Evaluation of the Use of Radiomics in Predicting Treatment Response in Oncology: (Review Artical)
Abstract
Background: Radiomics involves the retrieval of numerical information from medical imaging, with the ability to describe the characteristics of a tumor. The radiomics technique has the ability to create prognostic models for therapy response, which is crucial for the advancement of personalized medicine. Aim of Study: This literature review provides a concise overview and assesses the scientific rigor and reporting stand-ards of radiomics research in predicting treatment response in non-small-cell lung cancer (NSCLC). Methods: An extensive literature search was performed us-ing the PubMed database. The radiomics quality score (RQS), a measure specifically designed for radiomics, was used to evalu-ate the scientific and reporting quality, following the parameters set by TRIPOD. Results: The studies included in the analysis revealed sev-eral predictive markers, including first-, second-, and high-or-der features. These characteristics included kurtosis, grey-level uniformity, and wavelet HLL mean, as well as PET-based met-abolic indicators. The studies exhibited significant variability as a result of variations in patient demographics, cancer stage, treatment methods, duration of follow-up, and radiomics pro- cessing protocols. Conclusion: The use of radiomics research in clinical prac- tice has not yet been implemented. To develop radiomic pre-dictors of response that can be reproduced, it is necessary to make efforts toward standardization and cooperation. In order for radiomic models to be used as a clinical decision-making tool for personalized treatment of patients with NSCLC, it is necessary to verify them externally and assess their effect with-in the therapeutic pathway.