Association of food habit with the COVID-19 severity and hospitalization: A cross-sectional study among the recovered individuals in Bangladesh.

Background: It was assumed that dietary habits might influence the status of COVID-19 patients. Aim: We aimed at the identification of association of dietary habits with the COVID-19 severity and hospitalization. Methods: It was a retrospective cross-sectional study (n = 1025). We used bivariate and multivariate analyses to correlate the association between self-reported dietary patterns and COVID-19 severity and hospitalization. Results: Dietary habits (black tea, milked tea, pickles, black caraway seeds, honey, fish, fruits, vegetables, garlic, onion and turmeric) were identified with lower risk of COVID-19 severity and hospitalization. Interestingly, the consumption frequency (one-, two- or three-times/day) of rice - the staple food in Bangladesh - was not associated with COVID-19 severity and hospitalization for comorbid patients. In contrast, a moderate rice-eating habit (two times/day) was strongly associated with the lower risk of severity and hospitalization for non-comorbid patients. However, for both comorbid and non-comorbid patients, consumption of black tea, milked tea, pickles and honey were associated with a lower likelihood of severity and hospitalization. Overall, a high consumption (three-times/day) of fish, fruits and vegetables, a moderate consumption of garlic, onion and turmeric spices and a daily intake of black/milked tea, and honey were associated with reduced risk of COVID-19 severity and hospitalization. Conclusions: To reduce the severity of COVID-19, a habitual practice of intaking black tea, milked tea, black caraway seeds and honey along with dietary habit (rice, fish and vegetables) and with a moderate consumption of ginger, garlic, onion, mixed aromatic spices (cinnamon + cardamom + cloves) and turmeric might be suggested.

Association of comorbidities with the COVID-19 severity and hospitalization: A study among the recovered individuals in Bangladesh.

Objectives: We aimed at the identification of the association of comorbidities with the COVID-19 severity and hospitalization. Methods: It is a retrospective cross-sectional study to investigate the variation in age, sex, dwelling, comorbidities, and medication with the COVID-19 severity and hospitalization by enrolling 1025 recovered individuals while comparing their time of recovery with or without comorbidities. Results: COVID-19 patients mostly suffered from fever. The predominant underlying medical conditions in them were hypertension (HTN) followed by diabetes mellitus (DM). Patients with cardiovascular disease (CVD) (54.3%) and hepatic disorders (HD) (43.6%) experienced higher severity. The risk of symptomatic cases was higher in aged (odds ratio, OR = 1.04, 95% CI = 1.02-1.06) and comorbid (OR = 1.87, 95% CI = 1.34-2.60) patients. T-test confirmed the differences between the comorbid and non-comorbid patients' recovery duration. The presence of multiple comorbidities increased the time of recovery (15-27 days) and hospitalization (20-40%). Increased symptomatic cases were found for patients having DM+HTN whereas CVD+Asthma patients were found with higher percentage of severity. Besides, DM+CKD (chronic kidney disease) was associated with higher hospitalization rate. Higher odds of severity were found for DM+CVD (OR = 4.42, 95% CI = 1.81-10.78) patients. Hospitalization risk was also increased for them (OR = 5.14, 95% CI = 2.02-13.07). Moreover, if they had HTN along with DM+CVD, they were found with even higher odds (OR = 6.82, 95% CI = 2.37-19.58) for hospitalization. Conclusion: Our study indicates that people who are aged, females, living in urban area and have comorbid conditions are at a higher risk for developing COVID-19 severity. Clinicians and health management authorities should prioritize these high-risk groups to reduce mortality attributed to the disease.

DCNN-based prediction model for detection of age-related macular degeneration from color fundus images.

Age-related macular degeneration (AMD) is a degenerative disorder in the macular region of the eye. AMD is the leading cause of irreversible vision loss in the elderly population. With the increase in aged population in the world, there is an urgent need to develop low-cost, hassle-free, and portable equipment diagnostic and analytical tools for early diagnosis. As AMD detection is done by examining the fundus images, its diagnosis is heavily dependent on medical personnel and their experience. To remove this issue, computer-aided algorithms may be used for AMD detection. The proposed work offers an effective solution to the AMD detection problem. It proposes a novel 13-layer deep convolutional neural network (DCNN) architecture to screen fundus images to spot direct signs of AMD. Five pairs of convolution and maxpool layers and three fully connected layers are utilized in the proposed network. Extensive simulations on original and augmented versions of two datasets (iChallenge-AMD and ARIA) consisting of healthy and diseased cases show a classification accuracy of 89.75%, 91.69%, and 99.45% on original and augmented versions of iChallenge-AMD and 90.00%, 93.03%, and 99.55% on ARIA, using a 10-fold cross-validation technique. It surpasses the best-known algorithm using DCNN by 2%.