Artificial Intelligence Augmented Cerebral Nuclear Imaging

Area of Science:

• Nuclear medicine and neuroscience
• Artificial intelligence in medical imaging
• Advanced data analysis techniques

Background:

• Nuclear neuroimaging techniques like SPECT and PET are crucial for brain disorder diagnosis.
• Current analysis methods can be limited by image quality and data interpretation complexity.
• The integration of AI offers potential solutions to these existing challenges.

Purpose of the Study:

• To explore the current and emerging applications of AI, ML, and DL in nuclear neuroimaging.
• To detail how AI enhances processing, analysis, and interpretation of brain SPECT and PET scans.
• To discuss the future impact of AI on precision medicine and research in nuclear cerebral imaging.

Main Methods:

• Review of current and emerging applications of ML and DL in nuclear neuroimaging.
• Analysis of AI's role in image segmentation, disease classification, and radiomic feature extraction.
• Exploration of DL for image reconstruction, attenuation correction, denoising, and radioligand design.

Main Results:

• AI enables automated segmentation, disease classification, and advanced radiomic feature extraction (1st to 4th order).
• DL improves image quality through reconstruction, pseudo-CT generation for attenuation correction, and denoising.
• Federated learning addresses data security, while generative AI promises solutions for data limitations and workflow efficiencies.

Conclusions:

• AI, ML, and DL are fundamentally transforming nuclear neuroimaging, enhancing diagnostic capabilities and research.
• These technologies improve image quality, data analysis, and research efficiency, paving the way for precision medicine.
• The continued innovation in AI is set to redefine the nuclear neuroimaging landscape and patient care.