The field of neuroimaging and brain tumor analysis is rapidly evolving, with a focus on developing innovative methods for accurate diagnosis, treatment planning, and patient care. Recent studies have explored the use of deep learning models, graph-based approaches, and multimodal fusion techniques to improve the analysis of brain tumors and neurological diseases. These advances have the potential to enable more personalized and effective treatment strategies, and to enhance our understanding of the underlying biology of brain tumors. Noteworthy papers in this area include: LV-Net, which introduces a novel framework for producing individualized 3D lateral ventricle meshes from brain MRI, demonstrating superior reconstruction accuracy and delivering more reliable shape descriptors. FoundBioNet, which proposes a foundation-based biomarker network for noninvasive prediction of IDH mutation status from multi-parametric MRI, achieving high predictive accuracy and outperforming baseline approaches. Fusion-Based Brain Tumor Classification, which presents an ensemble-based deep learning framework for classifying brain tumors, integrating explainable AI and rule-based reasoning to enhance transparency and clinical trust. SAGCNet, which develops a spatial-aware graph completion network for missing slice imputation in population CMR imaging, demonstrating superior performance in synthesizing absent CMR slices. Automatic and standardized surgical reporting for central nervous system tumors, which introduces a comprehensive pipeline for postsurgical reporting, enabling robust automated segmentation, MR sequence classification, and standardized report generation. Hi-SMGNN, which proposes a hierarchical framework for predicting genotype of glioma, integrating structural and morphological connectomes from regional to modular levels, and demonstrating improved robustness and effectiveness. Multimodal Sheaf-based Network, which develops a novel sheaf-based framework for structure-aware and consistent fusion of MRI and histopathology data, contributing to the development of virtual biopsy tools for rapid diagnostics. Automated Segmentation of Coronal Brain Tissue Slabs, which presents a deep learning model for automating the segmentation of tissue from photographs, achieving high performance and approaching inter-/intra-rater levels.