Domain adaptation (DA) and domain generalization (DG) have emerged as a solution to the domain shift problem where the distribution of the source and target data is different. The task of DG is more challenging than DA as the target data is totally unseen during the training phase in DG scenarios. The current state-of-the-art employs adversarial techniques, however, these are rarely considered for the DG problem. Furthermore, these approaches do not consider correlation alignment which has been proven highly beneficial for minimizing domain discrepancy. In this paper, we propose a correlation-aware adversarial DA and DG framework where the features of the source and target data are minimized using correlation alignment along with adversarial learning.Incorporating the correlation alignment module along with adversarial learning helps to achieve a more domain agnostic model due to the improved ability to reduce domain discrepancy with unlabeled target data more effectively. Experiments on benchmark datasets serve as evidence that our proposed method yields improved state-of-the-art performance.

Domain adaption (DA) and domain generalization (DG) are two closely related methods which are both concerned with the task of assigning labels to an unlabeled data set. The only dissimilarity between these approaches is that DA can access the target data during the training phase, while the target data is totally unseen during the training phase in DG. The task of DG is challenging as we have no earlier knowledge of the target samples. If DA methods are applied directly to DG by a simple exclusion of the target data from training, poor performance will result for a given task. In this paper, we tackle the domain generalization challenge in two ways. In our first approach, we propose a novel deep domain generalization architecture utilizing synthetic data generated by a Generative Adversarial Network (GAN). The discrepancy between the generated images and synthetic images is minimized using existing domain discrepancy metrics such as maximum mean discrepancy or correlation alignment. In our second approach, we introduce a protocol for applying DA methods to a DG scenario by excluding the target data from the training phase, splitting the source data to training and validation parts, and treating the validation data as target data for DA. We conduct extensive experiments on four cross-domain benchmark datasets. Experimental results signify our proposed model outperforms the current state-of-the-art methods for DG.

Vietnamese populations in Vietnam and the United States have a high prevalence of smoking. The associations among behavioral risk factors, acculturation, and smoking among the Vietnamese population living in the United States are not well documented. The present study aimed to identify the factors associated with smoking behavior among Vietnamese men living in Santa Clara County, California. A cross-sectional random-digit-dialed telephone survey was conducted. The sampling frame consisted of 27 Vietnamese surnames from the Santa Clara County telephone directory. A total of 660 adult respondents were interviewed to collect information on general health status, alcohol and tobacco use, HIV/AIDS, sexual behavior, injury control, hypertension, cholesterol screening, and acculturation. Of the 660 adults interviewed, 364 (55.2%) were male and 296 (44.8%) were female. Among males, 31.9% were current smokers, and among females, only one woman reported smoking. Univariate analyses revealed that having less than a college education, having poor English language skills, using Vietnamese at home and with friends, being less acculturated, not having a routine physical or blood cholesterol check, and being a binge drinker were significantly associated with an increased likelihood of smoking. Multivariate analysis revealed two independently associated factors: Respondents who were more acculturated were less likely to smoke (OR = 0.38, 95% CI = 0.18-0.83), and those not having cholesterol checked were more likely to smoke (OR = 2.48, 95% CI = 1.30-4.71). Acculturation level was inversely associated with smoking among Vietnamese adult men in Santa Clara County. Other health risk behaviors coexisted with smoking behavior and should be considered in prevention programs.

Graphical abstract

No abstract

The internet has become an essential part of our daily life. But excessive usage can have a negative impact on the physical health of its users. Over the last decade, the use of Social Media (Facebook) has been increasing rapidly and the younger generations getting addicted to it. But all possible health impacts of excessive use of internet are yet to be thoroughly evaluated, especially in such a developing country as Bangladesh. The present study aims to understand possible health deteriorations from excessive use of Facebook in a cohort of university students of Bangladesh. A cross-sectional study was conducted on 1186 students from two public universities and 1472 from several private universities of Bangladesh using a comprehensive questionnaire. The data were analyzed using the chi-square test to understand the association between Facebook usage behaviors and physical health status. We found that ~70% of the students used the internet for at least 4–6 hours/day, and ~27% of them used Facebook for >3 hrs. Students frequently use social media (mostly Facebook) for news and social communication. About 50% of the students reported wasting time on Facebook and going to sleep late because of it. Importantly, 47.3% students reported that excessive use of Facebook results sleeping disturbance and has a negative impact on the concentration of daily works/studies (p < 0.001). In addition, they experienced several other health problems, including worsening eyesight (71.2%), headaches (15.4%), back and neck pain (28%). Although not statistically important, a fair number of students sought medical attention due to the daily excessive use of internet (p-value = 0.112). These findings demands better understanding of the all possible impacts of using excessive internet among the University students, which can help take the necessary initiatives to encourage good use of the internet. Further extension of this study is suggested at all education levels to reveal the full scenario of degree of excessive internet use and its impact on the healths of Bangladeshi students.

In the presence of large sets of labeled data, Deep Learning (DL) has accomplished extraordinary triumphs in the avenue of computer vision, particularly in object classification and recognition tasks. However, DL cannot always perform well when the training and testing images come from different distributions or in the presence of domain shift between training and testing images. They also suffer in the absence of labeled input data. Domain adaptation (DA) methods have been proposed to make up the poor performance due to domain shift. In this paper, we present a new unsupervised deep domain adaptation method based on the alignment of second order statistics (covariances) as well as maximum mean discrepancy of the source and target data with a two stream Convolutional Neural Network (CNN). We demonstrate the ability of the proposed approach to achieve state-of-the-art performance for image classification on three benchmark domain adaptation datasets: Office-31 [27], Office-Home [37] and Office-Caltech [8].

As the coronavirus disease 2019 , caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), rages across the world, killing hundreds of thousands and infecting millions, researchers are racing against time to elucidate the viral genome. Some Bangladeshi institutes are also in this race, sequenced a few isolates of the virus collected from Bangladesh. Here, we present a genomic analysis of 14 isolates. The analysis revealed that SARS-CoV-2 isolates sequenced from Dhaka and Chittagong were the lineage of Europe and the Middle East, respectively. Our analysis identified a total of 42 mutations, including three large deletions, half of which were synonymous. Most of the missense mutations in Bangladeshi isolates found to have weak effects on the pathogenesis. Some mutations may lead the virus to be less pathogenic than the other countries. Molecular docking analysis to evaluate the effect of the mutations on the interaction between the viral spike proteins and the human ACE2 receptor, though no significant interaction was observed. This study provides some preliminary insights into the origin of Bangladeshi SARS-CoV-2 isolates, mutation spectrum and its possible pathomechanism, which may give an essential clue for designing therapeutics and management of COVID-19 in Bangladesh.