South Africa Integrates AI-Powered Cancer Treatment Planning in First Clinical Use Outside US

The University of the Free State (UFS) and Universitas Academic Hospital in Bloemfontein have become the first clinical site outside the United States to integrate the Radiation Planning Assistant (RPA), an AI-powered, web-based platform developed at the University of Texas MD Anderson Cancer Center.

The system is designed to streamline one of the most time-consuming steps in cancer care — creating patient-specific radiotherapy treatment plans — by automating key components of the process.

Initially rolled out for cervical cancer, which accounts for the largest proportion of patients receiving radiotherapy at Universitas, the RPA is now being used for other cancers, including breast, head and neck, and primary brain tumors.

“This system holds significant promise for broader application across nearly all major tumor types treated with external beam radiotherapy,” said Dr. William Shaw, deputy manager of UFS’s Department of Medical Physics.

The RPA uses a patient’s planning CT scan as its sole imaging input. Once uploaded via a secure web interface, advanced machine learning algorithms automatically identify tumor volumes and critical normal tissues before generating a full treatment plan. The platform is intended to support both high- and low-resource settings, improving the quality, consistency, and timeliness of care while reducing strain on specialized staff.

Shaw said the technology can lead to faster treatment starts, fewer interruptions, reduced side effects, and better overall outcomes. Standardized plans also help reduce variability between institutions and enable consistent protocols for multi-center research.

Since its introduction in Bloemfontein, the RPA has been used to treat nearly 50 patients. The positive results have prompted MD Anderson to extend access to other radiotherapy centers in South Africa, with plans for expansion to other countries.

“This represents a major advancement for oncology services regionally and nationally,” Shaw said. “It’s a shift from research collaboration to real-world application, where AI improves access to safe, high-quality cancer care.”

The initiative is supported by the Nuclear Technologies in Medicine and the Biosciences Initiative, managed by the South African Nuclear Energy Corporation and funded by the Technology Innovation Agency.