Phenotype Image Data and Digital Learning Innovation (PhIDDLI)

Aim:

Feasibility study of commercial off-the-shelf (COTS) artificial intelligence and machine learning (AI/ML) solutions for automated cell isolation and identification of drug-treated malaria parasites from fluorescent microscope images to prepare them for future AI/ML-based phenotypic analysis.

Objectives:

1. Using a pre-acquired dataset of raw microscope images of malaria parasites, evaluate and optimise AI/ML algorithms to efficiently and accurately extract cells of interest.
2. Determine which AI/ML approach is most appropriate for this problem.

Apply solution to a newly generated dataset.