Artificial Intelligence in CT and MR Imaging for Oncological Applications.

Area of Science:

• Oncologic Imaging
• Artificial Intelligence in Radiology
• Medical Image Analysis

Background:

• Cancer patient management heavily relies on imaging modalities like computed tomography (CT) and magnetic resonance imaging (MRI).
• These modalities offer high-resolution anatomical and physiological insights crucial for diagnosis and treatment monitoring.
• The integration of artificial intelligence (AI) presents new opportunities and challenges in oncologic imaging.

Purpose of the Study:

• To summarize recent advancements in AI applications for CT and MRI in oncology.
• To address the benefits, challenges, and opportunities of AI in oncologic imaging.
• To highlight the need for robust quantitative imaging metrics and collaborative efforts.

Main Methods:

• Review of recent AI applications in CT and MRI for oncologic imaging.
• Illustration of challenges and solutions using novel methods like image synthesis, auto-segmentation, and image reconstruction.
• Examples provided from lung CT and abdomen, pelvis, and head and neck MRI.

Main Results:

• AI offers significant potential for improving cancer care through advanced imaging analysis.
• Key challenges include clinical integration, data accuracy assessment, and ensuring research integrity.
• Novel methods demonstrate promise in addressing these challenges, particularly in image synthesis and segmentation.

Conclusions:

• AI in oncologic CT and MRI holds great promise for personalized cancer patient management.
• Overcoming challenges requires robust validation of imaging biomarkers, data sharing, and interdisciplinary collaboration.
• Embracing quantitative metrics beyond lesion size is essential for advancing AI in oncology.