Program Findings Brochure (2020)
Project design and results for the 2020 calendar year are summarized in this brief document.
Formative Research Report (2018)
This report presents findings from formative research conducted prior to the surveillance system’s implementation, in order to inform the design of the surveillance system. Download formative report.
Regularized Bayesian transfer learning for population-level etiological distributions
Datta et al. Biostatistics (2020)
Summary excerpt: Computer-coded verbal autopsy (CCVA) algorithms predict cause of death from high- dimensional family questionnaire data (verbal autopsy) of a deceased individual, which are then aggregated to generate national and regional estimates of cause-specific mortality fractions. These estimates may be inaccurate if CCVA is trained on non-local training data different from the local population of interest.
Generalized Bayesian Quantification Learning
Fiksel et al. (Under review)
Abstract excerpt: Quantification learning is the task of prevalence estimation for a test population using predictions from a classifier trained on a differen population. Commonly used quantification methods either assume perfect sensitivity and specificity of the classifier, or use the training data to both train the classifier and also estimate its missclassification rates… We proposed a generalized Bayesian quantification learning (GBQL) approach that uses the entire compositional predictions from probabilistic classifiers and allows for uncertainty in true class labeled test data. Download preprint article.
Johns Hopkins Bloomberg School
of Public Health (JHSPH)
615 N. Wolfe Street, Baltimore, MD 21205
Instituto Nacional de Saúde (INS)
Vila De Marracuene, Estrada Nacional N°1,
Província De Maputo – Moçambique
Instituto Nacional de Estatística (INE)
Av. 24 De Julho 1989 Caixa Postal 493,
Cidade De Maputo – Moçambique