Radiomics for Preoperative Assessment of Pituitary Adenoma Consistency with T2-Weighted MRI: A Multicenter Study.
Authors
Affiliations (11)
Affiliations (11)
- Division of Neurosurgery, Department of Medical and Surgical Specialties, Radiological Sciences and Public Health, University of Brescia, Brescia, Italy.
- Department of Medicine, Surgery and Dentistry, University of Salerno, Baronissi, Italy.
- Unit of Otorhinolaryngology - Head and Neck Surgery, ASST Spedali Civili, Department of Surgical and Medical Specialties, Radiological Sciences, and Public Health, University of Brescia, School of Medicine, Brescia, Italy.
- Unit of Neurosurgery, ARNAS Ospedale Civico of Palermo, Palermo, Italy.
- Unit of Neurosurgery, Ospedale Ca' Foncello, Treffviso, Italy.
- Neurosurgery Division, Department of Neuroscience, "Sapienza" University of Rome, Rome, Italy.
- Division of Neurosurgery, Department of Neurosciences, Reproductive and Odontostomatological Sciences, Università degli Studi di Napoli Federico II, Naples, Italy.
- Division of Neurological Surgery, Department of Biotechnology and Life Sciences, University of Insubria-Varese, ASST Sette Laghi, Ospedale di Circolo e Fondazione Macchi, Varese, Italy.
- Facoltà di Medicina e Chirurgia, Università Cattolica del Sacro Cuore, Rome, Italy.
- Division of Neurosurgery, Department of Neuroscience, Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy.
- Department of Advanced Medical and Surgical Sciences, University of Campania "Luigi Vanvitelli," Naples, Italy.
Abstract
Pituitary adenoma (PA) consistency significantly influences the outcomes of endoscopic endonasal surgery. Radiomics represents a promising tool for objective and quantitative assessment using T2-weighted magnetic resonance imaging (MRI). A multicenter retrospective database was collected (2012-2023), including 394 patients with preoperative T2-weighted MRI and histologically confirmed PAs after endoscopic endonasal surgical removal. Tumor segmentation was performed manually on coronal T2-weighted images using ITK-SNAP software. Radiomic features were extracted with Pyradiomics. A 60:40 dataset split was used to train an Extra Trees classifier, and recursive feature elimination was used to select features. Model performance was assessed using sensitivity, specificity, and the area under the curve of receiver operating characteristic (AUC-ROC) curve metrics. From 1,106 extracted radiomic features, 65 were identified as most predictive following variance and correlation filtering. The sensitivity, specificity, and accuracy of the ET classifier were 74%, 74%, and 63% (±10%), respectively. The AUC-ROC curve was 0.59. Despite its moderate accuracy and AUC-ROC curve, the ET model showed promising performance to predict preoperative PA consistency, underlying the power of radiomics-driven models in PA surgical planning.