Prediction of Pathological Tremor Signals Using Long Short-Term Memory Neural Networks

Abstract

Previous implementations of closed-loop peripheral electrical stimulation (PES) strategies have provided evidence about the effect of the stimulation timing on tremor reduction. However, these strategies have used traditional signal processing techniques that only consider phase prediction and might not model the non-stationary behavior of tremor. Here, we tested the use of long shortterm memory (LSTM) neural networks to predict tremor signals using kinematic data recorded from Essential Tremor (ET) patients. A dataset comprising wrist flexionextension data from 12 ET patients was pre-processed to feed the predictors. A total of 180 models resulting from the combination of network (neurons and layers of the LSTM networks, length of the input sequence and prediction horizon) and training parameters (learning rate) were trained, validated and tested. Predicted tremor signals using LSTM-based models presented high correlation values (from 0.709 to 0.998) with the expected values, with a phase delay between the predicted and real signals below 15 ms, which corresponds approximately to 7.5% of a tremor cycle. The prediction horizon was the parameter with a higher impact on the prediction performance. The proposed LSTM-based models were capable of predicting both phase and amplitude of tremor signals outperforming results from previous studies (32-56% decreased phase prediction error compared to the out-of-phase method), which might provide a more robust PES-based closed-loop control applied to PES-based tremor reduction.

Method of diagnosing some diseases of the neuro-muscular system and features of data processing in software

“…Let's use the method of the learning model or the refe rence model, which belongs to the parametric methods of system identification [17,18]. In accordance with this method, the reaction of the studied system to the test impact is compared with the reaction of the model of a given species to the same impact [19,20]. The comparison error is used to adjust the parameters of the model according to the cri terion of minimizing the deviation of the response of the system and the model.…”

Discovering Themes in Deep Brain Stimulation Research Using Explainable Artificial Intelligence

“…For example, closed-loop deep brain stimulation systems can use neural activity to detect symptoms and adapt stimulation parameters in real-time [ 8 ]. These systems could utilize computational models of brain activity that help decide how to alter stimulation and inhibit an unwanted behavior before it begins, such as a tremor [ 9 ]. This adaptive stimulation approach follows a sensor-trigger protocol that alters the stimulation parameters according to the difference between ongoing neural activity and a desired motif of neural activity.…”

Classification of Kinematic and Electromyographic Signals Associated with Pathological Tremor Using Machine and Deep Learning

“…The application of Artificial Neural Networks (ANNs) and deep learning models to model pathological tremors has gained interest in recent years [28]. The higher complexity of some of these models, such as auto-encoders, Recurrent Neural Networks or Long Short-Term Memory (LSTM) neural networks, allows the identification of temporal patterns in a time series [29]. However, whether the increase in the complexity of these models translates into more accurate outcomes compared to traditional machine learning approaches is yet to be proven [30].…”