Classification of EEG Signals Reveals a Focal Aftereffect of 10 Hz Motor Cortex Transcranial Alternating Current Stimulation.

Transcranial alternating current stimulation (tACS) modulates oscillations in a frequency- and location-specific manner and affects cognitive and motor functions. This effect appears during stimulation as well as "offline," following stimulation, presumably reflecting neuroplasticity. Whether tACS produces long-lasting aftereffects that are physiologically meaningful, is still of current debate. Thus, for tACS to serve as a reliable method for modulating activity within neural networks, it is important to first establish whether "offline" aftereffects are robust and reliable. In this study, we employed a novel machine-learning approach to detect signatures of neuroplasticity following 10-Hz tACS to two critical nodes of the motor network: left motor cortex (lMC) and right cerebellum (rCB). To this end, we trained a classifier to distinguish between signals following lMC-tACS, rCB-tACS, and sham. Our results demonstrate better classification of electroencephalography (EEG) signals in both theta (θ, 4-8 Hz) and alpha (α, 8-13 Hz) frequency bands to lMC-tACS compared with rCB-tACS/sham, at lMC-tACS stimulation location. Source reconstruction allocated these effects to premotor cortex. Stronger correlation between classification accuracies in θ and α in lMC-tACS suggested an association between θ and α efffects. Together these results suggest that EEG signals over premotor cortex contains unique signatures of neuroplasticity following 10-Hz motor cortex tACS.

Support Vector Machine Classifiers Show High Generalizability in Automatic Fall Detection in Older Adults.

Falls are a major cause of morbidity and mortality in neurological disorders. Technical means of detecting falls are of high interest as they enable rapid notification of caregivers and emergency services. Such approaches must reliably differentiate between normal daily activities and fall events. A promising technique might be based on the classification of movements based on accelerometer signals by machine-learning algorithms, but the generalizability of classifiers trained on laboratory data to real-world datasets is a common issue. Here, three machine-learning algorithms including Support Vector Machine (SVM), k-Nearest Neighbors (kNN), and Random Forest (RF) were trained to detect fall events. We used a dataset containing intentional falls (SisFall) to train the classifier and validated the approach on a different dataset which included real-world accidental fall events of elderly people (FARSEEING). The results suggested that the linear SVM was the most suitable classifier in this cross-dataset validation approach and reliably distinguished a fall event from normal everyday activity at an accuracy of 93% and similarly high sensitivity and specificity. Thus, classifiers based on linear SVM might be useful for automatic fall detection in real-world applications.

Evaluation of Three Machine Learning Algorithms for the Automatic Classification of EMG Patterns in Gait Disorders.

Gait disorders are common in neurodegenerative diseases and distinguishing between seemingly similar kinematic patterns associated with different pathological entities is a challenge even for the experienced clinician. Ultimately, muscle activity underlies the generation of kinematic patterns. Therefore, one possible way to address this problem may be to differentiate gait disorders by analyzing intrinsic features of muscle activations patterns. Here, we examined whether it is possible to differentiate electromyography (EMG) gait patterns of healthy subjects and patients with different gait disorders using machine learning techniques. Nineteen healthy volunteers (9 male, 10 female, age 28.2 ± 6.2 years) and 18 patients with gait disorders (10 male, 8 female, age 66.2 ± 14.7 years) resulting from different neurological diseases walked down a hallway 10 times at a convenient pace while their muscle activity was recorded via surface EMG electrodes attached to 5 muscles of each leg (10 channels in total). Gait disorders were classified as predominantly hypokinetic (n = 12) or ataxic (n = 6) gait by two experienced raters based on video recordings. Three different classification methods (Convolutional Neural Network-CNN, Support Vector Machine-SVM, K-Nearest Neighbors-KNN) were used to automatically classify EMG patterns according to the underlying gait disorder and differentiate patients and healthy participants. Using a leave-one-out approach for training and evaluating the classifiers, the automatic classification of normal and abnormal EMG patterns during gait (2 classes: "healthy" and "patient") was possible with a high degree of accuracy using CNN (accuracy 91.9%), but not SVM (accuracy 67.6%) or KNN (accuracy 48.7%). For classification of hypokinetic vs. ataxic vs. normal gait (3 classes) best results were again obtained for CNN (accuracy 83.8%) while SVM and KNN performed worse (accuracy SVM 51.4%, KNN 32.4%). These results suggest that machine learning methods are useful for distinguishing individuals with gait disorders from healthy controls and may help classification with respect to the underlying disorder even when classifiers are trained on comparably small cohorts. In our study, CNN achieved higher accuracy than SVM and KNN and may constitute a promising method for further investigation.

Linking Individual Movements to a Skilled Repertoire: Fast Modulation of Motor Synergies by Repetition of Stereotyped Movements.

Motor skills emerge when practicing individual movements enables the motor system to extract building instructions that facilitate the generation of future diverse movements. Here we asked how practicing stereotyped movements for minutes affects motor synergies that encode human motor skills acquired over years of training. Participants trained a kinematically highly constrained combined index-finger and thumb movement. Before and after training, finger movements were evoked at rest by transcranial magnetic stimulation (TMS). Post-training, the angle between posture vectors describing TMS-evoked movements and the training movements temporarily decreased, suggesting the presence of a short-term memory for the trained movement. Principal component analysis was used to identify joint covariance patterns in TMS-evoked movements. The quality of reconstruction of training or grasping movements from linear combinations of a small subset of these TMS-derived synergies was used as an index of neural efficiency of movement generation. The reconstruction quality increased for the trained movement but remained constant for grasping movements. These findings suggest that the motor system rapidly reorganizes to enhance the coding efficiency of a difficult movement without compromising the coding efficiency of overlearned movements. Practice of individual movements may drive an unsupervised bottom-up process that ultimately shapes synergistic neuronal organization by constant competition of action memories.