Artificial intelligence-based radiomics in bone tumors: Technical advances and clinical application

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Radiomics involves extracting quantitative features from medical images.
• Artificial intelligence (AI) algorithms can process these features to identify complex patterns.
• Bone tumors, including primary and metastatic types, present significant clinical challenges.

Purpose of the Study:

• To review the evolution and opportunities of AI in medical imaging for bone tumors.
• To focus on both hand-crafted and deep learning-based radiomics approaches.
• To summarize current AI-based radiomics applications in primary and metastatic bone tumors.

Main Methods:

• Extraction of predefined mathematical features from medical images (radiomics).
• Application of AI algorithms to analyze radiomic features for pattern recognition.
• Review of existing literature on AI and radiomics in bone tumor management.

Main Results:

• Radiomics models demonstrate high performance in diagnosing and grading bone tumors.
• AI-based radiomics can predict overall survival, metastasis, and recurrence.
• Radiomics aids in therapeutic planning, particularly for neoadjuvant chemotherapy.

Conclusions:

• AI-based radiomics offers innovative solutions for primary and metastatic bone tumors.
• Understanding these emerging approaches is crucial for personalized medicine in oncology.
• Future opportunities lie in further integrating AI-driven radiomics into clinical decision-making for bone tumors.