Home » News » HeAL partners with the Africa Centre to investigate AI techniques for modeling and predicting HIV drug resistance

HeAL partners with the Africa Centre to investigate AI techniques for modeling and predicting HIV drug resistance

The Health Architecture Lab (HeAL) was part of a UKZN consortium, led by Professors Tulio de Oliveira and Frank Tanser from UKZN’s Africa Centre for Health and Population studies, which was recently awarded a R16.5 million grant to fight HIV drug resistance and transmission in rural KwaZulu-Natal. The grant, funded by the Medical Research Council (MRC) Flagship Projects is part of the MRC’s strategy to sustain vibrant medical research in South Africa.

Researchers from the HeAL and KRR will aim to use machine learning and knowledge representation techniques to build prediction models for HIV drug resistance and treatment and knowledge discovery and extraction from clinical and genomic databases. Software prototypes may result from the research but would need further development through the lab’s external collaborators, such as Jembi Health Systems. Previous work carried out by Mr Pascal Brandt, as part of his MSc research in the HeAL, focused on the application of machine learning techniques to the analysis of patient demographic and treatment data as individual predictors of the development of HIV drug resistance and as a supplement to the more expensive genotyping diagnostic. The depth and scope of the analysis can be increased and extended to include other variables and predictors, including individual drugs and epitopes as well as including public health parameters of interest, such as cost. In addition, the machine learning techniques will be generally useful to other data analysis tasks within the multi-disciplinary consortium.

For the full story see:
http://ndabaonline.ukzn.ac.za/UkzndabaStory/NdabaOnline-Vol1-Issue10/Mul...