Ethnic variation in outcome of people hospitalised during the first COVID-19 epidemic wave in Wales (UK): an analysis of national surveillance data using Onomap, a name-based ethnicity classification tool

ObjectiveTo identify ethnic differences in proportion positive for SARS-CoV-2, and proportion hospitalised, proportion admitted to intensive care and proportion died in hospital with COVID-19 during the first epidemic wave in Wales.DesignDescriptive analysis of 76 503 SARS-CoV-2 tests carried out in Wales to 31 May 2020. Cohort study of 4046 individuals hospitalised with confirmed COVID-19 between 1 March and 31 May. In both analyses, ethnicity was assigned using a name-based classifier.SettingWales (UK).Primary and secondary outcomesAdmission to an intensive care unit following hospitalisation with a positive SARS-CoV-2 PCR test. Death within 28 days of a positive SARS-CoV-2 PCR test.ResultsUsing a name-based ethnicity classifier, we found a higher proportion of black, Asian and ethnic minority people tested for SARS-CoV-2 by PCR tested positive, compared with those classified as white. Hospitalised black, Asian and minority ethnic cases were younger (median age 53 compared with 76 years; p<0.01) and more likely to be admitted to intensive care. Bangladeshi (adjusted OR (aOR): 9.80, 95% CI 1.21 to 79.40) and ‘white – other than British or Irish’ (aOR: 1.99, 95% CI 1.15 to 3.44) ethnic groups were most likely to be admitted to intensive care unit. In Wales, older age (aOR for over 70 years: 10.29, 95% CI 6.78 to 15.64) and male gender (aOR: 1.38, 95% CI 1.19 to 1.59), but not ethnicity, were associated with death in hospitalised patients.ConclusionsThis study adds to the growing evidence that ethnic minorities are disproportionately affected by COVID-19. During the first COVID-19 epidemic wave in Wales, although ethnic minority populations were less likely to be tested and less likely to be hospitalised, those that did attend hospital were younger and more likely to be admitted to intensive care. Primary, secondary and tertiary COVID-19 prevention should target ethnic minority communities in Wales.

Recognizing Human Races through Machine Learning—A Multi-Network, Multi-Features Study

The human face holds a privileged position in multi-disciplinary research as it conveys much information—demographical attributes (age, race, gender, ethnicity), social signals, emotion expression, and so forth. Studies have shown that due to the distribution of ethnicity/race in training datasets, biometric algorithms suffer from “cross race effect”—their performance is better on subjects closer to the “country of origin” of the algorithm. The contributions of this paper are two-fold: (a) first, we gathered, annotated and made public a large-scale database of (over 175,000) facial images by automatically crawling the Internet for celebrities’ images belonging to various ethnicity/races, and (b) we trained and compared four state of the art convolutional neural networks on the problem of race and ethnicity classification. To the best of our knowledge, this is the largest, data-balanced, publicly-available face database annotated with race and ethnicity information. We also studied the impact of various face traits and image characteristics on the race/ethnicity deep learning classification methods and compared the obtained results with the ones extracted from psychological studies and anthropomorphic studies. Extensive tests were performed in order to determine the facial features to which the networks are sensitive to. These tests and a recognition rate of 96.64% on the problem of human race classification demonstrate the effectiveness of the proposed solution.

A Deep Learning-Based Framework for Accurate Facial Ethnicity Classification and Efficient Query Retrieval

Facial images carry important demographic information such as ethnicity and gender. Ethnicity is an essential part of human identity and serves as a useful identifier for numerous applications ranging from biometric recognition, targeted advertising to social media profiling. Recent years have seen a huge spike in the use of convolutional neural networks (CNNs) for various visual, face recognition problems. The ability of the CNN to take advantage of the hierarchical pattern in data makes it a suitable model for facial ethnicity classification. As facial datasets lack ethnicity information it becomes extremely difficult to classify images. In this chapter a deep learning framework is proposed that classifies the individual into their respective ethnicities which are Asian, African, Latino, and White. The performances of various deep learning techniques are documented and compared for accuracy of classification. Also, a simple efficient face retrieval model is built which retrieves similar faces. The aim of this model is to reduce the search time by 1/3 of the original retrieval model.

Ethnicity classification systems for public health surveys: experiences from HIV behavioural surveillance among men who have sex with men

Background Race and ethnicity classification systems have considerable implications for public health, including the potential to reveal or mask inequities. Given increasing “super-diversity” and multiple racial/ethnic identities in many global settings, especially among younger generations, different ethnicity classification systems can underrepresent population heterogeneity and can misallocate and render invisible Indigenous people and ethnic minorities. We investigated three ethnicity classification methods and their relationship to sample size, socio-demographics and sexual health indicators. Methods We examined data from New Zealand’s HIV behavioural surveillance programme for men who have sex with men (MSM) in 2006, 2008, 2011, and 2014. Participation was voluntary, anonymous and self-completed; recruitment was via community venues and online. Ethnicity allowed for multiple responses; we investigated three methods of dealing with these: Prioritisation, Single/Combination, and Total Response. Major ethnic groups included Asian, European, indigenous Māori, and Pacific. For each classification method, statistically significant associations with ethnicity for demographic and eight sexual health indicators were assessed using multivariable logistic regression. Results Overall, 10,525 MSM provided ethnicity data. Classification methods produced different sample sizes, and there were ethnic disparities for every sexual health indicator. In multivariable analysis, when compared with European MSM, ethnic differences were inconsistent across classification systems for two of the eight sexual health outcomes: Māori MSM were less likely to report regular partner condomless anal intercourse using Prioritisation or Total Response but not Single/Combination, and Pacific MSM were more likely to report an STI diagnosis when using Total Response but not Prioritisation or Single/Combination. Conclusions Different classification approaches alter sample sizes and identification of health inequities. Future research should strive for equal explanatory power of Indigenous and ethnic minority groups and examine additional measures such as socially-assigned ethnicity and experiences of discrimination and racism. These findings have broad implications for surveillance and research that is used to inform public health responses.

Ethnicity Classification Based On Facial Features Using Viola-Jones Algorithm

Biometric recognition system can use race classification to identify the human globally with a particular identity. This paper proposes Support Vector Machine and will compare the result with K-Nearest Neighbor for classification of people into two major races namely Indonesian western and eastern races. Firstly, the proposed classification method extracts the distinct primary facial feature and skin color model of the given face with Viola-Jones Algorithm to effectively classify the races. To increase the accuracy, the sample must not contain any background of other people skin, no movement and the pictures were taken from the mobile camera with no beauty filter.

Ethnic variation in outcome of people hospitalised with Covid-19 in Wales (UK): A rapid analysis of surveillance data using Onomap, a name-based ethnicity classification tool

There is growing evidence that ethnic minorities in Europe are disproportionately affected by Covid-19. Using a name-based ethnicity classifier, we found that hospitalised Black, Asian and minority ethnic cases were younger and more likely to be admitted to intensive care (ICU). Pakistani, Bangladeshi and White - other than British or Irish, ethnic groups were most at risk. In this study, older age and male gender, but not ethnicity, were associated with death in hospitalised patients.