Nomogram to predict 1-year cognitive decline after stent placement for unruptured intracranial aneurysms.

New iatrogenic cerebral infarcts (NICIs) are common findings on diffusion-weighted magnetic resonance imaging (DWI) following stent placement for unruptured intracranial aneurysms (UIAs) and may contribute to cognitive decline (CD). Using posttreatment DWI and clinical features, we developed a deep learning radiomics nomogram (DLRN) to predict 1-year CD in NICI patients. In a multicenter cohort of 526 patients, CD rates at 1 year were 18.7% (47/251) in the training cohort, 19.8% (33/167) in the external validation cohort, and 20.4% (22/108) in the prospective cohort. The DLRN achieved excellent predictive performance with areas under the curve of 0.975, 0.942, and 0.841 for the respective cohorts. Calibration and decision curve analyses confirmed its reliability and clinical utility. This tool could facilitate early identification of high-risk patients, enabling timely, and tailored interventions to protect cognitive function.

Duke Scholars

Author David Hasan Neurosurgery