delahunt@uw.edu
Welcome! My work focuses on AI - (1) applications to global health care in LMICs (low and middle income countries), and (2) basic research. Below please find links to papers. (*) indicates lead author.
Read More
Resumé
Machine learning papers: health care in LMICs
-
(*) Beyond validation loss: Clinically-tailored optimization metrics improve a model's clinical performance. In review, 2026
(*) Designing AI Algorithms to Suit Local Context, MICCAI MIRASOL, 2025
(*) Reducing Poisson error can offset classification error: a technique to meet clinical performance requirements. ML4H, 2024
(*) Metrics to guide development of machine learning algorithms for malaria diagnosis. Frontiers Mal, 2024
NTDscope: A multi-contrast portable microscope for disease diagnosis. PLOS Global Public Health, 2026
Multi-contrast machine learning improves schistosomiasis diagnostic performance. PLOS NTDs, 2025
A comprehensive metadata repository of available malaria blood film image datasets. Bioimage Informatics, 2025
A schistosomiasis dataset with bright- and darkfield images. MICCAI Open Data, 2024
Evaluation of an automated microscope using ML for the detection of malaria in travelers returned to the UK. Frontiers Mal, 2023
How good are synthetic medical images? An empirical study with lung ultrasound. MICCAI SASHIMI, 2023
Deep learning video classification of lung ultrasound features associated with pneumonia. CVPR, 2023
Field evaluation of the diagnostic performance of EasyScan GO: a digital malaria microscopy device based on ML. Malaria J, 2022
Performance of a fully‐automated system on a WHO malaria microscopy evaluation slide set. Malaria J, 2021
(*) Algorithms to predict moisture content of grain using relative humidity time-series. IEEE GHTC, 2020
(*) Fully-automated patient-level malaria assessment on field-prepared thin blood film microscopy images. IEEE GHTC, 2019
Automated microscopy for malaria: a field comparison in Peru. Malaria J, 2018
Automated malaria diagnosis using CNNs. ICCV, 2017
(*) Automated microscopy and machine learning for malaria. IEEE GHTC, 2015
(*) Limitations of haemozoin-based diagnosis. Malaria J, 2014
-
(*) A toolkit for data-driven discovery of governing equations in high noise regimes. IEEE Access, 2021
(*) Built to Last: Functional and structural mechanisms in the moth olfactory network mitigate effects of neural injury. Brain Sciences, 2021
(*) Predicting United States policy outcomes with Random Forests. INET, 2020
(*) Money on the Table: Statistical information ignored by Softmax can improve classifier accuracy. arXiv, 2019
(*) Putting a bug in ML: The moth learns to read MNIST. Neural Networks, 2019
(*) Insect cyborgs: Bio-mimetic feature generators improve ML accuracy on limited data. NeurIPS Real neurons and hidden units, 2019
(*) Engineered for Function: The Power of Biologically Constrained Neural Networks for Neurosensory Integration. SIAM News, July 2019
(*) A moth brain learns to read MNIST. ICLR workshop, 2018
(*) Biological Mechanisms for Learning. Frontiers Neuroscience, 2018