Related Experiment Video
Updated: Jul 16, 2026
Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics
Published on: January 8, 2018
Artificial Intelligence-Driven Radiomics in Head and Neck Cancer: Current Status and Future Prospects
Rasheed Omobolaji Alabi1, Mohammed Elmusrati2, Ilmo Leivo3
1Research Program in Systems Oncology, Faculty of Medicine, University of Helsinki, Helsinki, Finland; Department of Industrial Digitalization, School of Technology and Innovations, University of Vaasa, Vaasa, Finland.
Artificial intelligence-based radiomics (AI-based radiomics) enhances head and neck cancer (HNC) management by extracting quantitative image features for improved diagnosis and prognosis. Challenges in implementation are being addressed for personalized oncology.
Area of Science:
- * Medical imaging analysis
- * Quantitative feature extraction
- * Artificial intelligence in oncology
Background:
- * Radiomics leverages medical images for quantitative feature extraction to characterize patient phenotypes.
- * AI techniques combined with radiomics improve diagnostic accuracy and clinical outcome prediction.
Purpose of the Study:
- * Review AI-based radiomics applications in head and neck cancer (HNC) management.
- * Explore AI-based radiomics workflow for personalized oncology in HNC.
- * Examine implementation challenges and solutions for AI-based radiomics in clinical oncology.
Main Methods:
- * Systematic literature search using PRISMA guidelines across multiple databases.
- * Quality and bias assessment of included studies using TRIPOD and PROBAST tools.
Main Results:
- * 45 studies met inclusion criteria, identifying AI-based radiomics as an ancillary tool for HNC decision-making.
- * Applications include cancer diagnosis (staging, grading, classification) and prognosis (treatment response, recurrence, metastasis, survival).
- * Key challenges include data imbalance, feature engineering, model generalizability, multi-modal fusion, and interpretability.
Conclusions:
- * AI-based radiomics offers quantitative insights beyond human visual interpretation, addressing interobserver variability.
- * Potential to revolutionize HNC oncology by enabling personalized, high-quality, and cost-effective patient care.
