scholarly journals Early detection of diabetic retinopathy based on deep learning and ultra-wide-field fundus images

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Kangrok Oh ◽  
Hae Min Kang ◽  
Dawoon Leem ◽  
Hyungyu Lee ◽  
Kyoung Yul Seo ◽  
...  
Keyword(s):  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Low Income ◽  
Detection System ◽  
Learning Technologies ◽  
Fundus Photography ◽  
Wide Field ◽  
Middle Income ◽  
Middle Income Countries ◽  
Automated Screening

AbstractVisually impaired and blind people due to diabetic retinopathy were 2.6 million in 2015 and estimated to be 3.2 million in 2020 globally. Though the incidence of diabetic retinopathy is expected to decrease for high-income countries, detection and treatment of it in the early stages are crucial for low-income and middle-income countries. Due to the recent advancement of deep learning technologies, researchers showed that automated screening and grading of diabetic retinopathy are efficient in saving time and workforce. However, most automatic systems utilize conventional fundus photography, despite ultra-wide-field fundus photography provides up to 82% of the retinal surface. In this study, we present a diabetic retinopathy detection system based on ultra-wide-field fundus photography and deep learning. In experiments, we show that the use of early treatment diabetic retinopathy study 7-standard field image extracted from ultra-wide-field fundus photography outperforms that of the optic disc and macula centered image in a statistical sense.

scholarly journals Efficacy of deep learning methods for predicting under-five mortality in 34 low-income and middle-income countries

BMJ Open ◽  
2020 ◽  
Vol 10 (8) ◽  
pp. e034524
Author(s):  
Adeyinka Emmanuel Adegbosin ◽  
Bela Stantic ◽  
Jing Sun
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Low Income ◽  
Maternal Education ◽  
Household Wealth ◽  
Secondary Outcome ◽  
Cross Sectional ◽  
Middle Income ◽  
Under Five ◽  
Middle Income Countries

ObjectivesTo explore the efficacy of machine learning (ML) techniques in predicting under-five mortality (U5M) in low-income and middle-income countries (LMICs) and to identify significant predictors of U5M.DesignThis is a cross-sectional, proof-of-concept study.Settings and participantsWe analysed data from the Demographic and Health Survey. The data were drawn from 34 LMICs, comprising a total of n=1 520 018 children drawn from 956 995 unique households.Primary and secondary outcome measuresThe primary outcome measure was U5M; secondary outcome was comparing the efficacy of deep learning algorithms: deep neural network (DNN); convolution neural network (CNN); hybrid CNN-DNN with logistic regression (LR) for the prediction of child’s survival.ResultsWe found that duration of breast feeding, number of antenatal visits, household wealth index, postnatal care and the level of maternal education are some of the most important predictors of U5M. We found that deep learning techniques are superior to LR for the classification of child survival: LR sensitivity=0.47, specificity=0.53; DNN sensitivity=0.69, specificity=0.83; CNN sensitivity=0.68, specificity=0.83; CNN-DNN sensitivity=0.71, specificity=0.83.ConclusionOur findings provide an understanding of determinants of U5M in LMICs. It also demonstrates that deep learning models are more efficacious than traditional analytical approach.

Assessment of county-level poverty alleviation progress by deep learning and satellite observations

2021 ◽  
Author(s):  
Liqiang Zhang ◽  
Yanxiao Jiang ◽  
Yang Li ◽  
Alicia J Zhou ◽  
Jing Cao ◽  
...  
Keyword(s):  
Growth Rate ◽  
Deep Learning ◽  
Low Income ◽  
Poverty Alleviation ◽  
County Level ◽  
Average Growth Rate ◽  
Middle Income ◽  
Average Growth ◽  
Middle Income Countries ◽  
Per Capita

Abstract Poverty alleviation is one of the greatest challenges faced by low-income and middle-income countries. China, which had the largest rural poverty-stricken population, has made tremendous efforts in alleviating poverty especially since the implementation of the targeted poverty alleviation (TPA) policy in 2014. Yet it remains unknown about the successfulness of the policy, because the official statistics are not timely available and in some cases questionable. This study combines deep learning with multiple satellite datasets to estimate county-level economic development from 2008 to 2019 and assess the effect of the TPA policy for 592 national poverty-stricken counties (NPCs) at country, provincial and county levels. Per capita gross domestic product (GDP) is used to measure the affluence level. From 2014 through 2019, the 592 NPCs experience an average growth rate of per capita GDP at 7.6%±0.4%, higher than the average growth rate of 310 adjacent non-NPC counties (7.3%±0.4%) and of the whole country (6.3%). This indicates an overall success of TPA policy so far. We also reveal 42 counties with weak growth recently and that the average affluence level of the NPCs in 2019 is still much lower than the national or provincial averages. The inexpensive, timely and accurate method proposed here can be applied to other low-income and middle-income countries for affluence assessment.

scholarly journals Tuberculosis and Non-Communicable Disease Multimorbidity: An Analysis of the World Health Survey in 48 Low- and Middle-Income Countries

2021 ◽  
Vol 18 (5) ◽  
pp. 2439
Author(s):  
Brendon Stubbs ◽  
Kamran Siddiqi ◽  
Helen Elsey ◽  
Najma Siddiqi ◽  
Ruimin Ma ◽  
...  
Keyword(s):  
Health Survey ◽  
Low Income ◽  
World Health ◽  
Low Income Countries ◽  
Multivariable Logistic Regression Analysis ◽  
World Health Survey ◽  
Middle Income ◽  
Middle Income Countries ◽  
The World ◽  
Low And Middle Income

Tuberculosis (TB) is a leading cause of mortality in low- and middle-income countries (LMICs). TB multimorbidity [TB and ≥1 non-communicable diseases (NCDs)] is common, but studies are sparse. Cross-sectional, community-based data including adults from 21 low-income countries and 27 middle-income countries were utilized from the World Health Survey. Associations between 9 NCDs and TB were assessed with multivariable logistic regression analysis. Years lived with disability (YLDs) were calculated using disability weights provided by the 2017 Global Burden of Disease Study. Eight out of 9 NCDs (all except visual impairment) were associated with TB (odds ratio (OR) ranging from 1.38–4.0). Prevalence of self-reported TB increased linearly with increasing numbers of NCDs. Compared to those with no NCDs, those who had 1, 2, 3, 4, and ≥5 NCDs had 2.61 (95% confidence interval (CI) = 2.14–3.22), 4.71 (95%CI = 3.67–6.11), 6.96 (95%CI = 4.95–9.87), 10.59 (95%CI = 7.10–15.80), and 19.89 (95%CI = 11.13–35.52) times higher odds for TB. Among those with TB, the most prevalent combinations of NCDs were angina and depression, followed by angina and arthritis. For people with TB, the YLDs were three times higher than in people without multimorbidity or TB, and a third of the YLDs were attributable to NCDs. Urgent research to understand, prevent and manage NCDs in people with TB in LMICs is needed.

scholarly journals Gestational weight gain in low-income and middle-income countries: a modelling analysis using nationally representative data

2020 ◽  
Vol 5 (11) ◽  
pp. e003423
Author(s):  
Dongqing Wang ◽  
Molin Wang ◽  
Anne Marie Darling ◽  
Nandita Perumal ◽  
Enju Liu ◽  
...  
Keyword(s):  
Weight Gain ◽  
North Africa ◽  
Low Income ◽  
Sub Saharan Africa ◽  
Middle Income ◽  
Middle Income Countries ◽  
Gestational Weight ◽  
Latin America And Caribbean ◽  
Nationally Representative ◽  
Sub Saharan

IntroductionGestational weight gain (GWG) has important implications for maternal and child health and is an ideal modifiable factor for preconceptional and antenatal care. However, the average levels of GWG across all low-income and middle-income countries of the world have not been characterised using nationally representative data.MethodsGWG estimates across time were computed using data from the Demographic and Health Surveys Program. A hierarchical model was developed to estimate the mean total GWG in the year 2015 for all countries to facilitate cross-country comparison. Year and country-level covariates were used as predictors, and variable selection was guided by the model fit. The final model included year (restricted cubic splines), geographical super-region (as defined by the Global Burden of Disease Study), mean adult female body mass index, gross domestic product per capita and total fertility rate. Uncertainty ranges (URs) were generated using non-parametric bootstrapping and a multiple imputation approach. Estimates were also computed for each super-region and region.ResultsLatin America and Caribbean (11.80 kg (95% UR: 6.18, 17.41)) and Central Europe, Eastern Europe and Central Asia (11.19 kg (95% UR: 6.16, 16.21)) were the super-regions with the highest GWG estimates in 2015. Sub-Saharan Africa (6.64 kg (95% UR: 3.39, 9.88)) and North Africa and Middle East (6.80 kg (95% UR: 3.17, 10.43)) were the super-regions with the lowest estimates in 2015. With the exception of Latin America and Caribbean, all super-regions were below the minimum GWG recommendation for normal-weight women, with Sub-Saharan Africa and North Africa and Middle East estimated to meet less than 60% of the minimum recommendation.ConclusionThe levels of GWG are inadequate in most low-income and middle-income countries and regions. Longitudinal monitoring systems and population-based interventions are crucial to combat inadequate GWG in low-income and middle-income countries.

scholarly journals Urgent needs of low-income and middle-income countries for COVID-19 vaccines and therapeutics

The Lancet ◽  
2021 ◽  
Vol 397 (10274) ◽  
pp. 562-564
Author(s):  
J Peter Figueroa ◽  
Maria Elena Bottazzi ◽  
Peter Hotez ◽  
Carolina Batista ◽  
Onder Ergonul ◽  
...  
Keyword(s):  
Low Income ◽  
Middle Income ◽  
Middle Income Countries
Sign in / Sign up

Export Citation Format

Share Document