Radiomics is a relatively new field and consists in extracting a large amount of data from high-resolution medical imaging in order to create diagnostic and prognostic models with applications in clinical practice. Oncology and radiation oncology benefit maximally from the potential of radiomics due to the intensive use of medical imaging for staging and quantication the treatment response. Widely used in medical imaging for diagnostic potential, radiomics and other artificial intelligence (AI) methods, including deep learning, can offer new horizons to the clinician by anticipating the response to oncological treatment, but also by optimally selecting the cases that will respond to a treatment or another, being thus feasible a stratification of the multimodal therapy. The design of treatment escalation or de-escalation strategies for the purpose of therapy optimization could be an indirect consequence of a multi-omics biomarkers approach including radiomics. Head and neck cancers represent a therapeutic challenge both due to the high rate of loco-regional relapses, but also due to the complexity of non-surgical oncological treatment associated with toxicities that often affect QoL. The essential role of radiotherapy in locally advanced stages of the disease make head and neck squamous cell carcinoma (HNSCC) an ideal candidate for the large-scale implementation of AI and in particular of deep learning and radiomics in the prediction of response to treatment and the optimization of therapeutic ratio. Complex models including both clinical, dosimetric and radiomic parametrics seem to have an increased accuracy compared to purely radiomic models. Radiomics can also be a potential radioresistance biomarker to guide the sub-volumes in which the irradiation dose should be escalated
Alan : Sağlık Bilimleri
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|