scholarly journals Multivariate random forest prediction of poverty and malnutrition prevalence

PLoS ONE ◽  
2021 ◽  
Vol 16 (9) ◽  
pp. e0255519
Author(s):  
Chris Browne ◽  
David S. Matteson ◽  
Linden McBride ◽  
Leiqiu Hu ◽  
Yanyan Liu ◽  
...  
Keyword(s):  
Random Forest ◽  
Transfer Learning ◽  
Predictive Accuracy ◽  
Monitoring And Evaluation ◽  
Early Warning Systems ◽  
Learning Methods ◽  
Middle Income ◽  
Middle Income Countries ◽  
Random Forest Models ◽  
Poverty Mapping

Advances in remote sensing and machine learning enable increasingly accurate, inexpensive, and timely estimation of poverty and malnutrition indicators to guide development and humanitarian agencies’ programming. However, state of the art models often rely on proprietary data and/or deep or transfer learning methods whose underlying mechanics may be challenging to interpret. We demonstrate how interpretable random forest models can produce estimates of a set of (potentially correlated) malnutrition and poverty prevalence measures using free, open access, regularly updated, georeferenced data. We demonstrate two use cases: contemporaneous prediction, which might be used for poverty mapping, geographic targeting, or monitoring and evaluation tasks, and a sequential nowcasting task that can inform early warning systems. Applied to data from 11 low and lower-middle income countries, we find predictive accuracy broadly comparable for both tasks to prior studies that use proprietary data and/or deep or transfer learning methods.

scholarly journals Facing up to (online) fashion and fads … Face-to-face contact is here to stay in M&E capacity building. Evidence from 35 National Evaluation Societies

2016 ◽  
Vol 4 (1) ◽  
Author(s):  
Sara Dewachter ◽  
Nathalie Holvoet
Keyword(s):  
Communities Of Practice ◽  
Capacity Building ◽  
Supply And Demand ◽  
Qualitative Comparative Analysis ◽  
Monitoring And Evaluation ◽  
Evaluation Capacity Building ◽  
Middle Income ◽  
Evaluation Capacity ◽  
Face To Face ◽  
Middle Income Countries

Background: Over the years, Communities of Practice have gained popularity as a capacity-building method among Monitoring and Evaluation practitioners. Yet, thus far, relatively little is known about their effectiveness.Objectives: This article focuses on National Evaluation Societies as Communities of Practice that aim to contribute to the monitoring and evaluation capacity building of their members.Method: Drawing upon a survey of 35 National Evaluation Societies in 33 low- and middle-income countries, we explore to what extent capacity building efforts have been successful and what factors explain the relative success or failure in capacity building. We rely upon Qualitative Comparative Analysis as we are particularly interested in different pathways to ensure successful National Evaluation Societies.Results: Our findings highlight that regular face-to-face contact is a particularly important element. This does not entirely come as a surprise, as monitoring and evaluation capacity building often implies tacit knowledge that is most effectively shared face-to-face. Furthermore, capacity building in conducting and, particularly, using evaluations entails building networks among the monitoring and evaluation supply and demand side which can most easily be done through regular face-to-face interaction.Conclusion: Our findings are not only theoretically interesting, they are also policy relevant; they hint at the fact that in an era of quick advances in technology, investing in face-to-face contact among members remains important.

scholarly journals A Multimodality Machine Learning Approach to Differentiate Severe and Nonsevere COVID-19: Model Development and Validation

10.2196/23948 ◽  
2021 ◽  
Vol 23 (4) ◽  
pp. e23948
Author(s):  
Yuanfang Chen ◽  
Liu Ouyang ◽  
Forrest S Bao ◽  
Qian Li ◽  
Lei Han ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Laboratory Test ◽  
Predictive Accuracy ◽  
Learning Approach ◽  
Test Results ◽  
Laboratory Test Results ◽  
Machine Learning Approach ◽  
Forest Models ◽  
Random Forest Models

Background Effectively and efficiently diagnosing patients who have COVID-19 with the accurate clinical type of the disease is essential to achieve optimal outcomes for the patients as well as to reduce the risk of overloading the health care system. Currently, severe and nonsevere COVID-19 types are differentiated by only a few features, which do not comprehensively characterize the complicated pathological, physiological, and immunological responses to SARS-CoV-2 infection in the different disease types. In addition, these type-defining features may not be readily testable at the time of diagnosis. Objective In this study, we aimed to use a machine learning approach to understand COVID-19 more comprehensively, accurately differentiate severe and nonsevere COVID-19 clinical types based on multiple medical features, and provide reliable predictions of the clinical type of the disease. Methods For this study, we recruited 214 confirmed patients with nonsevere COVID-19 and 148 patients with severe COVID-19. The clinical characteristics (26 features) and laboratory test results (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest models based on all the features in each modality as well as on the top 5 features in each modality combined were developed and validated to differentiate COVID-19 clinical types. Results Using clinical and laboratory results independently as input, the random forest models achieved >90% and >95% predictive accuracy, respectively. The importance scores of the input features were further evaluated, and the top 5 features from each modality were identified (age, hypertension, cardiovascular disease, gender, and diabetes for the clinical features modality, and dimerized plasmin fragment D, high sensitivity troponin I, absolute neutrophil count, interleukin 6, and lactate dehydrogenase for the laboratory testing modality, in descending order). Using these top 10 multimodal features as the only input instead of all 52 features combined, the random forest model was able to achieve 97% predictive accuracy. Conclusions Our findings shed light on how the human body reacts to SARS-CoV-2 infection as a unit and provide insights on effectively evaluating the disease severity of patients with COVID-19 based on more common medical features when gold standard features are not available. We suggest that clinical information can be used as an initial screening tool for self-evaluation and triage, while laboratory test results should be applied when accuracy is the priority.

scholarly journals Mapping access to basic hygiene services in low- and middle-income countries: A cross-sectional case study of geospatial disparities

2020 ◽  
Author(s):  
Weiyu Yu ◽  
Robert E. S. Bain ◽  
Jie Yu ◽  
Victor Alegana ◽  
Winfred Dotse-Gborgbortsi ◽  
...  
Keyword(s):  
Random Forest ◽  
Scale Up ◽  
Random Forest Model ◽  
Demographic And Health Surveys ◽  
Ensemble Model ◽  
Cross Sectional ◽  
Middle Income ◽  
Forest Model ◽  
Middle Income Countries ◽  
Low And Middle Income

Handwashing with water and soap, is among the most a cost-effective interventions to improve public health. Yet billions of people globally lacking handwashing facilities with water and soap on premises, with gaps particularly found in low- and middle-income countries. Targeted efforts to expand access to basic hygiene services require data at geospatially explicit scales. Drawing on country-specific cross-sectional Demographic and Health Surveys with georeferenced hygiene data, we developed an ensemble model to predict the prevalence of basic hygiene facilities in Malawi, Nepal, Nigeria, Pakistan and Uganda. The ensemble model was based on a multiple-level stacking structure, where five predictive modelling algorithms were used to produce sub-models, and a random forest model was used to generalise the final predictions. An inverse distance weighted interpolation was incorporated in the random forest model to account for spatial autocorrelation. Local coverage and a local dissimilarity index were calculated to examine the geographic disparities in access. Our methodology produced robust outputs, as evidenced by performance evaluations (all R2 were above 0.8 with the exception of Malawi where R2 = 0.6). Among the five study countries, Pakistan had the highest overall coverage, whilst Malawi had the poorest coverage. Apparent disparities in basic hygiene services were found across geographic locations and between urban and rural settings. Nigeria had the highest level of inequalities in basic hygiene services, whilst Malawi showed the least segregation between populations with and without basic hygiene services. Both educational attainment and wealth were important predictors of the geospatial distribution of basic hygiene services. By producing geospatially explicit estimates of the prevalence of handwashing facilities with water and soap, this study provides a means of identifying geographical disparities in basic hygiene services. The method and outputs can be useful tools to identify areas of low coverage and to support efficient and precise targeting of efforts to scale up access to handwashing facilities and shift social and cultural norms on handwashing.

scholarly journals Improving surgical systems in low- and middle-income countries: an inclusive framework for monitoring and evaluation

2015 ◽  
Vol 7 (6) ◽  
pp. 380-383 ◽  
Author(s):  
Peter G. Bendix ◽  
Jamie E. Anderson ◽  
John A. Rose ◽  
Emilia V. Noormahomed ◽  
Stephen W. Bickler
Keyword(s):  
Monitoring And Evaluation ◽  
Middle Income ◽  
Middle Income Countries ◽  
Low And Middle Income

scholarly journals Advocacy is essential to supporting women with pre-eclampsia

2017 ◽  
Vol 10 (1) ◽  
pp. 33-35 ◽  
Author(s):  
Eleni Tsigas
Keyword(s):  
Maternal Health ◽  
Critical Role ◽  
Monitoring And Evaluation ◽  
Patient Advocacy ◽  
Advocacy Groups ◽  
Future Directions ◽  
Middle Income ◽  
Middle Income Countries ◽  
Concrete Action ◽  
Patient Advocacy Groups

Advocacy has a critical role in advancing the maternal health agenda. Patient advocacy groups can hold governments and other stakeholders accountable and ensure that commitments are translated into concrete action. This article highlights the advocacy efforts of the Preeclampsia Foundation, a patient advocacy organisation that aims to improve the diagnosis, management, and prevention of pre-eclampsia through research and improved healthcare practices. A number of challenges continue to face maternal health advocacy especially in low- and middle-income countries. Future directions include developing a strategic focus for advocacy, effectively engaging citizens to build a culture of accountability, and monitoring and evaluation of advocacy efforts.

MAPPING OF HEALTH TECHNOLOGY ASSESSMENT IN SELECTED COUNTRIES

2013 ◽  
Vol 29 (4) ◽  
pp. 424-434 ◽  
Author(s):  
Wija Oortwijn ◽  
Pieter Broos ◽  
Hindrik Vondeling ◽  
David Banta ◽  
Lora Todorova
Keyword(s):  
Health Technology Assessment ◽  
Technology Assessment ◽  
Monitoring And Evaluation ◽  
Relevant Information ◽  
Health Technology ◽  
Dissemination And Implementation ◽  
Policy And Practice ◽  
Middle Income ◽  
Middle Income Countries ◽  
Country Level

Objectives: The aim of this study was to develop and apply an instrument to map the level of health technology assessment (HTA) development at country level in selected countries. We examined middle-income countries (Argentina, Brazil, India, Indonesia, Malaysia, Mexico, and Russia) and countries well-known for their comprehensive HTA programs (Australia, Canada, and United Kingdom).Methods: A review of relevant key documents regarding the HTA process was performed to develop the instrument which was then reviewed by selected HTAi members and revised. We identified and collected relevant information to map the level of HTA in the selected countries. This was supplemented by information from a structured survey among HTA experts in the selected countries (response rate: 65/385).Results: Mapping of HTA in a country can be done by focusing on the level of institutionalization and the HTA process (identification, priority setting, assessment, appraisal, reporting, dissemination, and implementation in policy and practice). Although HTA is most advanced in industrialized countries, there is a growing community in middle-income countries that uses HTA. For example, Brazil is rapidly developing effective HTA programs. India and Russia are at the very beginning of introducing HTA. The other middle-income countries show intermediate levels of HTA development compared with the reference countries.Conclusions: This study presents a set of indicators for documenting the current level and trends in HTA at country level. The findings can be used as a baseline measurement for future monitoring and evaluation. This will allow a variety of stakeholders to assess the development of HTA in their country, help inform strategies, and justify expenditure for HTA.

Advances in monitoring and evaluation in low- and middle-income countries

2021 ◽  
pp. 101994
Author(s):  
James C. Thomas ◽  
Kathy Doherty ◽  
Stephanie Watson-Grant ◽  
Manish Kumar
Keyword(s):  
Monitoring And Evaluation ◽  
Middle Income ◽  
Middle Income Countries ◽  
Low And Middle Income

scholarly journals Detecting COVID-19 in chest images based on deep transfer learning and machine learning algorithms

2021 ◽  
Vol 52 (1) ◽  
Author(s):  
Seyed Masoud Rezaeijo ◽  
Mohammadreza Ghorvei ◽  
Razzagh Abedi-Firouzjah ◽  
Hesam Mojtahedi ◽  
Hossein Entezari Zarch
Keyword(s):  
Machine Learning ◽  
Transfer Learning ◽  
Learning Algorithms ◽  
Chest Ct ◽  
Machine Learning Algorithms ◽  
Limited Resources ◽  
Learning Models ◽  
Laboratory Equipment ◽  
Middle Income ◽  
Middle Income Countries

Abstract Background This study aimed to propose an automatic prediction of COVID-19 disease using chest CT images based on deep transfer learning models and machine learning (ML) algorithms. Results The dataset consisted of 5480 samples in two classes, including 2740 CT chest images of patients with confirmed COVID-19 and 2740 images of suspected cases was assessed. The DenseNet201 model has obtained the highest training with an accuracy of 100%. In combining pre-trained models with ML algorithms, the DenseNet201 model and KNN algorithm have received the best performance with an accuracy of 100%. Created map by t-SNE in the DenseNet201 model showed not any points clustered with the wrong class. Conclusions The mentioned models can be used in remote places, in low- and middle-income countries, and laboratory equipment with limited resources to overcome a shortage of radiologists.

Sign in / Sign up

Export Citation Format

Share Document