Will AI help treat cancer? RSU researcher on new possibilities in cancer diagnosis

Researchers from Rīga Stradiņš University (RSU), Riga Technical University (RTU), and the University of Latvia (LU) are collaborating on a study aimed at improving the accuracy and efficiency of cancer diagnosis. The project focuses on developing AI-driven solutions for diagnosing spinal metastases, enhancing both detection speed and precision. It involves creating an application capable of identifying both lytic and sclerotic types of metastases in the vertebrae, helping doctors reduce diagnostic time and minimize human error.

Edgars Edelmers with his colleagues from the project. Photo: Courtesy of Edelmers

Project manager and RSU researcher Edgars Edelmers explains:

‘This project demonstrates how interdisciplinary collaboration between engineers, medical professionals, and researchers can lay the foundation not only for significant scientific advancements but also for practical solutions with strong commercialisation potential.

I firmly believe that multilateral cooperation is essential for turning innovative ideas into tangible solutions. This approach enables us to gain a deeper understanding of clinical processes and develop techniques that could form the basis of modern medical products in the future.

Right now, we are creating a practical and sustainable solution that will significantly accelerate cancer diagnostics, reduce the risk of human error, and advance modern medical practice in Latvia.

At the same time, this project will drive broader changes in the healthcare system, emphasizing the importance of timely and accurate diagnostics, which can be achieved through modern technology and data-driven decision-making.’

Results of the study after four months

The study is being conducted in collaboration with Riga East University Hospital (RAKUS). After four months of development, the AI algorithm achieved 68% accuracy in detecting lytic metastases and 57% accuracy in detecting sclerotic metastases. To further enhance diagnostic quality, the project team plans to expand the system’s capabilities by integrating magnetic resonance imaging (MRI) and refining the algorithm.

Faster and more effective cancer diagnosis planned for the future

The project is funded by the Fundamental and Applied Research Programme (FLPP) and will continue into 2025 to improve accuracy and clinical applicability. The results of this initiative offer hope for faster and more effective cancer diagnostics in the future.