Ankle torque forecasting using time-delayed neural networks.

A method for ankle torque prediction ahead of the current time is proposed in this paper. The mean average value of EMG signals from four muscles, alongside the joint angle and angular velocity of the right ankle, were used as input parameters to train a time-delayed artificial neural network. Data collected from five healthy subjects were used to generate the dataset to train and test the model. The model predicted ankle torque for five different future times from zero to 2 seconds. Model predictions were compared to torque calculated from inverse dynamics for each subject. The model predicted ankle torque up to 1 second ahead of time with normalized root mean squared error of less than 15 percent while the coefficient of determination was over 0.85.Clinical Relevance- the potential of the model for predicting joint torque ahead of time is helpful to establish an intuitive interaction between human and assistive robots. This model has application to assist patients with neurological disorders.

Comparative Study of a Biomechanical Model-based and Black-box Approach for Subject-Specific Movement Prediction.

The performance and safety of human robot interaction (HRI) can be improved by using subject-specific movement prediction. Typical models include biomechanical (parametric) or black-box (non-parametric) models. The current work aims to investigate the benefits and drawbacks of these approaches by comparing elbow-joint torque predictions based on electromyography signals of the elbow flexors and extensors. To this end, a parameterized biomechanical model is compared to a non-parametric (Gaussian-process) approach. Both models showed adequate results in predicting the elbow-joint torques. While the non-parametric model requires minimal modeling effort, the parameterized biomechanical model can lead to deeper insight of the underlying subject specific musculoskeletal system.

Application of BCI systems in neurorehabilitation: a scoping review.

PURPOSE: To review various types of electroencephalographic activities of the brain and present an overview of brain-computer interface (BCI) systems' history and their applications in rehabilitation. METHODS: A scoping review of published English literature on BCI application in the field of rehabilitation was undertaken. IEEE Xplore, ScienceDirect, Google Scholar and Scopus databases were searched since inception up to August 2012. All experimental studies published in English and discussed complete cycle of the BCI process was included in the review. RESULTS AND DISCUSSION: In total, 90 articles met the inclusion criteria and were reviewed. Various approaches that improve the accuracy and performance of BCI systems were discussed. Based on BCI's clinical application, reviewed articles were categorized into three groups: motion rehabilitation, speech rehabilitation and virtual reality control (VRC). Almost half of the reviewed papers (48%) concentrated on VRC. Speech rehabilitation and motion rehabilitation made up 33% and 19% of the reviewed papers, respectively. Among different types of electroencephalography signals, P300, steady state visual evoked potentials and motor imagery signals were the most common. CONCLUSIONS: This review discussed various applications of BCI in rehabilitation and showed how BCI can be used to improve the quality of life for people with neurological disabilities. It will develop and promote new models of communication and finally, will create an accurate, reliable, online communication between human brain and computer and reduces the negative effects of external stimuli on BCI performance. Implications for Rehabilitation The field of brain-computer interfaces (BCI) is rapidly advancing and it is expected to fulfill a critical role in rehabilitation of neurological disorders and in movement restoration in the forthcoming years. In the near future, BCI has notable potential to become a major tool used by people with disabilities to control locomotion and communicate with surrounding environment and, consequently, improve the quality of life for many affected persons. Electrical field recording at the scalp (i.e. electroencephalography) is the most likely method to be of practical value for clinical use as it is simple and non-invasive. However, some aspects need future improvements, such as the ability to separate close imagery signal (motion of extremities and phalanges that are close together).