The Evolution of Artificial Intelligence in Nuclear Medicine

Area of Science:

• Nuclear medicine
• Artificial intelligence
• Medical imaging

Background:

• Nuclear medicine continuously evolves, improving disease diagnosis and treatment.
• Artificial intelligence (AI) integration marks a significant advancement in the field.
• AI promises breakthroughs in diagnosis, prognosis, segmentation, image enhancement, and theranostics.

Purpose of the Study:

• To review the advancements and impact of AI in nuclear medicine.
• To highlight the evolution from machine learning to deep learning and generative AI.
• To discuss challenges and future directions for AI in clinical practice.

Main Methods:

• Review of AI applications in nuclear medicine, including machine learning and deep learning.
• Analysis of advancements in diagnostic accuracy, image segmentation, low-dose imaging, and dosimetry.
• Exploration of generative AI for medical language and image processing.

Main Results:

• AI has improved diagnostic accuracy and outcome prediction using machine learning.
• Deep learning models (CNNs, transformers) enhance segmentation, low-dose imaging, and personalized dosimetry.
• Generative AI facilitates complex medical language and image generation and interpretation.

Conclusions:

• AI integration offers significant potential to enhance nuclear medicine diagnosis and treatment.
• Overcoming challenges like data scarcity, heterogeneity, and ethical concerns is crucial for clinical translation.
• Interdisciplinary collaboration is key to broader AI adoption and improved patient outcomes.