眼表疾病眨眼分析技术范围综述:9种设备中6种缺乏准确性验证
A scoping review of integrated blink analysis technologies for ocular surface diseases
一项范围综述从3275条记录中纳入42项研究,识别出9种用于眼表疾病评估的整合式眨眼分析技术,分为台式设备、可穿戴设备和手机或电脑应用三类。LipiView与IDRA的研究最多,部分眨眼率在干眼等眼表疾病人群中呈现差异,但各研究结果并不一致。9种设备中有6种没有与人工判读对照的准确性验证,眨眼与部分眨眼的判定标准在各平台间也缺乏统一。
Introduction
The blink of the human eye plays a vital role in maintaining ocular surface health [1]. Blinking activates the lacrimal pump, assists with tear film propagation, and facilitates tear drainage. This ensures tear film stability and sustains visual clarity. Blink dynamics are determined by parameters including blink rate, inter-blink interval, amplitude, velocity, and completeness. These metrics can vary substantially with age, cognitive load, emotional state, neuropsychiatric diseases, eyelid movement disorders, and ocular surface conditions [1,2,3,4]. Blink kinematics have also been shown to differ significantly between voluntary and involuntary blinks [5].
Dry eye disease (DED) is the most prevalent ocular surface disease globally [2]. The TFOS DEWS III 2025 report defines DED as a multifactorial, symptomatic disease characterised by a loss of tear film and/or ocular surface homeostasis, driven by tear film instability and hyperosmolarity, ocular surface inflammation and damage, and neurosensory abnormalities [6]. Abnormal blink patterns have been implicated in the pathophysiology of tear film dyshomeostasis. Sustained visual attention, particularly during visually demanding tasks, can prolong interblink intervals and reduce blink frequency, leading to reduced meibomian gland expression and increased evaporative tear loss [7]. Incomplete blinks are also more frequent during visual display terminal (VDT) use [3, 4, 7], which has been reported as a significant risk factor for DED [3, 7, 8]. Reported prevalence rates of DED among VDT users reach up to 87.5% [9], with a corresponding rise in dry eye related symptoms among children and adolescents [10, 11].
Although abnormal blinking is recognised as a major contributing factor to DED and abnormal blink parameters have been identified as predictors for DED, objective blink analysis is not widely practised clinically [12,13,14]. Reported studies vary widely in blink measurement methods, and there is little consensus on the diagnostic value of specific blink metrics. In practice, the focus of clinical evaluation of DED and other ocular surface disorders predominantly relies on static measures such as symptom questionnaires, ocular surface staining, tear volume quantification (e.g. Schirmer’s test, phenol red thread test), tear-film quality assessments (e.g. tear-film break-up patterns, lipid-layer thickness), and biomarker analysis (e.g. tear osmolarity, cytokines, matrix metalloproteinase-9).
Barriers to the clinical adoption of blink analysis include the cost, complexity and limited portability of established assessment methods such as electrooculography, electromyography, videography and infrared oculography [1, 15]. These methods require extended in-person testing within a controlled environment, which may not reflect real-world blink behaviour. Ocular surface diseases like DED are dynamic in nature, with symptoms and clinical signs that fluctuate over time and are influenced by a multitude of environmental factors. These signs may not always be evident in static, cross-sectional assessments. There are therefore benefits to integrated blink analysis technologies, which our review defines as end-user targeted solutions designed to be seamless and intuitive, capable of capturing blink metrics across a variety of settings. Our review aims to report on these existing platforms, evaluate their clinical applications in ocular surface disease assessment and management, and identify current limitations and gaps to inform future research and technological development.
Methods
Eligibility criteria
We included studies describing use of integrated technologies capable of capturing and analysing blink metrics for the assessment of dry eye disease and other ocular surface diseases. Non-English publications and publications focused solely on the description or validation of technical specifications, including engineering, device, and software-related aspects, were excluded.
Search methods
A comprehensive literature search was conducted on August 17, 2025 via PubMed MEDLINE, Embase, and Cochrane Controlled Register of Trials (CENTRAL). Search terms included “Device”, “Technology”, “Appliance”, “Application”, “Blink”, “Blinking” and “Eye Blink”. The search included articles between January 1, 2000 and August 17, 2025, without language or publication-type restrictions. Reference lists of eligible studies were hand-searched to identify additional relevant studies. EndNote (Clarivate Analytics) was used to manage all identified publications and remove duplicates. Search results were recorded according to PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines [16].
Study selection
Two independent reviewers (CWO and CHLL) screened all titles, abstracts, and full texts. Discrepancies regarding eligibility were resolved through consensus among all authors.
Data collection and analysis
Extracted data included technology type and technical specifications, method of blink detection, criteria for blink determination, blink parameters evaluated, study design, aim, population characteristics, sample size, results, and outcomes.
Results
A total of 3275 unique records were identified. Following screening of titles and abstracts, 315 full-text publications were assessed, of which 42 met final eligibility (Fig. 1). Amongst these studies, 9 distinct blink-analysis technologies were identified and categorised into three main types: desk-bound devices, wearable devices, and smartphone or computer applications (see Table 1).
PRISMA flow diagram.
Desk-bound devices
Intelligent Vision Sensor (IVS)
Two studies evaluated the Intelligent Vision Sensor (IVS; Hamamatsu Photonics K.K., Hamamatsu, Japan) [17]. IVS is a high-speed imaging system that can capture up to 1000 images per second for real-time analysis of blink dynamics. Participants maintain a fixed head position using a chin rest and forehead support with a gaze-fixation point, with blink data recorded over two 40-second epochs. Blink parameters such as rate, phase amplitudes, phase durations, and eyelid velocities were derived from upper eyelid motion.
Kato et al. reported a significant correlation between increased upper-eyelid closing maximum velocity and conjunctival epithelial damage in 45 DED patients (r = 0.210, P = 0.021) [18].
In a subsequent study of 16 patients with severe aqueous-deficient DED, significant increase in upper-eyelid closing amplitudes (P = 0.025), opening amplitudes (P = 0.023), upper-eyelid closing maximum velocity (P = 0.039) and upper-eyelid opening maximum velocity (P = 0.016) was observed one month after punctal occlusion. These reported findings were attributed to an increase in tear volume which led to a resultant decrease in friction during blinking [19].
LipiView™
The LipiView™ ocular surface interferometer (Johnson & Johnson Vision, Jacksonville, FL, USA) is a desk-mounted instrument that quantifies lipid-layer thickness (LLT), performs meibography, and analyses blink parameters from a 20-s high-definition video recording. The proprietary onboard software detects and analyses the total number of blinks, complete blinks, and partial or incomplete blinks. The partial blink rate is also reported and derived from the ratio of partial blinks to total blinks.
Twenty-seven studies employed the LipiView™ system. The most common ocular surface disease evaluated was DED (n = 17). Other conditions included meibomian gland dysfunction (n = 3), blepharospasm (n = 3), neurotrophic keratitis (n = 1), allergic conjunctivitis (n = 1), thyroid eye disease (n = 1), and abnormal blinking (n = 1). Several other studies assessed the effect of interventions on the ocular surface, including orthokeratology lens wear (n = 3), transcutaneous upper eyelid blepharoplasty (n = 1), Laser-Assisted In Situ Keratomileusis (LASIK) (n = 1), and silicone intubation for nasolacrimal duct obstruction (NLDO) (n = 1). A summary of these studies and their relevant findings can be found in Table 2.
Partial blink rate was the most frequently reported parameter and was assessed in 24 of 27 (88.9%) of studies. This was reported to be significantly higher in younger patients aged 20-40 years compared to older patients with DED (younger age 0.621 ± 0.350 vs middle age 0.528 ± 0.358 vs older age 0.500 ± 0.369, P < 0.001) [20], neurotrophic keratitis (0.79 ± 0.29 vs healthy fellow eyes 0.28 ± 0.25, P < 0.001 and vs control eyes 0.21 ± 0.23, P < 0.001) [21], allergic conjunctivitis (0.637 ± 0.391 vs controls 0.506 ± 0.379, P = 0.024) [22], thyroid eye disease-related DED (0.533 ± 0.345 vs non-thyroidal DED 0.346 ± 0.363, P = 0.006) [23], and post-transcutaneous upper eyelid blepharoplasty (0.81 ± 0.03 at baseline vs 0.58 ± 0.05, P < 0.001) [24]. Partial blink rates were also significantly reduced when measured after therapeutic interventions such as vectored thermal pulsation (baseline 0.607 vs post-intervention 0.179, P = 0.002) [25] and blink exercises (baseline 0.54 ± 0.36 vs 5 min post-blinking exercises 0.43 ± 0.34, P = 0.04 and 28 days post-blinking exercises 0.34 ± 0.29, P < 0.001) [26].
No significant difference in partial blink rate was observed across certain conditions, such as between dry eye patients with and without Sjogren’s Syndrome [27], and active and inactive thyroid eye disease [23]. Likewise, no significant difference was found between healthy controls and DED [28], incomplete NLDO patients [29], orthokeratology lens wearers [30, 31], and blepharospasm patients [32]. Partial blink rates were significantly negatively correlated with age in DED (r = −0.076, P = 0.001) [33], Ocular Surface Disease Index scores (β = −24.931, p = 0.032), and meibomian gland dropout (β = 0.158, P = 0.000) [12].
IDRA
IDRA (SBM Sistemi, Turin, Italy) has the capability to assess blink quality, lipid layer thickness, tear meniscus height, and meibography. It requires subjects’ heads to be positioned in a chinrest, looking into the device camera while blinking naturally. Videos are recorded with the on-board software detecting and analysing full and partial blinks to provide a proprietary eye blink quality score. However, the composite of this score has not been released.
Seven studies evaluated IDRA in DED (n = 5), DED in psoriasis patients (n = 1), and ocular surface changes with topical anti-hypertensive medication use (n = 1). A summary of these studies and their relevant findings can be found in Table 3.
Blink quality scores were generally non-discriminatory between disease groups and controls such as those with topical anti-hypertensive use, psoriasis, and mild DED versus moderate-severe DED [34,35,36]. Blink quality score was assessed in 5 of 7 studies, while 2 remaining studies assessed the number of total, complete, and partial blinks and partial blink rates instead of composite blink scores.
Jeon et al. reported that partial blink rates correlated strongly with MG dropout (r = 0.811, P < 0.01) and moderately with Standard Patient Evaluation of Eye Dryness (SPEED) Questionnaire scores (r = 0.596, P < 0.01) [37].
Diagnostic performance of the IDRA varied across studies. Rinert et al. reported good overall discrimination for DED with an AUC of 0.868 (95% CI [0.809 – 0.927]). Pathologic meibography, abnormal interferometry, and tear meniscus had the strongest associations for DED, giving an estimated probability of 96% for DED [38]. In contrast, Vigo et al. reported poor diagnostic accuracy of the blink quality score with an AUC = 0.533 (sensitivity = 0.48, and specificity = 0.59) and no significant difference in blink quality score between DED and controls (85.0 ± 19.5 vs 87.2 ± 18.8, P = 0.382) [39].
Wearable devices
Wink glass
Wink Glass (Masunaga group, Japan) is a smart-glass system that delivers visual cues via opacity changes to prompt blinking once blink rates drop below 12 per min. The “smart glass” is made up of two polyethylene sheets sandwiching a polymer network liquid crystal over one eye. It can appear either transparent or translucent and is activated by a blink-detecting sensor. The user is forced to blink to restore its transparency below the ideal blink rate. Ang et al. evaluated the effect of Wink Glass on blink rates, ocular surface symptoms, and tear stability during visual display terminal use, reporting that use of Wink Glass increased blink rate (9 vs 15 blinks per min, P = 0.001) and reduced ocular surface symptoms during VDT use [40].
Blink-sensing glasses
Chen et al. incorporated iontronic pressure sensors in nose pads of smart glasses capable of monitoring blink patterns through orbicularis oculi movements as a proxy [41]. This setup was capable of measuring blink rates, average blink durations, and maximum blink intervals (MBI). Measured blink rates demonstrated high correlation with manual measurements derived from video analysis (r = 0.99). A low mean error of −0.14 blinks per min between both methods of measurement was reported, with more than 95% of plotted points falling within agreement limits. A study involving 3 DED subjects and 3 healthy controls reported findings of significantly higher blink rates (20.5 ± 4.0 vs 15.6 ± 3.0 per min, P = 0.001) and blink duration (0.69 ± 0.13 vs 0.60 ± 0.12 s, P = 0.013), but shorter MBI in DED subjects compared to controls (8.07 ± 2.74 vs 9.45 ± 2.36 s, P = 0.041).
Smartphone and computer applications
HealthyRoad
“HealthyRoad” is a facial biometric software that relies on a webcam to detect blink rates by capturing webcam-obtained numerical data on eye-opening percentages without storing photographic or videographic data.
Lapa et al. explored the ability of HealthyRoad to predict computer vision syndrome (CVS) on a real-time basis [42]. HealthyRoad was installed on the computers of 13 students, who also completed the Computer Vision Syndrome Questionnaire (CVS-Q) [43]. In this cohort, blink rate was significantly inversely correlated with CVS scores (β = −1.26, P = 0.019, 95% CI [−2.249, −0.275]), suggesting an association between decreased blinking rates and CVS.
DryEyeRhythm
DryEyeRhythm is a smartphone application developed using Apple Inc’s (Cupertino, CA, USA) open-source framework ResearchKit by Juntendo University Graduate School of Medicine, Tokyo, Japan, and InnoJin Inc., Tokyo, Japan. It facilitates administration of the Japanese version of the Ocular Surface Disease Index questionnaire (J-OSDI) and measures MBI with the front-facing smartphone camera, utilising the CIFaceFeature of the iOS interface for facial detection.
Okumura et al. analysed the concurrent validity of DryEyeRhythm’s measurements of MBI, finding a significant positive correlation between app-based and visually-confirmed MBI (r = 0.329, P = 0.003). However, app-based MBI measurements were reported to be significantly longer than visually-confirmed MBI (P = 0.003), with Bland-Altman analysis demonstrating a difference of 3.62 s (95% LOA [24.6–17.4]) [44].
Despite these findings, the discriminant validity of app-based MBI measurements in distinguishing DED and non-DED groups was satisfactory; app-based MBI measurements were significantly lower for DED participants (13.2 ± 9.3 vs 19.0 ± 11.1 s, P = 0.011) [44]. Concurrent use of app-based J-OSDI and MBI provided further diagnostic discrimination for DED (AUC = 0.910, 95% CI [0.846-0.973]).
Fujio et al. reported the internal consistency and reliability of app-based MBI measurements when deployed across different smartphone platforms (iOS and Android) [45]. The Cronbach’s alpha coefficients of correlation between app-based and visually confirmed MBIs were 0.999 for iOS and 1.000 for Android, while the intraclass correlation coefficients of app-based and visually confirmed MBIs were 0.996 (95% CI 0.994–0.998) for iOS and 0.998 (95% CI 0.997–0.999] for Android, respectively. Concurrent validity assessment with Bland-Altman analysis showed minimal differences between app-based and visually confirmed MBI measurements. Significant positive correlations were identified between app-based and visually confirmed MBI measurements (r = 0.999, P < 0.001 for both iOS and Android). Discriminant validity assessment identified significantly shorter app-based MBI measurements in the DED group compared to the non-DED group (P = 0.021 for iOS, P = 0.028 for Android).
Dry eye or not?
“Dry eye or not?” is a smartphone application designed using Flutter by Google, utilising Google’s machine learning kit for facial landmark detection to measure blink rate, MBI, and visual acuity based on recordings from the front camera at a viewing distance of about 40 cm. In a cross-sectional study of users in Thailand, Kasetsuwan et al. found that subjects with symptomatic DED had a higher blink rate (Median = 18, IQR 16 vs Median 16, IQR 16 blinks per min, P < 0.001), and reduced MBI measurements (Median = 8.90, IQR 10.8 vs Median = 8.9, IQR 14.8 s) compared to those without. These findings were confirmed in a multivariate analysis, where DED was found to be significantly associated with increased blink rates (multivariate-adjusted OR 1.01; 95% CI [1.01–1.02], P < 0.001) and MBI measurements (multivariate-adjusted OR 0.98; 95% CI [0.98–0.99], P < 0.001) [46].
EyeScore
EyeScore is an iOS mHealth application integrating blink metrics, symptoms, and demographics into a 10-point dry eye scoring system. Factors incorporated as part of EyeScore’s scoring system include a blink rate of >30 per min (3-points), partial blink rate of >5 per min (3-points), sensation of dryness and grit (2-points), women over the age of 50 (1-point) and regular contact lens wear (1–point). The following cut-offs were reported: a score of 1–3 was defined as normal, 4–6 as mild, and 7–10 as severe DED. Blink data is captured through a 1-min video recording by the iPhone camera at arm’s length, and analysed with an “eye aspect ratio” (EAR) based algorithm [47] to determine full blinking and partial blinking.
Zhang and Echegoyen evaluated use of the EyeScore in 10 DED patients and 10 controls, with counts of full blinks derived from the app demonstrating perfect agreement with visually confirmed data [48]. A significantly higher partial blink rate was observed in DED versus controls (P < 0.05).
Discussion
Blink analysis has gained increasing recognition as a valuable tool for understanding ocular surface conditions, with emerging evidence supporting its potential therapeutic role through blink behaviour modification [26, 49]. In this review, several technological platforms capable of blink analysis in patients with ocular surface diseases have been identified, with studies demonstrating an association between increased partial blink rates and ocular symptoms, consistent with existing literature.
Accurate and reliable assessment of blink parameters is essential to investigate the intricate relationship between blink dynamics, tear film stability, and ocular surface health. While traditional approaches such as electrooculography, electromyography, and magnetic search coils are able to perform precise measurements, these approaches require specialised equipment and controlled environments, limiting their feasibility in routine clinical practice. Furthermore, although computer vision algorithms and high-speed imaging systems have facilitated advanced automated blink detection [50,51,52,53,54], few studies have assessed their integration into user-friendly clinical or consumer platforms.
Desk-mounted platforms, particularly LipiView™ and IDRA, remain the most widely studied systems for blink analysis in ocular surface disease. Their strength lies in the ability to integrate blink assessment with comprehensive tear film and ocular surface evaluation. However, high cost, limited portability, and reliance on static clinical settings restrict accessibility and real-world adoption.
Smartphone applications such as EyeScore, Dry eye or not? and DryEyeRhythm leverage upon the widespread availability of smartphone cameras to capture blinks. Their strengths lie in accessibility, scalability, and community-based deployment. However, blink detection accuracy is dependent upon the capabilities of integrated hardware and external factors such as masks, eyewear, and headgear, and environmental conditions such as lighting may limit reliability in uncontrolled settings [51, 55].
Wearables are an emerging trend in the medical field and have the potential to monitor and diagnose diseases, and empower patients to take charge of their own health [56]. Wink Glass and Blink-sensing Glasses are promising solutions for real-time, longitudinal blink monitoring. Beyond monitoring, Wink Glass is designed to actively modify blink behaviour for therapeutic purposes. While preliminary results are encouraging and demonstrate improvements in blink rates, further studies are needed to assess long-term user adherence, tolerability, and sustained clinical benefit.
While most included studies examined the impact of ocular surface disease on blink behaviour, there is an increasing awareness of ophthalmic interventions like cataract and refractive surgery as iatrogenic causes of abnormal blink behaviour [57, 58]. Our review included a few studies evaluating the use of LipiView™ and IDRA to identify changes in blink patterns following interventions like Ortho-K lens use, refractive and eyelid surgery, albeit with inconclusive findings. Further studies will be needed to establish the role of integrated blink analysis technologies in quantifying iatrogenic blink abnormalities.
As non-invasive diagnostic tools, integrated blink analysis technologies can be useful in ophthalmic and optometry practices, complementing the clinical examination and thereby empowering practitioners with the ability to screen for and diagnose ocular surface diseases. The use of smartphone- and webcam-based integrated blink analysis solutions may facilitate the practice of tele-optometry by providing remote assessment of blink patterns and treatment response. In tertiary eye care centres, the automated and user-friendly nature of these integrated technologies can be incorporated into existing clinical workflows without impacting upon clinic effiency.
Despite the potential of existing integrated technologies for blink analysis, barriers to their widespread use remain. Most notably, there is insufficient validation of the accuracy of several devices in detecting and determining blink parameters. Rigorous validation is a prerequisite for clinicians to confidently adopt these technologies in clinical practice. As summarised in Table 4, there is significant heterogeneity in the methods of blink capture adopted and lack of standardised criteria to define what constitutes a blink, or a partial blink. The criteria for blink determination were not reported across several platforms. Heterogeneity of methods adopted results in variability in efficacy and measurements, leading to challenges in integrating findings across platforms. There is also limited information on inter-device agreement, with only one study comparing LipiView™ and IDRA reporting significant differences in measurements [59]. Assessment of test-retest repeatability, intra-observer, and inter-observer reliability is also lacking.
Validation studies to determine the accuracy of measured blink parameters were absent for six of the nine devices identified. Only three devices (Blink-sensing Glasses, DryEyeRhythm, and EyeScore) had accuracy rates published against visually confirmed counts for several blink metrics, while the accuracy for other blink parameters, such as partial blinks, which have a recognised association with dry eye disease, was not reported [7, 10].
The diagnostic performance of these devices for ocular surface diseases also remains inadequately established. In the limited number of studies assessing diagnostic accuracy, equivocal differences in measured blink parameters between patients afflicted with ocular surface conditions and control groups have been reported [30, 32, 34, 36]. These findings translate to challenges in applying devices for clinical screening and diagnosis.
Finally, as most blink analysis technologies rely on capturing identifiable facial features, robust privacy safeguards are essential. Secure data handling, anonymisation, and regulatory compliance are prerequisites for widespread clinical adoption.
Strengths and limitations
This scoping review adheres to the PRISMA Extension for Scoping Reviews [16] framework. This review represents the first comprehensive summary of blink analysis technologies applied to ocular surface evaluation. These findings provide a roadmap for the community by identifying key trends, gaps, and opportunities for future development. This review was limited to devices with published validation data for ocular surface evaluation. Consequently, platforms such as the MYAH optical biometer and corneal topographer (Topcon Healthcare Inc., Tokyo, Japan), and TERA dry eye imager (Topcon Healthcare Inc., Tokyo, Japan) were excluded due to a lack of studies in this domain; however, these platforms remain candidates for future research. Finally, our review does not include a formal critical appraisal of the methodological quality of the included studies, which is beyond the scope of this review. However, it can be noted that a large proportion of included studies are cross-sectional by design with small sample sizes, and the geographic distribution of the included literature was skewed significantly, with the majority of studies conducted in Asia.
Conclusion
Integrated blink analysis technologies represent a highly promising frontier for the assessment of ocular surface conditions, with certain devices demonstrating therapeutic potential through active blink behaviour modification. Although desk-mounted devices like LipiViewTM and IDRA currently dominate the evidence base, there is a clear paradigm shift toward portable and wearable solutions capable of performing continuous real-world monitoring. To achieve widespread clinical integration, future research must establish standardised diagnostic criteria and prioritise rigorous comparative studies to validate the accuracy, reliability, and clinical utility of these emerging tools.
References
Cruz AAV, Garcia DM, Pinto CT, Cechetti SP. Spontaneous eyeblink activity. Ocul Surf. 2011;9:29–41. https://doi.org/10.1016/S1542-0124(11)70007-6.
Nakamori K, Odawara M, Nakajima T, Mizutani T, Tsubota K. Blinking is controlled primarily by ocular surface conditions. Am J Ophthalmol. 1997;124:24–30. https://doi.org/10.1016/s0002-9394(14)71639-3.
Oganov A, Yazdanpanah G, Jabbehdari S, Belamkar A, Pflugfelder S. Dry eye disease and blinking behaviors: a narrative review of methodologies for measuring blink dynamics and inducing blink response. Ocul Surf. 2023;29:166–74. https://doi.org/10.1016/j.jtos.2023.05.011.
Talens-Estarelles C, Esteve-Taboada JJ, Sanchis-Jurado V, Pons ÁM, García-Lázaro S. Blinking kinematics characterization during digital displays use. Graefes Arch Clin Exp Ophthalmol. 2022;260:1183–93. https://doi.org/10.1007/s00417-021-05490-9.
Sanchis-Jurado V, Talens-Estarelles C, Esteve-Taboada JJ, Pons ÁM, García-Lázaro S. Non-invasive high-speed blinking kinematics characterization. Graefes Arch Clin Exp Ophthalmol. 2020;258:2701–14. https://doi.org/10.1007/s00417-020-04782-w.
Wolffsohn JS, Benítez-Del-Castillo J, Loya-Garcia D, Inomata T, Iyar G, Liang L, et al. TFOS DEWS III Diagnostic Methodology. Am J Ophthalmol. 2025: 40451408. https://doi.org/10.1016/j.ajo.2025.05.033.
Fjærvoll K, Fjærvoll H, Magno M, Nøland ST, Dartt DA, Vehof J, et al. Review on the possible pathophysiological mechanisms underlying visual display terminal-associated dry eye disease. Acta Ophthalmol. 2022;100:861–77. https://doi.org/10.1111/aos.15150.
Uchino M, Yokoi N, Uchino Y, Dogru M, Kawashima M, Komuro A, et al. Prevalence of dry eye disease and its risk factors in visual display terminal users: the Osaka study. Am J Ophthalmol. 2013;156:759–66. https://doi.org/10.1016/j.ajo.2013.05.040.
Courtin R, Pereira B, Naughton G, Chamoux A, Chiambaretta F, Lanhers C, et al. Prevalence of dry eye disease in visual display terminal workers: a systematic review and meta-analysis. BMJ Open. 2016;6:e009675. https://doi.org/10.1136/bmjopen-2015-009675.
Craig JP, Nichols KK, Akpek EK, Caffery B, Dua HS, Joo CK, et al. TFOS DEWS II definition and classification report. Ocul Surf. 2017;15:276–83. https://doi.org/10.1016/j.jtos.2017.05.008.
Marshall SJ, Gorely T, Biddle SJ. A descriptive epidemiology of screen-based media use in youth: a review and critique. J Adolesc. 2005;29:333–49. https://doi.org/10.1016/j.adolescence.2005.08.016.
Jie Y, Sella R, Feng J, Gomez ML, Afshari NA. Evaluation of incomplete blinking as a measurement of dry eye disease. Ocul Surf. 2019;17:440–6. https://doi.org/10.1016/j.jtos.2019.05.007.
Hirosawa K, Inomata T, Sung J, Nakamura M, Okumura Y, Midorikawa-Inomata A, et al. Diagnostic ability of maximum blink interval together with Japanese version of Ocular Surface Disease Index score for dry eye disease. Sci Rep. 2020;10:1. https://doi.org/10.1038/s41598-020-75193-4.
Johnston PR, Rodriguez J, Lane KJ, Ousler G, Abelson MB. The interblink interval in normal and dry eye subjects. Clin Ophthalmol (Auckl, NZ). 2013;7:253. https://doi.org/10.2147/OPTH.S39104.
Nyström M, Andersson R, Niehorster DC, Hessels RS, Hooge ITC. What is a blink? Classifying and characterizing blinks in eye openness signals. Behav Res Methods. 2024;56:3280–99. https://doi.org/10.3758/s13428-023-02333-9.
Tricco AC, Lillie E, Zarin W, O’Brien KK, Colquhoun H, Levac D, et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med. 2018;169:467–73. https://doi.org/10.7326/M18-0850.
Kimura N, Watanabe A, Suzuki K, Toyoda H, Hakamata N, Fukuoka H, et al. Measurement of spontaneous blinks in patients with Parkinson’s disease using a new high-speed blink analysis system. J Neurol Sci. 2017;380:200–4. https://doi.org/10.1016/j.jns.2017.07.035.
Kato H, Yokoi N, Watanabe A, Komuro A, Sonomura Y, Sotozono C, et al. Relationship between ocular surface epithelial damage, tear abnormalities, and blink in patients with dry eye. Cornea. 2019;38:318–24. https://doi.org/10.1097/ico.0000000000001841.
Kato H, Yokoi N, Watanabe A, Komuro A, Sonomura Y, Sotozono C, et al. Effect of punctal occlusion on blinks in eyes with severe aqueous deficient dry eye. Diagnostics. 2023;14:38201312. https://doi.org/10.3390/diagnostics14010003.
Weng HY, Ho WT, Chiu CY, Tsai TY, Chang SW. Characteristics of tear film lipid layer in young dry eye patients. J Formos Med Assoc. 2021;120:1478–84. https://doi.org/10.1016/j.jfma.2020.10.028.
Xie C, Liu B, Zhao X, He Q, Liu L, Wei R. Characteristics of the ocular surface in neurotrophic keratitis induced by trigeminal nerve injury following neurosurgery. Int Ophthalmol. 2023;43:1229–40. https://doi.org/10.1007/s10792-022-02521-0.
Yang B, Wen K, Li J, Zhang S, Fan Z, Liang X, et al. Quantitative evaluation of lipid layer thickness and blinking in children with allergic conjunctivitis. Graefes Arch Clin Exp Ophthalmol. 2021;259:2795–805. https://doi.org/10.1007/s00417-021-05199-9.
Park J, Baek S. Dry eye syndrome in thyroid eye disease patients: The role of increased incomplete blinking and Meibomian gland loss. Acta Ophthalmol. 2019;97:e800–6. https://doi.org/10.1111/aos.14000.
Zhang S, Yan Y, Lu Y, Zhou Y, Fu Y. Effect of transcutaneous upper eyelid blepharoplasty on blink parameters and lipid layer thickness. Front Med. 2021;8:732041. https://doi.org/10.3389/fmed.2021.732041.
Chan AYY, Chuang JC, Wong VWY. Evaluation of meibomian gland dysfunction among ophthalmic healthcare workers. Clin Ophthalmol. 2021;15:1201–6. https://doi.org/10.2147/OPTH.S299338.
Kim AD, Muntz A, Lee J, Wang MTM, Craig JP. Therapeutic benefits of blinking exercises in dry eye disease. Cont Lens Anterior Eye. 2021;44:101329. https://doi.org/10.1016/j.clae.2020.04.014.
Zang S, Cui Y, Cui Y, Fei W. Meibomian gland dropout in Sjögren’s syndrome and non-Sjögren’s dry eye patients. Eye. 2018;32:1681–7. https://doi.org/10.1038/s41433-018-0149-5.
Tangmonkongvoragul C, Chokesuwattanaskul S, Khankaeo C, Punyasevee R, Nakkara L, Moolsan S, et al. Prevalence of symptomatic dry eye disease with associated risk factors among medical students at Chiang Mai University due to increased screen time and stress during COVID-19 pandemic. PLoS One. 2022;17:e0265733. https://doi.org/10.1371/journal.pone.0265733.
Lee SM, Chung SJ, Lew H. Evaluation of tear film lipid layer thickness measurements obtained using an ocular surface interferometer in nasolacrimal duct obstruction patients. Korean J Ophthalmol. 2018;32:445–50. https://doi.org/10.3341/kjo.2018.0012.
Lee J, Hwang G, Ha M, Kim HS, Han K, Na KS. Evaluation of the meibomian glands using the tear interferometer wearing orthokeratology lenses. BMC Ophthalmol. 2022;22:133. https://doi.org/10.1186/s12886-022-02365-3.
Yu H, Yuan Y, Wu W, Zeng W, Tong L, Zhang Y, et al. Orthokeratology lens wear for 2 years in children did not alter tear film lipid thickness by non-invasive interferometry. Front Med. 2022;9:821106. https://doi.org/10.3389/fmed.2022.821106.
Jang J, Lew H. Blink index as a response predictor of blepharospasm to botulinum neurotoxin-A treatment. Brain Behav. 2021;11:e2374. https://doi.org/10.1002/brb3.2374.
Wu WL, Chang SW. Dermatochalasis aggravates meibomian gland dysfunction related dry eyes. J Clin Med. 2022;11:2379. https://doi.org/10.3390/jcm11092379.
Castro C, Marta A, Marques JH, Ferreira A, José D, Sousa P, et al. Ocular surface changes in primary open angle glaucoma patients treated with topical antihypertensive drugs. J Glaucoma. 2023;32:e113–20. https://doi.org/10.1097/ijg.0000000000002272.
Nowowiejska J, Ordon AJ, Baran A, Izdebska J, Woźniak B, Kaminski TW, et al. Dry eye syndrome symptoms in patients with psoriasis. J Eur Acad Dermatol Venereol. 2023;38:1522–30. https://doi.org/10.1111/jdv.19773.
Barbosa Ribeiro B, Marques JH, Baptista PM, Sousa PJM, Pires S, Menéres P, et al. Corneal epithelial thickness correlation with dry eye symptom severity: a cross-sectional study. Clin Ophthalmol. 2024;18:3313–20. https://doi.org/10.2147/opth.S480704.
Jeon YJ, Song MY, Kim KY, Hwang KY, Kwon YA, Koh K. Relationship between the partial blink rate and ocular surface parameters. Int Ophthalmol. 2021;41:2601–8. https://doi.org/10.1007/s10792-021-01819-9.
Rinert J, Branger G, Bachmann LM, Pfaeffli O, Iselin K, Kaufmann C, et al. Accuracy of a new noninvasive automatic ocular surface analyzer for the diagnosis of dry eye disease-two-gate design using healthy controls. Cornea. 2022;42:416–22. https://doi.org/10.1097/ico.0000000000003052.
Vigo L, Pellegrini M, Bernabei F, Carones F, Scorcia V, Giannaccare G. Diagnostic performance of a novel noninvasive workup in the setting of dry eye disease. J Ophthalmol. 2020;2020:5804123. https://doi.org/10.1155/2020/5804123.
Ang CK, Mohidin N, Chung KM. Effects of wink glass on blink rate, nibut and ocular surface symptoms during visual display unit use. Curr Eye Res. 2014;39:879–84. https://doi.org/10.3109/02713683.2013.859273.
Chen R, Zhang Z, Deng K, Wang D, Ke H, Cai L, et al. Blink-sensing glasses: A flexible iontronic sensing wearable for continuous blink monitoring. iScience. 2021;24:102399. https://doi.org/10.1016/j.isci.2021.102399.
Lapa I, Ferreira S, Mateus C, Rocha N, Rodrigues MA. Real-time blink detection as an indicator of computer vision syndrome in real-life settings: an exploratory study. Int J Environ Res Public Health. 2023;20:4569. https://doi.org/10.3390/ijerph20054569.
Seguí Mdel M, Cabrero-García J, Crespo A, Verdú J, Ronda E. A reliable and valid questionnaire was developed to measure computer vision syndrome at the workplace. J Clin Epidemiol. 2015;68:662–73. https://doi.org/10.1016/j.jclinepi.2015.01.015.
Okumura Y, Inomata T, Midorikawa-Inomata A, Sung J, Fujio K, Akasaki Y, et al. DryEyeRhythm: a reliable and valid smartphone application for the diagnosis assistance of dry eye. Ocul Surf. 2022;25:19–25. https://doi.org/10.1016/j.jtos.2022.04.005.
Fujio K, Nagino K, Huang T, Sung J, Akasaki Y, Okumura Y, et al. Clinical utility of maximum blink interval measured by smartphone application DryEyeRhythm to support dry eye disease diagnosis. Sci Rep. 2023;13:1. https://doi.org/10.1038/s41598-023-40968-y.
Kasetsuwan N, Suwan-Apichon O, Lekhanont K, Chuckpaiwong V, Reinprayoon U, Chantra S, et al. Assessing the risk factors for diagnosed symptomatic dry eye using a smartphone app: cross-sectional study. JMIR Mhealth Uhealth. 2022;10:e31011. https://doi.org/10.2196/31011.
Soukupová T, Cech J. Real-time eye blink detection using facial landmarks. In. 2016 [cited 2025 Jun 22]. Available from: https://www.semanticscholar.org/paper/Real-Time-Eye-Blink-Detection-using-Facial-Soukupov%C3%A1-Cech/4fa1ba3531219ca8c39d8749160faf1a877f2ced.
Zhang S, Echegoyen J. Design and Usability Study of a Point of Care mHealth App for Early Dry Eye Screening and Detection. J Clin Med. 2023;12:6479 https://doi.org/10.3390/jcm12206479.
Fujita H, Sano K, Baba T, Tanaka T, Ohno-Matsui K. Blind working time in visual display terminal users. J Occup Health. 2019;61:175–81. https://doi.org/10.1002/1348-9585.12027.
Bernard F, Deuter CE, Gemmar P, Schachinger H. Eyelid contour detection and tracking for startle research related eye-blink measurements from high-speed video records. Comput Methods Prog Biomed. 2013;112:22–37. https://doi.org/10.1016/j.cmpb.2013.06.003.
Dewi C, Chen RC, Jiang X, Yu H. Adjusting eye aspect ratio for strong eye blink detection based on facial landmarks. PeerJ Comput Sci. 2022;8:e943. https://doi.org/10.7717/peerj-cs.943.
Espinosa J, Domenech B, Vázquez C, Pérez J, Mas D. Blinking characterization from high speed video records. Application to biometric authentication. PLoS One. 2018;13:e0196125. https://doi.org/10.1371/journal.pone.0196125.
Kwon KA, Shipley RJ, Edirisinghe M, Ezra DG, Rose G, Best SM, et al. High-speed camera characterization of voluntary eye blinking kinematics. J R Soc Interface. 2013;10:20130227. https://doi.org/10.1098/rsif.2013.0227.
Navascues-Cornago M, Morgan PB, Maldonado-Codina C, Read ML. Characterisation of blink dynamics using a high-speed infrared imaging system. Ophthalmic Physiol Opt. 2020;40:519–28. https://doi.org/10.1111/opo.12694.
Xiong J, Dai W, Wang Q, Dong X, Ye B, Yang J. A review of deep learning in blink detection. PeerJ Comput Sci. 2025;11:e2594. https://doi.org/10.7717/peerj-cs.2594.
Kang HS, Exworthy M. Wearing the future-wearables to empower users to take greater responsibility for their health and care: scoping review. JMIR Mhealth Uhealth. 2022;10:e35684. https://doi.org/10.2196/35684.
Talens-Estarelles C, Díez-Ajenjo MA, Pons ÁM, Peris-Martínez C, García-Lázaro S. Effects of cataract surgery on blinking. J Cataract Refract Surg. 2023;49:177–83. https://doi.org/10.1097/j.jcrs.0000000000001081.
Hei Lai KK, Hu Z, Chiu JT, Aljufairi FMAA, Sebastian JU, Cheng GPM, et al. Long-term evaluation of ocular surface and meibomian gland function after laser-assisted in situ keratomileusis surgery. Ophthalmic Res. 2025;68:84–9. https://doi.org/10.1159/000542985.
Lee JM, Jeon YJ, Kim KY, Hwang KY, Kwon YA, Koh K. Ocular surface analysis: a comparison between the LipiView® II and IDRA®. Eur J Ophthalmol. 2021;31:2300–6. https://doi.org/10.1177/1120672120969035.
Ren Y, Wen H, Bai F, Huang B, Wang Z, Zhang S, et al. Comparison of deep learning-assisted blinking analysis system and LipiView interferometer in dry eye patients: a cross-sectional study. Eye Vis. 2024;11:7. https://doi.org/10.1186/s40662-024-00373-6.
Chou YB, Fan NW, Lin PY. Value of lipid layer thickness and blinking pattern in approaching patients with dry eye symptoms. Can J Ophthalmol. 2019;54:735–40. https://doi.org/10.1016/j.jcjo.2019.03.005.
Zhao H, Wu SN, Cheng Z, Xiao D, Shu HY, Ge QM, et al. Mean tear-film lipid layer thickness and video display terminal time as risk factors for abnormal blinking in children. Front Med (Lausanne). 2021;8:785901. https://doi.org/10.3389/fmed.2021.785901.
Li Y, Sang X, Yang L, Wang XR, Liu JH, He XJ, et al. Low concentration of sodium hyaluronate temporarily elevates the tear film lipid layer thickness in dry eye patients with lipid deficiency. Int J Ophthalmol. 2018;11:389–94. https://doi.org/10.18240/ijo.2018.03.07.
Zhang J, Wu Z, Sun L, Liu XH. Function and morphology of the meibomian glands using a LipiView interferometer in rotating shift medical staff. J Ophthalmol. 2020;2020:3275143. https://doi.org/10.1155/2020/3275143.
Zhao G, Wang J, Ma Q, Hao Y, Tian L, Jie Y. Comparison of upper eyelid pressure and ocular surface parameters in dry eye disease and benign essential Blepharospasm: a cross-sectional study. Contact Lens Anterior Eye. 2024;47:102298. https://doi.org/10.1016/j.clae.2024.102298.
Chang L, Guo Q, Pazo EE, Li F, Zhang Z, Fang Z, et al. Prevalence of dry eye in people over 50 years old in Beichen district, Tianjin city: a cross-sectional population-based survey. BMC Public Health. 2024;24:2111. https://doi.org/10.1186/s12889-024-19616-1.
Paik JS, Kwon MJ, Nam GH, Han K, Whang WJ, Hwang HS, et al. Tear film lipid layer thickness and blink dynamics in patients with blepharospasm. BMC Ophthalmol. 2025;25:22. https://doi.org/10.1186/s12886-025-03852-z.
Feng J, Gu Y, Wen Y, Wang Y, Wang J, Li A, et al. Influence of overnight orthokeratology on ocular surface and blink patterns in children and adolescents-a prospective cohort study. Eye Contact Lens. 2025;51:375–9. https://doi.org/10.1097/ICL.0000000000001197.
Shen J, Xu Z, Wang Y, Ye J, Wei L, Xia J, et al. Influential signs of dry eye-related ocular symptoms in participants with unstable tear film. Cornea. 2024;43:751–6. https://doi.org/10.1097/ICO.0000000000003473.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Open access funding provided by {0}, WA Health Libraries Network Consortium.
Author information
Authors and Affiliations
Department of Ophthalmology, National University Hospital, Singapore, Singapore
Chee Wui Ong, Blanche Xiaohong Lim & Chris Hong Long Lim
Lee Kong Chian School of Medicine, National Technological University, Singapore, Singapore
Blanche Xiaohong Lim
Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore
Blanche Xiaohong Lim
Department of Computer Science, University of Missouri-Saint Louis, Saint Louis, MO, USA
Sanjiv K. Bhatia & Tjaart van der Walt
College of Optometry, University of Missouri-Saint Louis, Saint Louis, MO, USA
Tareq Nabhan
Center for Sustainable Medicine, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore
Chris Hong Long Lim
Cornea and Oculoplastics Units, Department of Ophthalmology, Royal Perth Hospital, Perth, WA, Australia
Chris Hong Long Lim
Lions Eye Institute, Nedlands, WA, Australia
Chris Hong Long Lim
Singapore Eye Research Institute, Singapore Eye Research Institute, Singapore, Singapore
Chris Hong Long Lim
Authors
- Chee Wui Ong
- Blanche Xiaohong Lim
- Sanjiv K. Bhatia
- Tjaart van der Walt
- Tareq Nabhan
- Chris Hong Long Lim
Contributions
CWO was responsible for conducting the search, screening potentially eligible studies, extracting and analysing data, interpreting results, and writing the manuscript. BXL, SKB, TW and TN contributed to analysing data, interpreting results and writing the manuscript. CHLL was responsible for designing the review protocol, screening potentially eligible studies, analysing data and writing the manuscript.
Corresponding author
Correspondence to Chris Hong Long Lim.
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Ong, C.W., Lim, B.X., Bhatia, S.K. et al. A scoping review of integrated blink analysis technologies for ocular surface diseases. Eye (2026). https://doi.org/10.1038/s41433-026-04873-8
Received: 31 December 2025
Revised: 10 August 2026
Accepted: 07 September 2026
Published: 05 October 2026
Version of record: 05 October 2026
DOI: https://doi.org/10.1038/s41433-026-04873-8
来源:Nature 健康护理主题 · nature.com