University of Edinburgh, School of Medicine, Deanery of Clinical Sciences, 47 Little France Crescent, EH164TJ Edinburgh, Scotland.
Institute of Scientific Instruments of the CAS, v. v. i., Královopolská 147, 612 00 Brno, Czech Republic.
Clin Neurophysiol. 2024 May;161:1-9. doi: 10.1016/j.clinph.2024.01.007. Epub 2024 Feb 18.
Interictal biomarkers of the epileptogenic zone (EZ) and their use in machine learning models open promising avenues for improvement of epilepsy surgery evaluation. Currently, most studies restrict their analysis to short segments of intracranial EEG (iEEG).
We used 2381 hours of iEEG data from 25 patients to systematically select 5-minute segments across various interictal conditions. Then, we tested machine learning models for EZ localization using iEEG features calculated within these individual segments or across them and evaluated the performance by the area under the precision-recall curve (PRAUC).
On average, models achieved a score of 0.421 (the result of the chance classifier was 0.062). However, the PRAUC varied significantly across the segments (0.323-0.493). Overall, NREM sleep achieved the highest scores, with the best results of 0.493 in N2. When using data from all segments, the model performed significantly better than single segments, except NREM sleep segments.
The model based on a short segment of iEEG recording can achieve similar results as a model based on prolonged recordings. The analyzed segment should, however, be carefully and systematically selected, preferably from NREM sleep.
Random selection of short iEEG segments may give rise to inaccurate localization of the EZ.
癫痫灶（EZ）的发作间期生物标志物及其在机器学习模型中的应用为改善癫痫手术评估开辟了有前景的途径。目前，大多数研究将其分析仅限于颅内 EEG（iEEG）的短片段。
我们使用 25 名患者的 2381 小时 iEEG 数据，系统地选择跨各种发作间期条件的 5 分钟片段。然后，我们使用在这些单个片段或跨片段内计算的 iEEG 特征测试用于 EZ 定位的机器学习模型，并通过精度-召回曲线下的面积（PRAUC）评估性能。
平均而言，模型的得分为 0.421（机会分类器的结果为 0.062）。然而，片段之间的 PRAUC 差异很大（0.323-0.493）。总体而言，NREM 睡眠得分最高，N2 时的最佳结果为 0.493。当使用所有片段的数据时，该模型的性能明显优于单个片段，除了 NREM 睡眠片段。
基于 iEEG 记录短片段的模型可以达到与基于延长记录的模型相似的结果。然而，应该仔细且系统地选择分析片段，最好来自 NREM 睡眠。
随机选择 iEEG 短片段可能导致 EZ 的定位不准确。
