Artificial intelligence-based motion tracking in cancer radiotherapy: A review

Area of Science:

• Radiation oncology and medical physics.
• Artificial intelligence in healthcare.
• Medical imaging and data analysis.

Background:

• Radiotherapy techniques like VMAT, SRS, SBRT, and proton therapy enhance dose precision.
• Intra-fraction motion management is crucial for verifying tumor position during treatment.
• Artificial intelligence (AI) offers potential for real-time tumor tracking in radiotherapy.

Purpose of the Study:

• To review AI algorithms for tumor motion management in radiotherapy for chest, abdomen, and pelvic regions.
• To summarize existing literature on AI-based tumor tracking in radiation oncology.
• To discuss limitations and propose improvements for AI applications in radiotherapy motion management.

Main Methods:

• Literature review of AI algorithms applied to radiotherapy motion management.
• Analysis of AI applications for tracking tumors in the chest, abdomen, and pelvis.
• Identification of challenges and limitations in current AI-based studies.

Main Results:

• AI demonstrates significant potential for real-time tumor tracking during radiotherapy.
• Various AI algorithms are being explored for motion management across different anatomical sites.
• Several challenges hinder widespread AI implementation, including data bias and workflow complexity.

Conclusions:

• AI-based tumor motion management in radiotherapy is a rapidly developing field.
• Addressing limitations in data, transparency, and workflow is essential for clinical translation.
• Further research and development are needed to optimize AI for precise and reliable radiotherapy delivery.