Background: Eye movement abnormalities are commonplace in neurological disorders. However, unaided eye movement assessments lack granularity. Although videooculography (VOG) improves diagnostic accuracy, resource intensiveness precludes its broad use. To bridge this care gap, we here validate a framework for smartphone video-based nystagmography capitalizing on recent computer vision advances.Methods: A convolutional neural network was fine-tuned for pupil tracking using > 550 annotated frames: ConVNG. In a cross-sectional approach, slow-phase velocity of optokinetic nystagmus was calculated in 10 subjects using ConVNG and VOG. Equivalence of accuracy and precision was assessed using the “two one-samplet-test” (TOST) and Bayesian interval-null approaches. ConVNG was systematically compared to OpenFace and MediaPipe as computer vision (CV) benchmarks for gaze estimation.Results: ConVNG tracking accuracy reached 9–15% of an average pupil diameter. In a fully independent clinical video dataset, ConVNG robustly detected pupil keypoints (median prediction confidence 0.85). SPV measurement accuracy was equivalent to VOG (TOSTp < 0.017; Bayes factors (BF) > 24). ConVNG, but not MediaPipe, achieved equivalence to VOG in all SPV calculations. Median precision was 0.30°/s for ConVNG, 0.7°/s for MediaPipe and 0.12°/s for VOG. ConVNG precision was significantly higher than MediaPipe in vertical planes, but both algorithms’ precision was inferior to VOG.Conclusions: ConVNG enables offline smartphone video nystagmography with an accuracy comparable to VOG and significantly higher precision than MediaPipe, a benchmark computer vision application for gaze estimation. This serves as a blueprint for highly accessible tools with potential to accelerate progress toward precise and personalized Medicine.

Tremor is one of the most common neurological symptoms. Its clinical and neurobiological complexity necessitates novel approaches for granular phenotyping. Instrumented neurophysiological analyses have proven useful, but are highly resource-intensive and lack broad accessibility. In contrast, bedside scores are simple to administer, but lack the granularity to capture subtle but relevant tremor features. We utilise the open-source computer vision pose tracking algorithm Mediapipe to track hands in clinical video recordings and use the resulting time series to compute canonical tremor features. This approach is compared to marker-based 3D motion capture, wrist-worn accelerometry, clinical scoring and a second, specifically trained tremor-specific algorithm in two independent clinical cohorts. These cohorts consisted of 66 patients diagnosed with essential tremor, assessed in different task conditions and states of deep brain stimulation therapy. We find that Mediapipe-derived tremor metrics exhibit high convergent clinical validity to scores (Spearman’s ρ = 0.55–0.86, p≤ .01) as well as an accuracy of up to 2.60 mm (95% CI [−3.13, 8.23]) and ≤0.21 Hz (95% CI [−0.05, 0.46]) for tremor amplitude and frequency measurements, matching gold-standard equipment. Mediapipe, but not the disease-specific algorithm, was capable of analysing videos involving complex configurational changes of the hands. Moreover, it enabled the extraction of tremor features with diagnostic and prognostic relevance, a dimension which conventional tremor scores were unable to provide. Collectively, this demonstrates that current computer vision algorithms can be transformed into an accurate and highly accessible tool for video-based tremor analysis, yielding comparable results to gold standard tremor recordings.

TOR1A is the most common inherited form of dystonia with still unclear pathophysiology and reduced penetrance of 30-40%. ∆ETorA rats mimic the TOR1A disease by expression of the human TOR1A mutation without presenting a dystonic phenotype. We aimed to induce dystonia-like symptoms in male ∆ETorA rats by peripheral nerve injury and to identify central mechanism of dystonia development. Dystonia-like movements (DLM) were assessed using the tail suspension test and implementing a pipeline of deep learning applications. Neuron numbers of striatal parvalbumin⁺, nNOS⁺, calretinin⁺, ChAT⁺ interneurons and Nissl⁺ cells were estimated by unbiased stereology. Striatal dopaminergic metabolism was analyzed via in vivo microdialysis, qPCR and western blot. Local field potentials (LFP) were recorded from the central motor network. Deep brain stimulation (DBS) of the entopeduncular nucleus (EP) was performed. Nerve-injured ∆ETorA rats developed long-lasting DLM over 12 weeks. No changes in striatal structure were observed. Dystonic-like ∆ETorA rats presented a higher striatal dopaminergic turnover and stimulus-induced elevation of dopamine efflux compared to the control groups. Higher LFP theta power in the EP of dystonic-like ∆ETorA compared to wt rats was recorded. Chronic EP-DBS over 3 weeks led to improvement of DLM. Our data emphasizes the role of environmental factors in TOR1A symptomatogenesis. LFP analyses indicate that the pathologically enhanced theta power is a physiomarker of DLM. This TOR1A model replicates key features of the human TOR1A pathology on multiple biological levels and is therefore suited for further analysis of dystonia pathomechanism.

Vertigo is a common neurological complaint, which can result in significant morbidity and decreased quality of life. While pathology to peripheral and subtentorial brain structures is a well-established cause of vertigo, cortical lesions have also been linked to vertigo and may lend insight into relevant neuroanatomy. Here, we investigate the supratentorial lesion locations associated with vertigo and test whether they map to a common brain network. We performed a systematic literature search and identified 23 cases of supratentorial brain lesions associated with vertigo. We mapped the lesion locations to a standard brain template and computed the network of brain regions functionally connected to each lesion location, using a ‘wiring diagram’ of the human brain termed the human connectome (n = 1000). Sensitivity was assessed by identifying the most common connection to lesion locations associated with vertigo, and specificity was assessed through comparison with control lesions associated with symptoms other than vertigo (n = 68). We found that functional connectivity between lesion locations and the bilateral ventral posterior insula was both sensitive (22/23 lesions) and specific (voxel-wise family-wise error-corrected P < 0.05) for lesion-induced vertigo. We computed connectivity with this hub region to define a lesion-based vertigo network, which included regions in the bilateral insula, somatosensory cortex, higher-level visual areas, cingulate sulcus, thalamus and multiple cerebellar regions in the territory of the posterior inferior cerebellar artery. Next, we used stereo-electroencephalography (80 stimulation sites across 17 patients) to test whether stimulation sites associated with vertigo mapped to this same network. We found that 36/42 (86%) of stimulation sites eliciting vertigo fell within the lesion-based vertigo network in contrast to 16/39 (41%) of stimulation sites that did not elicit vertigo. Connectivity between stimulation sites and our lesion-based hub in the ventral posterior insula was also significantly associated with vertigo (P < 0.0001). We conclude that cortical lesions and direct electrical stimulation sites associated with vertigo map to a common brain network, offering insights into the causal neuroanatomical substrate of vertigo.

Dystonia is a neurological movement disorder characterised by abnormal involuntary movements and postures, particularly affecting the head and neck. However, current clinical assessment methods for dystonia rely on simplified rating scales which lack the ability to capture the intricate spatiotemporal features of dystonic phenomena, hindering clinical management and limiting understanding of the underlying neurobiology. To address this, we developed a visual perceptive deep learning framework that utilizes standard clinical videos to comprehensively evaluate and quantify disease states and the impact of therapeutic interventions, specifically deep brain stimulation. This framework overcomes the limitations of traditional rating scales and offers an efficient and accurate method that is rater-independent for evaluating and monitoring dystonia patients. To evaluate the framework, we leveraged semi-standardized clinical video data collected in three retrospective, longitudinal cohort studies across seven academic centres. We extracted static head angle excursions for clinical validation and derived kinematic variables reflecting naturalistic head dynamics to predict dystonia severity, subtype, and neuromodulation effects. The framework was also applied to a fully independent cohort of generalised dystonia patients for comparison between dystonia sub-types. Computer vision-derived measurements of head angle excursions showed a strong correlation with clinically assigned scores. Across comparisons, we identified consistent kinematic features from full video assessments encoding information critical to disease severity, subtype, and effects of neural circuit interventions, independent of static head angle deviations used in scoring. Our visual perceptive machine learning framework reveals kinematic pathosignatures of dystonia, potentially augmenting clinical management, facilitating scientific translation, and informing personalized precision neurology approaches.

Disorders of consciousness are characterized by severe impairments in arousal and awareness. Deep brain stimulation is a potential treatment, but outcomes vary—possibly due to differences in patient characteristics, electrode placement, or the specific brain network engaged. We describe 40 patients with disorders of consciousness undergoing deep brain stimulation targeting the thalamic centromedian-parafascicular complex. Improvements in consciousness are associated with better-preserved gray matter, particularly in the striatum. Electric field modeling reveals that stimulation is most effective when it extends below the centromedian nucleus, engaging the inferior parafascicular nucleus and the adjacent ventral tegmental tract—a pathway that connects the brainstem and hypothalamus and runs along the midbrain-thalamus border. External validation analyzed show that effective stimulation engages a brain network overlapping with disrupted patterns of brain activity observed in two independent cohorts with impaired consciousness: one with arousal-impairing stroke lesions and the other with awareness-impairing seizures. Together, these findings advance the field by informing patient selection, refining stimulation targets, and identifying a brain network linked to recovery that may have broader therapeutic relevance across consciousness-impairing conditions.