Automated and AI-driven analysis techniques in PET brain imaging

Positron emission tomography (PET) brain imaging is a pivotal tool in neuroscience, enabling precise visualization and quantification of physiological processes in the human brain. However, traditional analysis of PET brain images relies heavily on manual workflows and expert interpretation, which can be both time-consuming and susceptible to inter- and intra-observer variability. The need for more robust, reproducible, and efficient analysis techniques has prompted considerable interest in automation and artificial intelligence (AI)-driven approaches. Recent advancements in machine learning, particularly deep learning, have demonstrated the potential to automate complex image processing tasks, improve diagnostic accuracy, and extract clinically relevant biomarkers from PET data. Despite these breakthroughs, integrating AI into clinical practice faces considerable hurdles, including the need for large annotated datasets, generalizability across diverse populations, and transparency in algorithmic decision-making. Existing studies have begun to show promising results in automated lesion detection, brain segmentation, and quantitative analytics, yet many challenges remain unaddressed, such as interpretability, validation against gold-standard methods, and regulatory acceptance. As the neuroscience community embraces these new tools, there is an urgent need to systematically evaluate the impact of AI and automated techniques on both research and clinical outcomes in PET brain imaging.This Research Topic aims to bring together cutting-edge research, critical reviews, and innovative methodologies that harness automation and artificial intelligence to enhance PET brain image analysis. The primary objectives are to foster a deeper understanding of how AI-driven tools can improve diagnostic workflows, facilitate reproducible research, and enable new insights into brain function and disease. Key questions include: What are the most promising AI and automation strategies for PET analysis? How can these systems be validated and safely integrated into clinical and research settings? What new challenges arise regarding data standardization, transparency, and bias when deploying these technologies?To gather further insights in the evolving landscape of AI and automation in PET brain imaging, this Research Topic welcomes submissions exploring technological, methodological, and clinical perspectives while delineating limitations regarding non-AI-based or non-automated methods. We welcome articles addressing, but not limited to, the following themes:o Development and validation of automated image processing pipelines for PET brain datao Machine learning and deep learning approaches for PET brain image interpretationo Integration of PET data with other imaging modalities using AI techniqueso Evaluation of reproducibility, accuracy, and clinical impact of AI-driven PET analysiso Challenges in data standardization, algorithm transparency, and model generalizabilityo Regulatory and ethical considerations in deploying automated AI tools for brain imaging Topic editors: Cláudia Régio Brambilla, Xin Di Source: https://www.frontiersin.org/research-topics/79778/automated-and-ai-driven-analysis-techniques-in-pet-brain-imaging