Towards increased data transmission rate for a three-class metabolic brain–computer interface based on transcranial Doppler ultrasound
Highlights
► We investigated the future viability of a three-class ultrasound-based BCI. ► We differentiated three mental states offline based on changes in blood flow. ► We determined the time window that maximized data transmission rate. ► Data transmission rate peaked when mental task length was 20 s. ► Classification accuracies exceeded 70% using time-domain features and LDA.
Introduction
Brain–computer interfaces (BCIs) allow users to generate control signals for external devices using only their thoughts [7]. Due to their ability to bypass typical output channels such as movement and speech, BCIs are of interest within the field of rehabilitation engineering [29]. Specifically, BCIs can be used as an alternative means of communication in individuals with severe physical disabilities resulting from conditions such as stroke and amyotrophic lateral sclerosis (ALS). In extreme cases, these disabilities can result in total immobility and inability to communicate while retaining full consciousness. This condition is referred to as “locked-in syndrome” (LIS) [26]. The provision of a means of communication for individuals with LIS continues to be an important goal of BCI research [8].
Previous non-invasive BCI research has focused on a small number of measurement modalities, of which the foremost has been electroencephalography (EEG) [3], [14]. Recent research has also investigated alternative measurement modalities such as functional magnetic resonance imaging (fMRI) and near-infrared spectroscopy (NIRS) [25], [32]. While EEG directly measures neuronal activity, fMRI and NIRS measure changes in blood hemoglobin concentrations [15]. Consequently, BCIs using the latter modalities are often referred to as hemodynamic or metabolic BCIs [19]. These BCIs do not generally possess the same temporal resolution as EEG BCIs, but have still attracted attention due to their intuitive training methods and robustness against electrical artifacts [10]. Recent research in this area has produced a number of real-time fMRI and NIRS-based BCIs [1], [4], [6], [11], [18]. These BCIs, many of which rely upon detection of motor imagery (e.g. imagined hand movement or finger tapping), suggest that metabolic BCIs are worthy of further study.
Another metabolic signal that may be suitable for BCI development is transcranial Doppler ultrasound (TCD) [20]. TCD measures cerebral blood flow velocity (CBFV) within the circle of Willis (the network of arteries that supply the brain) [30]. Cognitive activation produces increases in CBFV within these arteries that can be detected using TCD [28]. These changes have been observed for a wide variety of different mental tasks [31], suggesting the potential to automatically detect mental activity on the basis of changes in CBFV. This possibility was investigated by Myrden et al. in [20], where it was shown that two different mental activities (word generation and mental rotation) can be differentiated from rest with greater than 80% accuracy. However, these results were achieved using very long durations for each activity (45 s), yielding a very low data transmission rate. This limits the practicality of such a BCI. Consequently, improvement of the data transmission rate is necessary in order to demonstrate the practical viability of a TCD-based BCI.
In BCIs, data transmission rate depends on three parameters – the number of potential classes (N), the classification accuracy (P), and the state duration – the length of time for which a mental activity is performed before it is classified. The first two variables determine the data transmission rate in bits per trial (B), which can be expressed as [22], [33]:
Using the state duration, data transmission rate can be converted to bits per second or bits per minute. It is clear that data transmission rate can be augmented by increasing either N or P, or by decreasing the state duration. The effects of each parameter on data transmission rate (in bits per minute) are shown in Fig. 1.
Increasing the number of classes and reducing state durations is likely to decrease classification accuracy. This limits the maximum achievable data transmission rate. In this paper, we investigate the net gain in data transmission rate that can be attained by varying these parameters for a TCD-based BCI. We have expanded the classification problem introduced in [20] to a three-class problem by attempting to differentiate word generation, mental rotation, and rest from each other. Furthermore, state durations have been limited to a range of durations between 5 and 30 s. If state durations can be substantially reduced without greatly decreasing classification accuracy, data transmission rate will be improved.
Section snippets
Participants
Nine able-bodied participants (6 female, mean age 25.6 ± 2.4 years) were recruited from the Bloorview Research Institute. All participants were right-handed, as quantified by the Edinburgh Handedness Inventory [21], with a mean score of 79.4 ± 16.3. Participants had no history of migraine and no known neurological, cardiopulmonary, or respiratory conditions. All participants gave informed written consent. This study was approved by the Research Ethics Boards of both Holland Bloorview Kids
Results
The mean classification accuracy across all participants using two and three-dimensional feature sets is displayed in Fig. 2 for state durations ranging from 5 to 30 s. Mean classification accuracy ranged between 40% and 69% for two-dimensional feature sets, and between 37% and 74% for three-dimensional feature sets. For both sets, classification accuracy increased with increasing state duration, but tended to stabilize as state duration exceeded 20 s. In Fig. 3, these curves have been converted
Discussion
In this study, we have shown that mean classification accuracies exceeding 70% can be achieved for a three-class problem within 20 s of the onset of cognitive activity using bilateral TCD measurements, time-domain features, and a linear classifier. This corresponds to a maximum data transmission rate of 1.2 bits per minute, compared to a maximum rate of 0.3 bits per minute previously reported for a TCD-based BCI [20]. This significant improvement highlights the advantages of a three-class BCI
Acknowledgments
This work was supported by the Canada Research Chairs program, the Natural Sciences and Engineering Research Council, Holland Bloorview Kids Rehabilitation Hospital, and Barbara and Frank Milligan.
References (33)
- et al.
Real-time imaging of human brain function by near-infrared spectroscopy using an adaptive general linear model
NeuroImage
(2009) - et al.
Self-modulation of primary motor cortex activity with motor and motor imagery tasks using real-time fMRI-based neurofeedback
NeuroImage
(2012) - et al.
Neurofeedback and brain–computer interface: clinical applications
International Review of Neurobiology
(2009) - et al.
Brain–computer interfaces in neurological rehabilitation
Lancet Neurology
(2008) - et al.
Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials
Electroencephalography and Clinical Neurophysiology
(1988) Decoding fMRI brain states in real-time
NeuroImage
(2011)The assessment and analysis of handedness: the Edinburgh inventory
Neuropsychologia
(1971)- et al.
Temporal classification of multichannel near-infrared spectroscopy signals of motor imagery for developing a brain–computer interface
NeuroImage
(2007) - et al.
Practice standards for transcranial Doppler ultrasound: Part I – test performance
Journal of Neuroimaging
(2007) - et al.
Brain–computer interface systems: progress and prospects
Expert Review of Medical Devices
(2007)
Assessment of cognitive neural correlates for a function near infrared-based brain computer interface system
Found Augmented Cognition. Neuroergonomics and Operational Neuroscience
Single-trial classification of antagonistic oxyhemoglobin responses during mental arithmetic
Medical & Biological Engineering & Computing
Breaking the silence: brain–computer interfaces (BCI) for communication and motor control
Psychophysiology
Brain–computer interface in paralysis
Current Opinion in Neurology
On the suitability of near-infrared (NIR) systems for next-generation brain–computer interfaces
Physiological Measurement
Brain–computer interface using a simplified near-infrared spectroscopy system
Journal of Neural Engineering
Cited by (13)
A brain-computer interface based on functional transcranial doppler ultrasound using wavelet transform and support vector machines
2018, Journal of Neuroscience MethodsCitation Excerpt :The studies suggested in (Goyal et al., 2016), (Myrden et al., 2012) reported accuracies of 62.40% within 15 s (Goyal et al., 2016) and 40% within 5 s (Myrden et al., 2012) of the onset of the cognitive task. Moreover, the maximum classification accuracy over the whole observation period obtained by our approach was 72.91% achieved in 11.27 s compared to the previously reported 73.11% obtained in 24.90 s (Myrden et al., 2012). A maximum bit rate of 3.83 and 3.09 bits per minute were achieved for the binary and the 3-class problems respectively using the moving window method, while the incremental window achieved maximum bit rates of 3.95 and 2.28 bits per minute.
An online three-class Transcranial Doppler ultrasound brain computer interface
2016, Neuroscience ResearchCitation Excerpt :However, lengthy task durations were required, limiting the data transmission rate. In a follow-up study using a three-class LDA, Myrden et al. (2012) varied the amount of time needed for the tasks, reporting that accuracies in excess of 70% could be attained with approximately 18 s task durations. Subsequently, Aleem and Chau (2013) proposed a user-independent TCD BCI.
Pattern classification to optimize the performance of Transcranial Doppler Ultrasonography-based brain machine interface
2015, Pattern Recognition LettersCitation Excerpt :This offline result exceeds the chance level and compares favorably to findings of previous studies, while considering the shortest data collection window to date and a combination of computationally efficient features and classifier. In fact, the present finding supports the methodological choices in [23] though SVM may be a viable classifier for TCD–BCI when used with WSFS in future online studies. Past TCD studies have shown that verbal fluency tasks tend to induce a left-lateralized increase in MCA CBFV in right-handed individuals [10].
Sequential hypothesis testing for automatic detection of task-related changes in cerebral perfusion in a brain-computer interface
2015, Neuroscience ResearchCitation Excerpt :The data transmission rate (DTR) incorporates the effects of both classification accuracy and decision time. It can be expressed in bits per trial as (Myrden et al., 2012): Classification accuracies, decision times, and data transmission rates for each participant are presented in Table 2.
Assessment of Resting-State Blood Flow Through Anterior Cerebral Arteries Using Trans-cranial Doppler Recordings
2013, Ultrasound in Medicine and Biology