Department of Neuroscience, Central Clinical School, Monash University, Melbourne, Victoria, Australia.
Department of Neurology, Alfred Health, Melbourne, Victoria, Australia.
J Clin Neurophysiol. 2024 Nov 1;41(7):618-624. doi: 10.1097/WNP.0000000000001039. Epub 2023 Oct 30.
Despite availability of commercial EEG software for automated epileptiform detection, validation on real-world EEG datasets is lacking. Performance evaluation of two software packages on a large EEG dataset of patients with genetic generalized epilepsy was performed.
Three epileptologists labelled IEDs manually of EEGs from three centres. All Interictal epileptiform discharge (IED) markings predicted by two commercial software (Encevis 1.11 and Persyst 14) were reviewed individually to assess for suspicious missed markings and were integrated into the reference standard if overlooked during manual annotation during a second phase. Sensitivity, precision, specificity, and F1-score were used to assess the performance of the software packages against the adjusted reference standard.
One hundred and twenty-five routine scalp EEG recordings from different subjects were included (total recording time, 310.7 hours). The total epileptiform discharge reference count was 5,907 (including spikes and fragments). Encevis demonstrated a mean sensitivity for detection of IEDs of 0.46 (SD 0.32), mean precision of 0.37 (SD 0.31), and mean F1-score of 0.43 (SD 0.23). Using the default medium setting, the sensitivity of Persyst was 0.67 (SD 0.31), with a precision of 0.49 (SD 0.33) and F1-score of 0.51 (SD 0.25). Mean specificity representing non-IED window identification and classification was 0.973 (SD 0.08) for Encevis and 0.968 (SD 0.07) for Persyst.
Automated software shows a high degree of specificity for detection of nonepileptiform background. Sensitivity and precision for IED detection is lower, but may be acceptable for initial screening in the clinical and research setting. Clinical caution and continuous expert human oversight are recommended with all EEG recordings before a diagnostic interpretation is provided based on the output of the software.
尽管有用于自动癫痫样放电检测的商业 EEG 软件，但缺乏对真实 EEG 数据集的验证。对遗传性全面性癫痫患者的大型 EEG 数据集上的两种软件包进行了性能评估。
三位癫痫专家分别对来自三个中心的 EEG 进行了 IED 手动标记。两位商业软件（Encevis 1.11 和 Persyst 14）预测的所有发作间期癫痫样放电（IED）标记均单独进行了审查，以评估可疑的遗漏标记，并在手动注释过程中在第二阶段忽略了这些标记，如果在手动注释过程中忽略了这些标记，则将其整合到参考标准中。使用敏感性、精度、特异性和 F1 分数来评估软件包对调整后的参考标准的性能。
共纳入 125 例不同受试者的常规头皮 EEG 记录（总记录时间 310.7 小时）。总癫痫样放电参考计数为 5907 次（包括棘波和棘慢波）。Encevis 检测 IED 的平均敏感性为 0.46（SD 0.32），平均精度为 0.37（SD 0.31），平均 F1 分数为 0.43（SD 0.23）。使用默认的中设置，Persyst 的敏感性为 0.67（SD 0.31），精度为 0.49（SD 0.33），F1 分数为 0.51（SD 0.25）。代表非 IED 窗口识别和分类的平均特异性为 Encevis 为 0.973（SD 0.08），Persyst 为 0.968（SD 0.07）。
自动软件对非癫痫样背景的检测具有高度特异性。IED 检测的敏感性和精度较低，但在临床和研究环境中进行初始筛查时可能是可以接受的。在提供基于软件输出的诊断解释之前，建议对所有 EEG 记录进行临床谨慎和持续的专家人工监督。
