Hogan Robert, Mathieson Sean R, Luca Aurel, Ventura Soraia, Griffin Sean, Boylan Geraldine B, O'Toole John M
CergenX Ltd, Dublin, Ireland.
INFANT Research Centre, University College Cork, Cork, Ireland.
NPJ Digit Med. 2025 Jan 8;8(1):17. doi: 10.1038/s41746-024-01416-x.
Neonatal seizures require urgent treatment, but often go undetected without expert EEG monitoring. We have developed and validated a seizure detection model using retrospective EEG data from 332 neonates. A convolutional neural network was trained and tested on over 50,000 hours (n = 202) of annotated single-channel EEG containing 12,402 seizure events. This model was then validated on two independent multi-reviewer datasets (n = 51 and n = 79). Increasing data and model size improved performance: Matthews correlation coefficient (MCC) and Pearson's correlation (r) increased by up to 50% (15%) with data (model) scaling. The largest model (21m parameters) achieved state-of-the-art on an open-access dataset (MCC = 0.764, r = 0.824, and AUC = 0.982). This model also attained expert-level performance on both validation sets, a first in this field, with no significant difference in inter-rater agreement when the model replaces an expert (∣Δκ∣ < 0.094, p > 0.05).
新生儿癫痫发作需要紧急治疗，但如果没有专业的脑电图监测，往往难以被发现。我们利用332名新生儿的回顾性脑电图数据开发并验证了一种癫痫发作检测模型。在超过50,000小时（n = 202）包含12,402次癫痫发作事件的单通道注释脑电图上对一个卷积神经网络进行了训练和测试。然后在两个独立的多评审员数据集（n = 51和n = 79）上对该模型进行了验证。增加数据和模型规模可提高性能：随着数据（模型）规模扩大，马修斯相关系数（MCC）和皮尔逊相关系数（r）分别提高了50%（15%）。最大的模型（2100万个参数）在一个开放获取数据集上达到了先进水平（MCC = 0.764，r = 0.824，AUC = 0.982）。该模型在两个验证集上也达到了专家水平的性能，这在该领域尚属首次，当模型取代专家时，评分者间一致性没有显著差异（∣Δκ∣ < 0.094，p > 0.05）。
