More articles from Adult Brain
- Convolutional Neural Network for Automated FLAIR Lesion Segmentation on Clinical Brain MR Imaging
This convolutional neural network was retrospectively trained on 295 brain MRIs to perform automated FLAIR lesion segmentation. Performance was evaluated on 92 validation cases using Dice scores and voxelwise sensitivity and specificity, compared with radiologists' manual segmentations. The authors' model demonstrated accurate FLAIR lesion segmentation performance (median Dice score, 0.79) on the validation dataset across a large range of lesion characteristics. Across 19 neurologic diseases, performance was significantly higher than existing methods (Dice, 0.56 and 0.41) and approached human performance (Dice, 0.81).
- One-Stop Management with Perfusion for Transfer Patients with Stroke due to a Large-Vessel Occlusion: Feasibility and Effects on In-Hospital Times
The authors report the first 15 consecutive transfer patients with stroke with externally confirmed large-vessel occlusions who underwent flat panel detector CT perfusion and thrombectomy in the same room. Preinterventional imaging consisted of noncontrast flat panel detector CT and flat panel detector CT perfusion, acquired with a biplane angiography system. The flat panel detector CT perfusion was used to reconstruct a flat panel detector CT angiography to confirm the large-vessel occlusions. After confirmation of the large-vessel occlusion, the patient underwent mechanical thrombectomy. Fifteen transfer patients underwent flat panel detector CT perfusion and were treated with mechanical thrombectomy from June 2017 to January 2019. The median time from symptom onset to admission was 241 minutes. Median door-to-groin time was 24 minutes. Compared with 23 transfer patients imaged with multidetector CT, time was reduced significantly (24 minutes versus53 minutes).