Artificial intelligence in oncological positron emission tomography: advancing image analysis and interpretation

Area of Science:

• Oncology
• Medical Imaging
• Artificial Intelligence

Background:

• Positron emission tomography (PET) imaging provides crucial functional and metabolic data for cancer patient management, including diagnosis, staging, and treatment evaluation.
• Clinical efficacy of PET imaging can be limited by variations in image quality and quantitative accuracy.
• Artificial intelligence (AI) is emerging as a transformative tool in oncological PET imaging.

Purpose of the Study:

• To review the current advancements and challenges of AI in oncological PET imaging.
• To provide a balanced perspective on AI's role in clinical translation for improved cancer care.

Main Methods:

• Review of recent research and literature on AI applications in oncological PET imaging.
• Analysis of AI's impact on image quality, quantitative metrics, diagnostic accuracy, and prognostic modeling.
• Identification of challenges hindering clinical implementation, including data standardization, explainability, and regulatory frameworks.

Main Results:

• AI significantly improves image quality and quantitative consistency in oncological PET.
• AI demonstrates value in enhancing diagnostic accuracy and prognostic modeling for cancer patients.
• Key challenges for clinical AI integration include data standardization, explainability, and regulatory approval.

Conclusions:

• AI holds substantial promise for advancing oncological PET imaging and personalized cancer care.
• Future progress relies on multimodal integration, federated learning, and probabilistic deep learning.
• Overcoming current challenges is essential for successful clinical translation of AI in PET imaging.