Deep Metric Learning With Locality Sensitive Mining for Self-Correcting Source Separation of Neural Spiking Signals.

Automated source separation algorithms have become a central tool in neuroengineering and neuroscience, where they are used to decompose neurophysiological signal into its constituent spiking sources. However, in noisy or highly multivariate recordings these decomposition techniques often make a large number of errors. Such mistakes degrade online human-machine interfacing methods and require costly post-hoc manual cleaning in the offline setting. In this article we propose an automated error correction methodology using a deep metric learning (DML) framework, generating embedding spaces in which spiking events can be both identified and assigned to their respective sources. Furthermore, we investigate the relative ability of different DML techniques to preserve the intraclass semantic structure needed to identify incorrect class labels in neurophysiological time series. Motivated by this analysis, we propose locality sensitive mining, an easily implemented sampling-based augmentation to typical DML losses which substantially improves the local semantic structure of the embedding space. We demonstrate the utility of this method to generate embedding spaces which can be used to automatically identify incorrectly labeled spiking events with high accuracy.

A myoelectric digital twin for fast and realistic modelling in deep learning.

Muscle electrophysiology has emerged as a powerful tool to drive human machine interfaces, with many new recent applications outside the traditional clinical domains, such as robotics and virtual reality. However, more sophisticated, functional, and robust decoding algorithms are required to meet the fine control requirements of these applications. Deep learning has shown high potential in meeting these demands, but requires a large amount of high-quality annotated data, which is expensive and time-consuming to acquire. Data augmentation using simulations, a strategy applied in other deep learning applications, has never been attempted in electromyography due to the absence of computationally efficient models. We introduce a concept of Myoelectric Digital Twin - highly realistic and fast computational model tailored for the training of deep learning algorithms. It enables simulation of arbitrary large and perfectly annotated datasets of realistic electromyography signals, allowing new approaches to muscular signal decoding, accelerating the development of human-machine interfaces.

Online Tracking of the Phase Difference Between Neural Drives to Antagonist Muscle Pairs in Essential Tremor Patients.

Transcutaneous electrical stimulation has been applied in tremor suppression applications. Out-of-phase stimulation strategies applied above or below motor threshold result in a significant attenuation of pathological tremor. For stimulation to be properly timed, the varying phase relationship between agonist-antagonist muscle activity during tremor needs to be accurately estimated in real-time. Here we propose an online tremor phase and frequency tracking technique for the customized control of electrical stimulation, based on a phase-locked loop (PLL) system applied to the estimated neural drive to muscles. Surface electromyography signals were recorded from the wrist extensor and flexor muscle groups of 13 essential tremor patients during postural tremor. The EMG signals were pre-processed and decomposed online and offline via the convolution kernel compensation algorithm to discriminate motor unit spike trains. The summation of motor unit spike trains detected for each muscle was bandpass filtered between 3 to 10 Hz to isolate the tremor related components of the neural drive to muscles. The estimated tremorogenic neural drive was used as input to a PLL that tracked the phase differences between the two muscle groups. The online estimated phase difference was compared with the phase calculated offline using a Hilbert Transform as a ground truth. The results showed a rate of agreement of 0.88 ± 0.22 between offline and online EMG decomposition. The PLL tracked the phase difference of tremor signals in real-time with an average correlation of 0.86 ± 0.16 with the ground truth (average error of 6.40° ± 3.49°). Finally, the online decomposition and phase estimation components were integrated with an electrical stimulator and applied in closed-loop on one patient, to representatively demonstrate the working principle of the full tremor suppression system. The results of this study support the feasibility of real-time estimation of the phase of tremorogenic neural drive to muscles, providing a methodology for future tremor-suppression neuroprostheses.

Deep Learning for Robust Decomposition of High-Density Surface EMG Signals.

Blind source separation (BSS) algorithms, such as gradient convolution kernel compensation (gCKC), can efficiently and accurately decompose high-density surface electromyography (HD-sEMG) signals into constituent motor unit (MU) action potential trains. Once the separation matrix is blindly estimated on a signal interval, it is also possible to apply the same matrix to subsequent signal segments. Nonetheless, the trained separation matrices are sub-optimal in noisy conditions and require that incoming data undergo computationally expensive whitening. One unexplored alternative is to instead use the paired HD-sEMG signal and BSS output to train a model to predict MU activations within a supervised learning framework. A gated recurrent unit (GRU) network was trained to decompose both simulated and experimental unwhitened HD-sEMG signal using the output of the gCKC algorithm. The results on the experimental data were validated by comparison with the decomposition of concurrently recorded intramuscular EMG signals. The GRU network outperformed gCKC at low signal-to-noise ratios, proving superior performance in generalising to new data. Using 12 seconds of experimental data per recording, the GRU performed similarly to gCKC, at rates of agreement of 92.5% (84.5%-97.5%) and 94.9% (88.8%-100.0%) respectively for GRU and gCKC against matched intramuscular sources.

Analysis of Retinal Segmentation Changes at High Altitude With and Without Acetazolamide.

Purpose: Our aim was to assess retinal venous diameter and segmented retinal layer thickness variation in acute systemic hypoxia with and without acetazolamide and to relate these changes to high altitude headache (HAH), as a proxy for intracerebral pathophysiology. Methods: A total of 20 subjects participated in a 4-day ascent to the Margherita Hut (4,559 m) on Monte Rosa in the Italian Alps. Each participant was randomized to either oral acetazolamide 250 mg twice daily or placebo. A combination of digital imaging and optical coherence tomography was used to measure retinal vessel diameter and retinal layer thickness. Clinically-assessed HAH was recorded. Results: A total of 18 participants had usable digital and OCT images, with 12 developing HAH. Significant thickening was seen only in the two inner layers of the retina, the retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL) at P = 0.012 and P = 0.010, respectively, independent of acetazolamide. There was a significant positive correlation between HAH and both retinal venous diameter (T = 4.953, P = 0.001) and retinal artery diameter (T = 2.865, P = 0.015), with both unaffected by acetazolamide (F = 0.439, P = 0.518). Conclusions: Retinal venous diameter correlates positively with HAH, adding further evidence for the proposed venous outflow limitation mechanism. The inner layers of the retina swelled disproportionately when compared to the outer layers under conditions of systemic hypoxia. Acetazolamide does not appear to influence altitudinal changes of retinal layers and vasculature.

Peer-led live research demonstrations: challenging medical student misconceptions about research.

Modern health care provision is now fundamentally evidence based, meaning competency in academic medicine is integral to medical training. The Integrated Academic Training pathway provides focussed training in this area at a postgraduate level but no such provision exists at an undergraduate level. A number of peer-led academic societies have emerged across the UK to provide education and support for undergraduates but there is little evidence about the type of peer-led interventions that are effective. We report here the findings of one such peer-led organization, the Warwick Academic Medicine Society. We found that traditional educational interventions, including didactic lectures and small-group teaching, are effective at inspiring students regarding academic medicine but poor at translating this enthusiasm into sustained involvement in research. We find this disparity to be centred on misconceptions amongst students regarding the time and skills required to meaningfully contribute to a research project. Further, we introduce the concept of the Live Research Demonstration (LRD), a novel peer-led educational intervention which aims to address these misconceptions and improve involvement of students in research. Initial pilots of the LRD concept have shown significant promise and we recommend a larger trial across multiple localities to confirm its educational benefits.