Machine learning redefines precision cancer treatment using smart radiopharmaceuticals

JMIR Publications featured news and perspective articles on technological advances in oncology. “AI-Designed Radiopharmaceuticals: How Machine Learning is Redefining Precision Cancer Treatment,” written by JMIR correspondent Benedette Cuffari, reports on the integration of deep learning and generative AI in radiopharmaceuticals, its impact on accelerating drug design, and how personalized dosimetry can improve patient outcomes.

Drug discovery using AI

Although radiopharmaceutical treatments are highly effective against some types of cancer, their development remains time-consuming and resource-intensive. Deep learning and generative AI models can rapidly identify new targets, predict chemical interactions, and design stable drug candidates. Dr. Kufari spoke to Dr. Sofia Michopoulou, a medical physicist and head of nuclear medicine physics at Southampton University Hospital, who said AI-driven computer simulations could “identify the most promising drug candidates early, reduce the amount of current preclinical work, and make early-stage assessments more focused and efficient.”

Personalized dosimetry and digital twin

The AI ​​model also optimizes dosimetry, which is the calculation of radiation absorbed by tissue to maximize tumor damage while sparing healthy organs. 3D convolutional neural networks can analyze medical images to predict biodistribution, while machine learning can also generate patient-specific digital twins to develop highly personalized treatment plans, Cuffari wrote.

Barriers to clinical implementation

Despite these advances, clinical translation is hampered by the lack of standardized, high-quality datasets to train AI models. Although techniques like federated learning can protect patient confidentiality across hospitals, extensive basic experimental research is still required before the model can be adequately generalized.