AI Streamlines Radiopharmaceutical Dosimetry in Prostate Cancer Care
At the 2025 UCSF-UCLA PSMA Conference, Dr. Irène Buvat, Director of the Laboratory of Translational Imaging in Oncology at Institut Curie, presented on the transformative role of artificial intelligence (AI) in simplifying personalized dosimetry for radiopharmaceutical therapy, particularly in prostate cancer treatment.
Dr. Buvat addressed the "catch-22" in clinical dosimetry: its complexity deters widespread clinical use, which in turn limits the accumulation of evidence needed to demonstrate its benefits. She emphasized that AI can break this cycle by streamlining each step of the dosimetry workflow.
Key applications of AI in this context include:
-
Enhancing the accuracy of SPECT imaging.
-
Improving image registration across different time points.
ADVERTISEMENT -
Automating bone marrow segmentation.
ADVERTISEMENT -
Directly calculating dose maps.
-
Reducing acquisition times for imaging.
These advancements aim to make personalized dosimetry more accessible and efficient, potentially leading to better patient outcomes.
However, Dr. Buvat cautioned that AI is not a panacea. While it can reduce variability in dosimetry calculations, it may also introduce biases, especially when protocols are simplified excessively. She stressed the importance of balancing simplification with accuracy to ensure reliable results.
In conclusion, integrating AI into dosimetry workflows holds promise for enhancing the precision and personalization of radiopharmaceutical therapies in prostate cancer, but careful implementation is crucial to maximize benefits and minimize risks.