16 June 2025 in Computerized Medical Imaging and Graphics
This study focuses on artificial intelligence-driven classification of glioma and Ki-67 leveling using T2w-FLAIR MRI, exploring the association of Ki-67 biomarkers with deep learning (DL) features through explainable artificial intelligence (XAI) and SHapley Additive exPlanations (SHAP). This IRB-approved study included 101 patients with glioma brain tumor acquired MR images with the T2W-FLAIR sequence. We extracted DL bottleneck features using ResNet50 from glioma MR images. Principal component analysis (PCA) was deployed for dimensionality reduction. XAI was used to identify potential features. The XGBosst classified the histologic grades of the glioma and the level of Ki-67. We integrated potential DL features with patient demographics (age and sex) and Ki-67 biomarkers, utilizing SHAP to determine the model's essential features and interactions. Glioma grade classification and Ki-67 level predictions achieved overall accuracies of 0.94 and 0.91, respectively. It achieved precision scores of 0.92, 0.94, and 0.96 for glioma grades 2, 3, and 4, and 0.88, 0.94, and 0.97 for Ki-67 levels (low: 5%≤Ki-6720%). Corresponding F1-scores were 0.95, 0.88, and 0.96 for glioma grades and 0.92, 0.93, and 0.87 for Ki-67 levels. SHAP analysis further highlighted a strong association between bottleneck DL features and Ki-67 biomarkers, demonstrating their potential to differentiate glioma grades and Ki-67 levels while offering valuable insights into glioma aggressiveness. This study demonstrates the precise classification of glioma grades and the prediction of Ki-67 levels to underscore the potential of AI-driven MRI analysis to enhance clinical decision-making in glioma management.
1 November 2024 in Surgical Neurology International
Background: The concurrent presentation of meningioma and intracranial aneurysm (IA) poses diagnostic and therapeutic challenges, with no standardized management protocol available. This study aims to address this through an updated systematic review, delineating optimal strategies for managing this dual pathology.Methods: A systematic review was conducted across PubMed, Web of Science, and Embase databases. Articles were screened independently by two reviewers. Treatment strategies and patient outcomes were comprehensively analyzed to formulate a treatment framework based on several characteristics. In addition, one concurrent meningioma and IA case from our institution was presented.Results: A total of 69 articles comprising 115 patients were included in the study. The cohort exhibited a female predominance (80%) with a mean age of 56 (±13) years. Meningiomas were primarily localized to the frontotemporal and sellar regions, while aneurysms favored the anterior circulation – notably, 16.5% of cases presented with ruptured aneurysms. Management strategies varied based on the spatial relationship between lesions and aneurysm rupture status. In unruptured cases, 34% underwent a single craniotomy for simultaneous resection of both pathologies, while endovascular intervention was favored when the IA originated from an artery feeding the meningioma (73%). Remarkably, postoperative aneurysm rupture occurred in 33% of cases managed solely through tumor resection (range 0–30 days postop).Conclusion: This study proposes a comprehensive treatment algorithm to guide neurosurgeons in managing concurrent meningioma and IA cases. By considering individual patient intricacies, the feasibility of simultaneous management, aneurysm rupture risk, and symptomatology, this framework is a valuable tool for clinical decision-making in these complex scenarios.