Pial arteriovenous fistulas (PAVFs) are rare high-flow vascular lesions requiring precise characterization of the fistulous angioarchitecture before treatment. We report a pediatric PAVF in which an enlarged artery adjacent to the fistula initially appeared suspicious for additional fistulous supply. Superselective angiography demonstrated normal cortical arterial arborization and parenchymal perfusion without early venous drainage, confirming that the enlarged vessel did not participate directly in the arteriovenous shunt. Selective coil embolization achieved complete immediate occlusion of the fistulous communication while preserving the adjacent normal cortical arterial supply. This case illustrates that arterial enlargement alone does not establish participation in a PAVF and highlights an angiographic pitfall relevant to treatment planning.

Hyperammonemic encephalopathy (HE) is a severe complication of sepsis-induced liver injury (SILI), often associated with sepsis. However, its neuroimaging manifestations remain poorly understood. The aim of this pilot case series is to describe the brain MRI findings in patients with sepsis and HE.

Predicting functional outcomes in acute ischemic stroke (AIS) patients after endovascular thrombectomy (EVT) remains challenging due to the complex and dynamic nature of disease progression. Existing methods typically rely on single-stage imaging or limited multimodal fusion, failing to fully capture disease progression and cross-modal complementarity. To address this issue, we propose a multimodal deep learning framework that integrates multi-stage imaging and clinical notes for 90-day modified Rankin Scale (mRS) prediction. The proposed model leverages a shared foundation model with Low-Rank Adaptation to encode pre- and post-treatment images. It further introduces a shared feature extraction module and a stage-specific feature extraction module to disentangle shared and stage-specific representations, enabling effective modeling of disease progression across stages. In addition, a bidirectional cross-attention module is employed to enhance semantic alignment between imaging features and clinical notes. Experiments on multi-stage NCCT and MRI datasets demonstrate that the proposed method consistently outperforms single-modality baselines, achieving AUCs of 0.9147 and 0.9479, respectively. Ablation studies further validate the effectiveness of both the multi-stage interaction design and the multimodal fusion strategy. These results suggest that integrating pre- and post-treatment imaging with clinical notes can improve prognostic performance, providing a practical solution for individualized outcome prediction in AIS patients following EVT.