7–11 Jul 2025
University of the Witwatersrand, Johannesburg
Africa/Johannesburg timezone
Registration open until 20 May 2025

Improving isotope production using machine learning techniqures at iThemba Labs

Not scheduled
2h 50m
Solomon Mahlangu House (University of the Witwatersrand, Johannesburg)

Solomon Mahlangu House

University of the Witwatersrand, Johannesburg

Poster Presentation Track F - Applied Physics Poster Session

Speaker

Donald Ngobeni

Description

The production of high-quality radioisotopes is essential for nuclear medicine, scientific research and various industries. These radioisotopes are produced using advanced particle accelerators at iThemba LABS and have become the leading organization for radioisotope production. The process requires precise control over the beam parameters, the target material, and the chemical processing. However, small changes in beam parameters, target material, or chemical processing can significantly impact the amount and quality of isotopes produced. To improve on the production, we investigate the use of machine learning (ML) techniques to make the production more efficient and reliable. These techniques will mainly focus on the intelligent knowledge systems, to optimize production pathways using historical production records and real-time beam related data to enhance isotope yield and reduce inefficiencies.

Apply for student award at which level: MSc
Consent on use of personal information: Abstract Submission Yes, I ACCEPT

Primary author

Donald Ngobeni

Co-authors

Bruce Mellado (University of the Witwatersrand and iThemba LABS) Edward Nkadimeng (NRF-iThemba LABS) Mukesh Kumar (School of Physics and Institute for Collider Particle Physics, University of the Witwatersrand)

Presentation materials

There are no materials yet.