This systematic review evaluates and summarizes the potential for radiomics, a computational tool that extracts quantitative features from imaging, to predict VS clinical outcomes and assess treatment responsiveness.

Vestibular schwannomas(VS) present a clinical challenge in management decision-making due to their unpredictable growth and potential impact on crucial neurological function. Radiomics involves extracting quantitative, reproducible data from medical images beyond what the human perception and can be used to predict outcomes such as patient survival, tumor growth, response to treatment, and hearing loss, to ultimately inform management.

Studies were extracted by searching PubMed, OVID Medline, and Web of Science databases. Included studies analyzed radiomic features from magnetic resonance imaging(MRI) as independent variables and varied in their methodology to predict clinical outcomes. Studies analyzed associations between radiomic features, pre-procedural clinical features, and post-procedural clinical and radiologic outcomes.

Thirteen retrospective studies met inclusion criteria; eleven of these used machine learning(ML) models to analyze radiomic MRI features. One non-ML study correlated longitudinal tumor volumetric changes with texture features. All segmentation workflows utilized manual or semi-automated approaches to determine the lesion’s region of interest. Models based on pre-procedural imaging demonstrated moderate predictive accuracy by Area Under the Receiver Operating Characteristic curve(AUC=0.66-0.7), while post-procedural models showed strong predictive capacity(AUC=0.75-1.0). One study employed a convolutional neural network evaluating post-operative facial nerve outcomes(AUC=0.89) that outperformed traditional ML models(AUC=0.64-0.85).

Radiomics-based predictive modeling in VS shows promise across a range of clinical outcomes. However, small sample sizes, retrospective designs, and lack of standardization in imaging and modeling hinder its widespread applicability. Addressing these limitations through larger, standardized datasets, consistent modeling approaches, and prospective predictive studies, potentially incorporating deep learning, will be essential to improve generalizability and support clinical integration.