Incremental benefits of size-zone matrix-based radiomics features for the prognosis of lung adenocarcinoma: advantage of spatial partitioning on tumor evaluation

This study investigates the incremental benefits of size-zone matrix (SZM) features in prognostic models for lung adenocarcinoma (ADC). A total of 298 patients' pretreatment CT images were analyzed using fivefold cross-validation. The study compared overall survival risk models using SZM features with conventional radiomics and clinical variable-based models. Dedicated lung imaging software (AVIEW, Coreline Soft) was used for tumor segmentation. Seven risk models were evaluated using the hazard ratio (HR) on the test fold. The clinical variable risk model showed an HR of 2.739. Incorporating SZM features with the radiomics signature improved the HR to 4.034, compared to 3.439 with the radiomics signature alone. Adding clinical variables to the radiomics and SZM model further improved the HR to 6.524, demonstrating the incremental benefits of SZM features. The study confirms that SZM features enhance prognostic accuracy for lung ADC, even when combined with clinical variables.