- 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).
- Long-Term Results and Follow-Up Examinations after Endovascular Embolization for Unruptured Cerebral Aneurysms
The appropriate period of follow-up examinations after endovascular embolization for cerebral aneurysms using time-of-flight MR angiography is not well-known. Between April 2006 and March 2011, one hundred forty-eight unruptured aneurysms were treated with endovascular coil embolization. Among them, the authors investigated 116 unruptured aneurysms, which were followed up for >5 years. Time-of-flight MRA was performed at 1 day, 3–6 months, 1 year after the procedure, and every year thereafter. The mean follow-up period was 7.0 years. Recanalization was observed in 19 (16.3%) aneurysms within 2 years. Among them, retreatment was performed in 8 (6.8%) aneurysms. No recanalization was detected in any aneurysms that had been stable in the first 2 years after embolization. They conclude that aneurysms in which recanalization was not observed within 2 years after endovascular coil embolization were stable during a mean follow-up of 7 years. This result may be helpful in considering the appropriate span or frequency of follow-up imaging for embolized cerebral aneurysms.