Journal article
Elizabeth Bondi-Kelly, A. Perrault, Fei Fang, Benjamin L. Rice, C. Golden, Milind Tambe
2020
2020
Semantic Scholar
Cite
Cite
APA
Click to copy
Bondi-Kelly, E., Perrault, A., Fang, F., Rice, B. L., Golden, C., & Tambe, M. (2020). Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency Prediction.
Chicago/Turabian
Click to copy
Bondi-Kelly, Elizabeth, A. Perrault, Fei Fang, Benjamin L. Rice, C. Golden, and Milind Tambe. “Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency Prediction” (2020).
MLA
Click to copy
Bondi-Kelly, Elizabeth, et al. Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency Prediction. 2020.
BibTeX Click to copy
@article{elizabeth2020a,
title = {Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency Prediction},
year = {2020},
author = {Bondi-Kelly, Elizabeth and Perrault, A. and Fang, Fei and Rice, Benjamin L. and Golden, C. and Tambe, Milind}
}
Abstract
The lack of micronutrients is a major threat to the health and development of populations, and it is challenging to detect such deficiency at large scale with low cost. In this work, we plan to use data from a study on micronutrient deficiency in Madagascar, which include blood draw results and corresponding questionnaires, along with satellite imagery, to determine whether there are certain cues visible in satellite imagery that could more easily and quickly suggest areas where people may be susceptible to micronutrient deficiency. We propose an approach that will (i) determine important predictors of micronutrient deficiency from blood draws and corresponding questionnaire data, such as type of food consumed, (ii) automatically detect related areas in satellite imagery, such as forest regions where sources of important food may be found, and (iii) use these to predict regions of micronutrient deficiency. As the prediction of micronutrient deficiency in satellite imagery will be done using meaningful predictors, and we anticipate using an inherently interpretable model for prediction based on these objects, such as logistic regression, we aim to create a model that will be intuitive to those in the public health community.1
