| Sumario: | Simple Summary: Inverted papillomas are a type of tumour that grows in the nose and nearby sinuses. Although these tumours are usually not cancerous, they can behave aggressively and sometimes change into cancer. It is difficult for doctors to know in advance which of these tumours might turn into cancer. This study looked at a new method called radiomics, which uses computer programs to analyse medical images, such as MRI scans, in great detail. These programs can find patterns in the images that may not be visible to the human eye. We reviewed five research studies to see if radiomics could help doctors predict which tumours are more likely to become cancer. The results showed that radiomics models were able to do this with a high level of accuracy, especially when used together with clinical information. This means that radiomics could become a helpful, non-invasive tool for planning treatment and follow-up care. However, more research is needed to test these models in larger and more varied patient groups. If confirmed, this approach could help personalise care, avoid unnecessary treatments, and improve outcomes for patients with these tumours. Introduction: Sinonasal inverted papilloma is a benign but aggressive tumour of the sinonasal tract. It has the potential for malignant transformation into sinonasal squamous cell carcinoma. Radiomics, which involves the extraction and analysis of quantitative imaging features, has emerged as a promising tool in prediction of tumour behaviour. This systematic review aims to critically evaluate the current literature on the application of radiomics in predicting the malignant transformation of sinonasal inverted papilloma. Methods: A comprehensive literature search was conducted across Medline (via PubMed), EMBASE, and Web of Science. Studies investigating the use of radiomics to predict malignant transformation in sinonasal inverted papilloma were selected based on predefined inclusion criteria. Methodological quality and risk of bias were assessed using the QUADAS-2 tool and the Radiomics Quality Score (RQS). Results: Five studies were included, encompassing 837 participants. All studies were retrospective and utilised MRI-based radiomics in the construction of their models. The radiomic models demonstrated satisfactory predictive performance. The median AUCs across the included studies were 0.954 (range: 0.901–0.987) in the training set and 0.914 (range: 0.8–0.989) in the validation set. Conclusions: This systematic review highlights the potential of radiomics as a predictive tool for the malignant transformation of sinonasal inverted papilloma. Radiomics shows promise as a non-invasive adjunct for clinical decision-making. However, further research is needed to standardise methodologies and validate these findings in larger multicentre cohorts.
|
|---|