Reconstructing animated eye movements from electrooculography data to aid the diagnosis of vestibular disorders.

OBJECTIVE: To develop a method of visualising electrooculography data to improve the interpretability of nystagmus eye-movements captured using the Continuous Ambulatory Vestibular Assessment (CAVA®) device. DESIGN: We are currently undertaking a clinical investigation to evaluate the capabilities of the CAVA® device to detect periods of pathological nystagmus. The work presented here was undertaken using unblinded data obtained from the preliminary phase of this investigation. STUDY SAMPLE: One patient with Ménière's disease and one with Benign Paroxysmal Positional Vertigo. RESULTS: Using the electrooculography data captured by the CAVA® device, we reconstructed 2D animations of patients' eye movements during attacks of vertigo. We were able to reanimate nystagmus produced as a consequence of two conditions. Concurrent video footage showed that the animations were visually very similar to the patient's actual eye-movements, excepting torsional eye-movements. CONCLUSIONS: The reconstructed animations provide an alternative presentation modality, enabling clinicians to largely interpret electrooculography data as if they were present during a vertigo attack. We were able to recreate nystagmus from attacks experienced in the community rather than a clinical setting. This information provides an objective record of a patient's nystagmus and could be used to complement a full neurotologic history when considering diagnosis and treatment options.

Detecting positional vertigo using an ensemble of 2D convolutional neural networks.

The aim of the work presented here was to develop a system that can automatically identify attacks of dizziness occurring in patients suffering from positional vertigo, which occurs when sufferers move their head into certain positions. We used our novel medical device, CAVA, to record eye- and head-movement data continually for up to 30 days in patients diagnosed with a disorder called Benign Paroxysmal Positional Vertigo. Building upon our previous work, we describe a novel ensemble of five 2D Convolutional Neural Networks, using composite recognition features, including eye-movement data and three-channel accelerometer data. We achieve an F 1 score of 0.63 across an 11-fold cross-fold validation experiment, demonstrating that the system can detect a few seconds of motion provoked dizziness from within over a 100 h of normal eye-movement data. We show that the system outperforms our previous 1D Neural Network approach, and that our ensemble classifier is superior to each of the individual networks it contains. We also demonstrate that our composite recognition features provide improved performance over results obtained using the individual data sources independently.

Clinical Techniques and Technology: Vestibular Telemetry.

When a patient presents to a clinician with dizziness, it can be difficult for the patient to describe their symptoms in a clear manner, and clinical examination often yields entirely normal results. Ideally, it would be favorable to measure key physiological parameters during their episodes of dizziness. From a clinical perspective, this would allow a more timely and more accurate diagnosis. From a research perspective, it would allow a greater understanding of how the vestibular system malfunctions as a consequence of vestibular disease. The authors of this report have been funded by the UK Medical Research Council to develop and test a novel technology to measure, record, and analyze key physiological parameters provided by the dizzy individual during an episode of dizziness while active in the community. We provide the context to evolving work in this field, the outcome of preliminary studies, and a consideration of future opportunities.

1D Convolutional Neural Networks for Detecting Nystagmus.

Vertigo is a type of dizziness characterised by the subjective feeling of movement despite being stationary. One in four individuals in the community experience symptoms of dizziness at any given time, and it can be challenging for clinicians to diagnose the underlying cause. When dizziness is the result of a malfunction in the inner-ear, the eyes flicker and this is called nystagmus. In this article we describe the first use of Deep Neural Network architectures applied to detecting nystagmus. The data used in these experiments was gathered during a clinical investigation of a novel medical device for recording head and eye movements. We describe methods for training networks using very limited amounts of training data, with an average of 11 mins of nystagmus across four subjects, and less than 24 hours of data in total, per subject. Our methods work by replicating and modifying existing samples to generate new data. In a cross-fold validation experiment, we achieve an average F1 score of 0.59 (SD = 0.24) across all four folds, showing that the methods employed are capable of identifying periods of nystagmus with a modest degree of accuracy. Notably, we were also able to identify periods of pathological nystagmus produced by a patient during an acute attack of Ménière's Disease, despite training the network on nystagmus that was induced by different means.

Nystagmus during an acute Ménière's attack: from prodrome to recovery.

OBJECTIVE: We are currently undertaking a clinical investigation to evaluate the diagnostic capability of a system for detecting periods of pathological dizziness. This article presents an analysis of the data captured during an acute attack of Ménière's disease. DESIGN: The Continuous Ambulatory Vestibular Assessment (CAVA) device is worn by patients in the community, and continuously records eye and head movement data (vestibular telemetry). STUDY SAMPLE: A 53-year-old lady with a fifteen-year history of left-sided unilateral Ménière's disease. RESULTS: The patient wore the device nearly continuously for thirty days. The data revealed a three-hour long attack of vertigo consisting of four separate phases of nystagmus. The duration, beat-direction and slow phase velocity of the nystagmus evolved through time. The first phase contained isolated nystagmus beats which preceded the patient's record of the vertigo attack onset but coincided with anticipation of an impending vertigo attack. CONCLUSIONS: CAVA provides a unique insight into the physiological parameters present during episodes of dizziness. Here, it has provided the first full example of an acute Ménière's attack, including a period of prodrome. These findings have implications for the prediction of vertigo attack onset, for the diagnosis of Ménière's disease and other diseases resulting in dizziness.

Automatic nystagmus detection and quantification in long-term continuous eye-movement data.

Symptoms of dizziness or imbalance are frequently reported by people over 65. Dizziness is usually episodic and can have many causes, making diagnosis problematic. When it is due to inner-ear malfunctions, it is usually accompanied by abnormal eye-movements called nystagmus. The CAVA (Continuous Ambulatory Vestibular Assessment) device has been developed to provide continuous monitoring of eye-movements to gain insight into the physiological parameters present during a dizziness attack. In this paper, we describe novel algorithms for detecting short periods of artificially induced nystagmus from the long-term eye movement data collected by the CAVA device. In a blinded trial involving 17 healthy subjects, each participant induced nystagmus artificially on up to eight occasions by watching a short video on a VR headset. Our algorithms detected these short periods with an accuracy of 98.77%. Additionally, data relating to vestibular induced nystagmus was collected, analysed and then compared to a conventional technique for assessing nystagmus during caloric testing. The results show that a range of nystagmus can be identified and quantified using computational methods applied to long-term eye-movement data captured by the CAVA device.

An investigation into the diagnostic accuracy, reliability, acceptability and safety of a novel device for Continuous Ambulatory Vestibular Assessment (CAVA).

Dizziness is a common condition that is responsible for a significant degree of material morbidity and burden on health services. It is usually episodic and short-lived, so when a patient presents to their clinician, examination is normal. The CAVA (Continuous Ambulatory Vestibular Assessment) device has been developed to provide continuous monitoring of eye-movements, allowing insight into the physiological parameters present during a dizziness attack. This article describes the first clinical investigation into the medical and technical aspects of this new diagnostic system. Seventeen healthy subjects wore the device near continuously for up to thirty days, artificially inducing nystagmus on eight occasions. 405 days' worth of data was captured, comprising around four billion data points. A computer algorithm developed to detect nystagmus demonstrated a sensitivity of 99.1% (95% CI: 95.13% to 99.98%) and a specificity of 98.6% (95% CI: 96.54% to 99.63%). Eighty-two percent of participants wore the device for a minimum of eighty percent of each day. Adverse events were self-limiting and mostly the consequence of skin stripping from the daily replacement of the electrodes. The device was shown to operate effectively as an ambulatory monitor, allowing the reliable detection of artificially induced nystagmus.

Ectopic BASL Reveals Tissue Cell Polarity throughout Leaf Development in Arabidopsis thaliana.

Tissue-wide polarity fields, in which cell polarity is coordinated across the tissue, have been described for planar organs such as the Drosophila wing and are considered important for coordinating growth and differentiation [1]. In planar plant organs, such as leaves, polarity fields have been identified for subgroups of cells, such as stomatal lineages [2], trichomes [3, 4], serrations [5], or early developmental stages [6]. Here, we show that ectopic induction of the stomatal protein BASL (BREAKING OF ASYMMETRY IN THE STOMATAL LINEAGE) reveals a tissue-wide epidermal polarity field in leaves throughout development. Ectopic GFP-BASL is typically localized toward the proximal end of cells and to one lobe of mature pavement cells, revealing a polarity field that aligns with the proximodistal axis of the leaf (base to tip). The polarity field is largely parallel to the midline of the leaf but diverges in more lateral positions, particularly at later stages in development, suggesting it may be deformed during growth. The polarity field is observed in the speechless mutant, showing that it is independent of stomatal lineages, and is observed in isotropic cells, showing that cell shape anisotropy is not required for orienting polarity. Ectopic BASL forms convergence and divergence points at serrations, mirroring epidermal PIN polarity patterns, suggesting a common underlying polarity mechanism. Thus, we show that similar to the situation in animals, planar plant organs have a tissue-wide cell polarity field, and this may provide a general cellular mechanism for guiding growth and differentiation.

Macro optical projection tomography for large scale 3D imaging of plant structures and gene activity.

Optical projection tomography (OPT) is a well-established method for visualising gene activity in plants and animals. However, a limitation of conventional OPT is that the specimen upper size limit precludes its application to larger structures. To address this problem we constructed a macro version called Macro OPT (M-OPT). We apply M-OPT to 3D live imaging of gene activity in growing whole plants and to visualise structural morphology in large optically cleared plant and insect specimens up to 60 mm tall and 45 mm deep. We also show how M-OPT can be used to image gene expression domains in 3D within fixed tissue and to visualise gene activity in 3D in clones of growing young whole Arabidopsis plants. A further application of M-OPT is to visualise plant-insect interactions. Thus M-OPT provides an effective 3D imaging platform that allows the study of gene activity, internal plant structures and plant-insect interactions at a macroscopic scale.