Geospatial methods for Public Health Surveillance in Urban Environments
Patrick Ken Kalonde – PhD Student
5th Symposium on Artificial Intelligence and Machine Learning in Malawi23 August 2024
SINK
SOURCE
Drivers
IMPACTS
A framework to help crafting of sustainable solutions
Response
Yang et al 2020). https://doi.org/10.1021/acs.est.0c04292
Reduced risk of exposure
How does proximity to urban waste (as potential reservoirs of pathogenic microbes) associate with exposure risk?
Research question
Hypothesis
Proximity to urban waste increases the risk of exposure to pathogenic microbes, such that people living in close proximity to waste have high risk of exposure than those that live far away.
Automate the process
Robert M., Karthikeyan Shanmugam, and Its' Hak Dinstein. "Textural features for image classification." IEEE Transactions on systems, man, and cybernetics 6 (1973): 610-621.
Kalonde, Patrick, "Geospatial Methods for Mapping Domestic Waste Piles and Macro Plastics" (2022). Culminating Projects in Geography and Planning. 13. https://repository.stcloudstate.edu/gp_etds/13
Patterns associated with waste dispersal
RSTMH.org | #RSTMHResearch
Before storm
After storm
TOTAL = 5938
| | | |
| | | |
| | | |
| | | |
N = 96,929
N = 111,817
TOTAL NUMBER OF PIXELS = 10,838,309,993
TOTAL NUMBER OF SEGMENTS = 1,362,981
Grouping similar pixels using Meanshift
BEST WAY TO LABEL THESE EFFICIENTLY TO ENABLE GROUNT TRUTHING WITHIN REASONABLE TIME
Current work and next steps
Lengau, Africa’s fastest computer
Ronneberger, Olaf, Philipp Fischer, and Thomas Brox. "U-net: Convolutional networks for biomedical image segmentation." In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, pp. 234-241. Springer International Publishing, 2015.
‘Science must respond to societal needs and global challenges’ ��Unknown
Questions, thoughts and suggestions