Growth and Immune Response Against Fusarium Wilt Disease.

Global food crisis due to climate change, pandemic COVID-19 outbreak, and Russia-Ukraine conflict leads to catastrophic consequences; almost 10 percent of the world's population go to bed hungry daily. Narrative solution for green agriculture with high vegetation and crop yield is mandatory; novel nanomaterials can improve plant immunity and restrain plant diseases. Iron is fundamental nutrient element; it plays vital role in enzyme activity and RNA synthesis; furthermore it is involved in photosynthesis electron-transfer chains. This study reports on the facile synthesis of colloidal ferric oxide nanoparticles as novel nano-fertilizer to promote vegetation and to suppress Fusarium wilt disease in tomato plant. Disease index, protection percent, photosynthetic pigments, and metabolic indicators of resistance in plant as response to induction of systemic resistance (SR) were recorded. Results illustrated that Fe2O3 NPs had antifungal activity against F. oxysporum. Fe2O3 NPs (at 20 µg/mL) was the best treatment and reduced percent disease indexes by 15.62 and gave highly protection against disease by 82.15% relative to untreated infected plants. Fe2O3 NPs treatments in either (non-infected or infected) plants showed improvements in photosynthetic pigments, osmolytes, and antioxidant enzymes activity. The beneficial effects of the synthesized Fe2O3 NPs were extended to increase not only photosynthetic pigments, osmolytes contents but also the activities of peroxidase (POD), polyphenol oxidase (PPO), catalase (CAT) and superoxide dismutase (SOD), enzymes of the healthy and infected tomato plants in comparison with control. For, peroxidase and polyphenol oxidase activities it was found that, application of Fe2O3 NPs (10 µg/mL) on challenged plants offered the best treatments which increased the activities of POD by (34.4%) and PPO by (31.24%). On the other hand, application of Fe2O3 NPs (20 µg/mL) on challenged plants offered the best treatments which increased the activities of CAT by (30.9%), and SOD by (31.33%). Supplementary Information: The online version contains supplementary material available at 10.1007/s10904-022-02442-6.

COVLIAS 1.0Lesion vs. MedSeg: An Artificial Intelligence Framework for Automated Lesion Segmentation in COVID-19 Lung Computed Tomography Scans.

BACKGROUND: COVID-19 is a disease with multiple variants, and is quickly spreading throughout the world. It is crucial to identify patients who are suspected of having COVID-19 early, because the vaccine is not readily available in certain parts of the world. METHODOLOGY: Lung computed tomography (CT) imaging can be used to diagnose COVID-19 as an alternative to the RT-PCR test in some cases. The occurrence of ground-glass opacities in the lung region is a characteristic of COVID-19 in chest CT scans, and these are daunting to locate and segment manually. The proposed study consists of a combination of solo deep learning (DL) and hybrid DL (HDL) models to tackle the lesion location and segmentation more quickly. One DL and four HDL models-namely, PSPNet, VGG-SegNet, ResNet-SegNet, VGG-UNet, and ResNet-UNet-were trained by an expert radiologist. The training scheme adopted a fivefold cross-validation strategy on a cohort of 3000 images selected from a set of 40 COVID-19-positive individuals. RESULTS: The proposed variability study uses tracings from two trained radiologists as part of the validation. Five artificial intelligence (AI) models were benchmarked against MedSeg. The best AI model, ResNet-UNet, was superior to MedSeg by 9% and 15% for Dice and Jaccard, respectively, when compared against MD 1, and by 4% and 8%, respectively, when compared against MD 2. Statistical tests-namely, the Mann-Whitney test, paired t-test, and Wilcoxon test-demonstrated its stability and reliability, with p < 0.0001. The online system for each slice was <1 s. CONCLUSIONS: The AI models reliably located and segmented COVID-19 lesions in CT scans. The COVLIAS 1.0Lesion lesion locator passed the intervariability test.

Cytokine Profiling among Children with Multisystem Inflammatory Syndrome versus Simple COVID-19 Infection: A Study from Northwest Saudi Arabia.

BACKGROUND: Multisystem Inflammatory Syndrome in Children (MIS-C) is a novel syndrome associated with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection with varying clinical features. This study aimed to analyze the expression profiles of cytokines in blood, report the important clinical characteristics, and correlate these with the short- and mid-term outcomes. METHODS: This cross-sectional study was conducted on hospitalized children with MIS-C from March 2021 to May 2022. Phenotypes were classified into two groups (A,B) according to the severity of the disease and the need for invasive respiratory support. Clinical features, laboratory parameters, and outcomes were reported. RESULTS: We identified 60 children with MIS-C (mean age of 7.4 ± 3.8 years) compared to 30 age- and sex-matched controls with simple COVID-19. The clinical manifestations of MIS-C patients were fever (100%), respiratory (83.3%), GIT (80%), and conjunctivitis (80%). Twenty-seven MIS-C children (45%) required PICU admission due to shock and needed mechanical ventilation. Anemia, lymphopenia, and elevated levels of inflammatory and tissue injury markers were observed in the MIS-C groups (mainly B). High cytokine levels (IL-1ß, IL-6, IFN-α, GM-CSF, and HMGB1) were observed acutely in the MIS-C children, and a persistent elevation of some cytokines were reported at midterm follow-up, especially in Group B. CONCLUSION: Robust inflammatory response to COVID-19 disease with elevated IL-1ß, IL-6, and GM-CSF levels might explain the severity and outcome of the clinical syndrome.

COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans

Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the 'COVLIAS 2.0-cXAI';system using four kinds of class activation maps (CAM) models. Methodology: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. Results: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. Conclusions: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.

COVLIAS 1.0Lesion vs. MedSeg: An Artificial Intelligence Framework for Automated Lesion Segmentation in COVID-19 Lung Computed Tomography Scans

Background: COVID-19 is a disease with multiple variants, and is quickly spreading throughout the world. It is crucial to identify patients who are suspected of having COVID-19 early, because the vaccine is not readily available in certain parts of the world. Methodology: Lung computed tomography (CT) imaging can be used to diagnose COVID-19 as an alternative to the RT-PCR test in some cases. The occurrence of ground-glass opacities in the lung region is a characteristic of COVID-19 in chest CT scans, and these are daunting to locate and segment manually. The proposed study consists of a combination of solo deep learning (DL) and hybrid DL (HDL) models to tackle the lesion location and segmentation more quickly. One DL and four HDL models-namely, PSPNet, VGG-SegNet, ResNet-SegNet, VGG-UNet, and ResNet-UNet-were trained by an expert radiologist. The training scheme adopted a fivefold cross-validation strategy on a cohort of 3000 images selected from a set of 40 COVID-19-positive individuals. Results: The proposed variability study uses tracings from two trained radiologists as part of the validation. Five artificial intelligence (AI) models were benchmarked against MedSeg. The best AI model, ResNet-UNet, was superior to MedSeg by 9% and 15% for Dice and Jaccard, respectively, when compared against MD 1, and by 4% and 8%, respectively, when compared against MD 2. Statistical tests-namely, the Mann–Whitney test, paired t-test, and Wilcoxon test-demonstrated its stability and reliability, with p < 0.0001. The online system for each slice was <1 s. Conclusions: The AI models reliably located and segmented COVID-19 lesions in CT scans. The COVLIAS 1.0Lesion lesion locator passed the intervariability test.

Cardiovascular Risk Stratification in Diabetic Retinopathy via Atherosclerotic Pathway in COVID-19/Non-COVID-19 Frameworks Using Artificial Intelligence Paradigm: A Narrative Review.

Diabetes is one of the main causes of the rising cases of blindness in adults. This microvascular complication of diabetes is termed diabetic retinopathy (DR) and is associated with an expanding risk of cardiovascular events in diabetes patients. DR, in its various forms, is seen to be a powerful indicator of atherosclerosis. Further, the macrovascular complication of diabetes leads to coronary artery disease (CAD). Thus, the timely identification of cardiovascular disease (CVD) complications in DR patients is of utmost importance. Since CAD risk assessment is expensive for low-income countries, it is important to look for surrogate biomarkers for risk stratification of CVD in DR patients. Due to the common genetic makeup between the coronary and carotid arteries, low-cost, high-resolution imaging such as carotid B-mode ultrasound (US) can be used for arterial tissue characterization and risk stratification in DR patients. The advent of artificial intelligence (AI) techniques has facilitated the handling of large cohorts in a big data framework to identify atherosclerotic plaque features in arterial ultrasound. This enables timely CVD risk assessment and risk stratification of patients with DR. Thus, this review focuses on understanding the pathophysiology of DR, retinal and CAD imaging, the role of surrogate markers for CVD, and finally, the CVD risk stratification of DR patients. The review shows a step-by-step cyclic activity of how diabetes and atherosclerotic disease cause DR, leading to the worsening of CVD. We propose a solution to how AI can help in the identification of CVD risk. Lastly, we analyze the role of DR/CVD in the COVID-19 framework.

Eight pruning deep learning models for low storage and high-speed COVID-19 computed tomography lung segmentation and heatmap-based lesion localization: A multicenter study using COVLIAS 2.0.

BACKGROUND: COVLIAS 1.0: an automated lung segmentation was designed for COVID-19 diagnosis. It has issues related to storage space and speed. This study shows that COVLIAS 2.0 uses pruned AI (PAI) networks for improving both storage and speed, wiliest high performance on lung segmentation and lesion localization. METHOD: ology: The proposed study uses multicenter ∼9,000 CT slices from two different nations, namely, CroMed from Croatia (80 patients, experimental data), and NovMed from Italy (72 patients, validation data). We hypothesize that by using pruning and evolutionary optimization algorithms, the size of the AI models can be reduced significantly, ensuring optimal performance. Eight different pruning techniques (i) differential evolution (DE), (ii) genetic algorithm (GA), (iii) particle swarm optimization algorithm (PSO), and (iv) whale optimization algorithm (WO) in two deep learning frameworks (i) Fully connected network (FCN) and (ii) SegNet were designed. COVLIAS 2.0 was validated using "Unseen NovMed" and benchmarked against MedSeg. Statistical tests for stability and reliability were also conducted. RESULTS: Pruning algorithms (i) FCN-DE, (ii) FCN-GA, (iii) FCN-PSO, and (iv) FCN-WO showed improvement in storage by 92.4%, 95.3%, 98.7%, and 99.8% respectively when compared against solo FCN, and (v) SegNet-DE, (vi) SegNet-GA, (vii) SegNet-PSO, and (viii) SegNet-WO showed improvement by 97.1%, 97.9%, 98.8%, and 99.2% respectively when compared against solo SegNet. AUC > 0.94 (p < 0.0001) on CroMed and > 0.86 (p < 0.0001) on NovMed data set for all eight EA model. PAI <0.25 s per image. DenseNet-121-based Grad-CAM heatmaps showed validation on glass ground opacity lesions. CONCLUSIONS: Eight PAI networks that were successfully validated are five times faster, storage efficient, and could be used in clinical settings.

A Powerful Paradigm for Cardiovascular Risk Stratification Using Multiclass, Multi-Label, and Ensemble-Based Machine Learning Paradigms: A Narrative Review.

Background and Motivation: Cardiovascular disease (CVD) causes the highest mortality globally. With escalating healthcare costs, early non-invasive CVD risk assessment is vital. Conventional methods have shown poor performance compared to more recent and fast-evolving Artificial Intelligence (AI) methods. The proposed study reviews the three most recent paradigms for CVD risk assessment, namely multiclass, multi-label, and ensemble-based methods in (i) office-based and (ii) stress-test laboratories. Methods: A total of 265 CVD-based studies were selected using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) model. Due to its popularity and recent development, the study analyzed the above three paradigms using machine learning (ML) frameworks. We review comprehensively these three methods using attributes, such as architecture, applications, pro-and-cons, scientific validation, clinical evaluation, and AI risk-of-bias (RoB) in the CVD framework. These ML techniques were then extended under mobile and cloud-based infrastructure. Findings: Most popular biomarkers used were office-based, laboratory-based, image-based phenotypes, and medication usage. Surrogate carotid scanning for coronary artery risk prediction had shown promising results. Ground truth (GT) selection for AI-based training along with scientific and clinical validation is very important for CVD stratification to avoid RoB. It was observed that the most popular classification paradigm is multiclass followed by the ensemble, and multi-label. The use of deep learning techniques in CVD risk stratification is in a very early stage of development. Mobile and cloud-based AI technologies are more likely to be the future. Conclusions: AI-based methods for CVD risk assessment are most promising and successful. Choice of GT is most vital in AI-based models to prevent the RoB. The amalgamation of image-based strategies with conventional risk factors provides the highest stability when using the three CVD paradigms in non-cloud and cloud-based frameworks.