PLoS One. 2026 Sep 23;21(9):e0358945. doi: 10.1371/journal.pone.0358945. eCollection 2026.
ABSTRACT
Intracranial aneurysms are dilatations of cerebral arteries that may rupture, leading to subarachnoid hemorrhage and high mortality. Endovascular coiling is widely used to prevent rupture; however, many treated aneurysms experience recanalization, which often necessitates follow-up interventions or surgical clipping. Despite its clinical importance, the segmentation and morphological characterization of recanalized coiled aneurysms remain largely unexplored. In this study, we introduce an automated framework for the segmentation of recanalized coiled aneurysms from three-dimensional rotational angiography (3D-RA) images and the extraction of morphological descriptors relevant to surgical planning. The method is based on the nnU-Net architecture, trained to segment the vascular tree, coil mass, and recanalized lumen. Quantitative postprocessing enables the computation of clinically relevant parameters, including coil-neck distance, recanalization volume, and maximum neck diameter, which contribute to the assessment of clipping feasibility. Preliminary results show encouraging segmentation performance and consistent estimation of morphological parameters across the evaluated cases, supporting the potential of the proposed framework for preoperative assessment of recanalized coiled aneurysms.
PMID:42776969 | DOI:10.1371/journal.pone.0358945