BioMag Laboratory, HUS Medical Imaging Center, Helsinki University Hospital, Helsinki University and Aalto University School of Science, P.O. Box 340, 00029 HUS Helsinki, Finland; Department of Neuroscience and Biomedical Engineering, Aalto University School of Science, Aalto University, P.O. Box 12200, 00760 AALTO, Finland.
BioMag Laboratory, HUS Medical Imaging Center, Helsinki University Hospital, Helsinki University and Aalto University School of Science, P.O. Box 340, 00029 HUS Helsinki, Finland; Department of Neuroscience and Biomedical Engineering, Aalto University School of Science, Aalto University, P.O. Box 12200, 00760 AALTO, Finland.
Clin Neurophysiol. 2023 Sep;153:79-87. doi: 10.1016/j.clinph.2023.06.010. Epub 2023 Jun 30.
Diagnosis of mild traumatic brain injury (mTBI) is challenging despite its high incidence, due to the unspecificity and variety of symptoms and the frequent lack of structural imaging findings. There is a need for reliable and simple-to-use diagnostic tools that would be feasible across sites and patient populations.
We evaluated linear machine learning (ML) methods' ability to separate mTBI patients from healthy controls, based on their sensor-level magnetoencephalographic (MEG) power spectra in the subacute phase (<2 months) after a head trauma. We recorded resting-state MEG data from 25 patients and 25 age-sex matched controls and utilized a previously collected data set of 20 patients and 20 controls from a different site. The data sets were analyzed separately with three ML methods.
The median classification accuracies varied between 80 and 95%, without significant differences between the applied ML methods or data sets. The classification accuracies were significantly higher with ML than with traditional sensor-level MEG analysis based on detecting pathological low-frequency activity.
Easily applicable linear ML methods provide reliable and replicable classification of mTBI patients using sensor-level MEG data.
Power spectral estimates combined with ML can classify mTBI patients with high accuracy and have high promise for clinical use.
尽管轻度创伤性脑损伤（mTBI）的发病率很高，但由于其症状的非特异性和多样性以及结构影像学发现的频繁缺乏，其诊断仍然具有挑战性。需要可靠且易于使用的诊断工具，这些工具在不同地点和患者群体中都可行。
我们评估了线性机器学习（ML）方法在头部外伤后亚急性期（<2 个月）根据其传感器水平脑磁图（MEG）功率谱分离 mTBI 患者与健康对照者的能力。我们记录了 25 名患者和 25 名年龄性别匹配的对照者的静息状态 MEG 数据，并利用来自不同地点的先前收集的 20 名患者和 20 名对照者的数据组进行分析。三个 ML 方法分别对数据集进行了分析。
中位数分类准确率在 80%至 95%之间变化，应用的 ML 方法或数据集之间没有显著差异。与基于检测病理性低频活动的传统传感器水平 MEG 分析相比，ML 的分类准确率明显更高。
易于应用的线性 ML 方法使用传感器水平 MEG 数据提供了 mTBI 患者的可靠和可复制分类。
功率谱估计与 ML 相结合，可以高精度地对 mTBI 患者进行分类，具有很高的临床应用前景。
