Measuring Saccade Latency Using Smartphone Cameras

Abstract

Accurate quantification of neurodegenerative disease progression is an ongoing challenge that complicates efforts to understand and treat these conditions. Clinical studies have shown that eye movement features may serve as objective biomarkers to support diagnosis and tracking of disease progression. Here, we demonstrate that saccade latency -an eye movement measure of reaction time -can be measured robustly outside of the clinical environment with a smartphone camera. Methods: To enable tracking of saccade latency in large cohorts of patients and control subjects, we combined a deep convolutional neural network for gaze estimation with a modelbased approach for saccade onset determination that provides automated signal-quality quantification and artifact rejection. Results: Simultaneous recordings with a smartphone and a highspeed camera resulted in negligible differences in saccade latency distributions. Furthermore, we demonstrated that the constraint of chinrest support can be removed when recording healthy subjects. Repeat smartphone-based measurements of saccade latency in eleven self-reported healthy subjects resulted in an intraclass correlation coefficient of 0.76, showing our approach has good to excellent test-retest reliability. Additionally, we conducted over 19,000 saccade latency measurements in 29 self-reported healthy subjects and observed significant intra-and inter-subject variability, which highlights the importance of individualized tracking. Lastly, we showed that with around 65 measurements we can estimate mean saccade latency to within less-than-10ms precision, which takes within four minutes with our setup. Conclusion and Significance: By enabling repeat measurements of saccade latency and its distribution in individual subjects, our framework opens the possibility of quantifying patient state on a finer timescale in a broader population than previously possible.

Eye Movement Alterations in Post-COVID-19 Condition: A Proof-of-Concept Study

“…The latency values depend on the target location tested as well as the method to assess it (gap, step or overlap) [70]. Several studies report the values of latency in visually guided saccade test with a large number of volunteers [71][72][73][74]. By comparing results given in Table 7 with those previously reported, we found similar latency values.…”

Digital Phenotyping in Clinical Neurology

“…A recent study presented stimuli on a tablet and used its built-in camera to identify facial landmarks and track head position to demonstrate that toddlers with autism orient to name less consistently and with longer delays compared with controls. 70 Studies have shown that it is possible to capture oculomotor information, including saccade latency in healthy individuals 71 and cerebellar smooth pursuit abnormalities in individuals with ataxia 72 using the high-speed camera present in some mobile phones combined with computer vision algorithms. These studies, although performed in the clinic or laboratory setting, show 40,81,[141][142][143][144][145] PD, HD, stroke Features of center of mass (COM) acceleration (reflecting COM excursions and total path and space covered); maximum and mean angular excursion from midline; total range of angular excursions; mean velocity, mean acceleration, and entropy of that it is possible to capture gaze and oculomotor information using cameras on everyday mobile devices.…”

App-based saccade latency and error determination across the adult age spectrum

“…In our previous work [15], we displayed the visual reaction task on a laptop and recorded the subjects with an iPhone. Synchronization of the recording and task display was achieved through a second screen that mirrored the laptop screen and was recorded alongside the subject's response [15]. Given the elaborate set-up, the recording was limited to our laboratory setting.…”

“…However, these features are commonly measured with dedicated infrared cameras and chinrests, which limits the measurements to the doctor's office or the neurophysiological laboratory. In our previous work [13]- [15], we showed that we can accurately and robustly determine saccade latency from recordings obtained with a smartphone camera.…”