FormalPara Key Summary Points

Phoropters are widely accepted for clinical use in refraction examination and visual function assessment. It can be time-consuming and tiring for optometrists.

The repeatability of the new Inspection Platform of Visual Function (IPVF) and the conventional equipment phoropter were high, and the IPVF instrument was slightly better in terms of positive relative accommodation (PRA) repeatability than the phoropter.

The agreement of phoria, negative relative accommodation (NRA)/PRA, and AMP measured by the new IPVF instrument and phoropter was also satisfactory.

The IPVF instrument could be an alternative for clinicians to obtain visual function measurements with improved efficiency and fewer subjective errors.

Introduction

Nonstrabismic binocular dysfunctions are common vision abnormalities, with a prevalence of 2–61.7% for accommodative insufficiency and that of 2.25–33% for convergence insufficiency [1]. They have been studied since the 1980s [1,2,3]. Their relevant indicators are accommodative anomalies, convergence and divergence anomalies, and phoria [1, 2, 4, 5]. Measurements of accommodation and phoria are important components of comprehensive eye examinations. Accommodation is the ability of the eyes to change the refractive power to focus objects on the retina at varying distances [6]. Phoria is the tendency of the eyes to not be directed towards the point of fixation, manifested in the absence or prevention of fusion. Amplitude of accommodation (AMP) is usually measured using the minus lens and push-up or push-down methods [7, 8]. Clinical measurement methods for phoria include the von Graefe, Maddox rod, Maddox wing, Thorington, and modified Thorington methods. These methods largely depend on the examiner’s training, skills, and experience, which leads to high inter-examiner variability [9, 10].

Phoropters are widely accepted for clinical use in refraction examination and visual function assessment. They are used by optometrists with professional training. In large-scale eye screening or busy hospital hours, patients must be inspected individually using conventional instruments, which can be time consuming and tiring for optometrists and cause long queuing time for patients.

The aim of this study was to evaluate the possibility of an alternative automatic diagnostic instrument to assess binocular visual function. Using this automatic instrument can avoid inter-examiner variability, helping to resolve the shortage of optometrists, and offer a better testing service to eye examinees. The Inspection Platform of Visual Function (IPVF, Youyan Zhiqiang, China) is a newly developed visual function measurement and diagnosis equipment that can automatically perform visual function tests, including Worth 4 Dot, stereopsis, phoria, range of positive/negative fusional vergence, vergence facility, negative/positive relative accommodation (NRA/PRA), accommodative response, accommodative amplitude (AMP), accommodative facility, and accommodation convergence/accommodation tests. Owing to time constraints, only the measurements of Phoria_D, Phoria_N, NRA/PRA, and AMP were included in this study. We assessed whether the repeatability of the inspection platform was equal to or better than that of the conventional equipment. Additionally, we evaluated whether the new and conventional equipment could yield consistent measurement results in the same visual function tests.

Methods

Study Population

This study followed the tenets of the Declaration of Helsinki, and approval of the experimental protocol was obtained from the ethics committee of the Eye Hospital of Wenzhou Medical University (approval No. 2022. 013). Healthy volunteers aged 18–25 years were enrolled. All participants provided consent before enrollment in the study. Inclusion criteria were: best-corrected Snellen visual acuity ≥ 1.0, no manifest strabismus/intermittent exotropia, no history of ocular trauma, no ocular diseases except refractory errors (confirmed using slit-lamp biomicroscopes by an ophthalmologist), no diabetes mellitus, and no current medication potentially affecting accommodation, such as topiramate, topical atropine, pilocarpine, or tropicamide.

Devices

Conventional manual measurements were performed using a phoropter (TOPCON VT-10, Japan). The phoropter is a commonly used instrument for eye examinations. The phoropter used in this study was manually operated. It could perform the following tests: Worth 4-dot, stereopsis, phoria, positive/negative fusion vergence, accommodative function (NRA/PRA; accommodative response, AMP), and accommodation convergence/accommodation.

The IPVF instrument consists of four parts: a lens unit (a standard refractor head controlled by an electric motor), an optotype unit (that can switch between distance and near modes under the control of the electric motor), a control pad, and an integrated device (Fig. 1). The first three parts are connected to the integrated device. The lens unit provides the lens used for all tests. The optotype unit is used to display targets at a distance of 40 cm or 5 m. The control pad uses voice to guide participants to the control handle or press buttons and receive feedback. The integrated device is the core control unit of the platform, and its core functions are as follows: to provide guidance to participants through voice prompts or graphical display, to integrate the functions of the lens and optotype units, to receive feedback from participants and adjust the test procedures accordingly, and to record and analyze the test data.

Fig. 1
figure 1

Assembled platform of the intelligent inspection platform of visual function, which includes: (1) lens unit, (2) optotype unit, (3) control pad, and (4) integrated device. Parts (1–3) are connected to part (4)

Full size image

Procedures

The same examiner with professional training worked on both the IPVF instrument and the phoropter. The same optometrist performed three consecutive tests. Whether the two instruments were used first for a participant was random. The participants were provided with 5 min to rest their eyes between tests. All participants were refractively corrected at distance. During the tests, 40 cm was used for close distance, and 5 m was used for long distance. The optotype used in NRA, PRA, and AMP tests corresponded to one line above the best-corrected visual acuity of the participant. The pupil distance was adjusted according to the test distance.

Using the phoropter, Phoria_D and Phoria_N were measured with the von Graefe method. A prism of 12 ∆BI in the right eye and 6 ∆BU in the left eye was used to split the target into two parts. The prismatic degree of the right eye was reduced at a rate of 2 ∆/s until the participant could align the two targets vertically. The final prismatic degree used was the participant’s phoria. The total value of the positive/negative lens added to both eyes to achieve the last clear vision before continuous blurriness was the participant’s NRA/PRA. The AMP of the right eye (ODAMP) was measured using the minus lens method. The sum of the negative lens used to reach the last clear vision before an unrecoverable blurriness plus 2.5 D was the final AMP.

Using the IPVF instrument, automatic measurements of Phoria_D, Phoria_N, NRA/PRA, and AMP were performed with the platform real-time interaction with the participant. During the testing process, no assistance from optometrists was required. Measurement procedures using the platform followed the same principle as those used in the phoropter; however, the optometrist’s operation was replaced by the platform’s automatic control, which was guided by the platform’s interaction with the participant.

Statistical Analysis

Statistical analyses were performed using SPSS version 20.0 (SPSS, Inc., Chicago, IL, USA). The data of three consecutive measurements of Phoria_D and Phoria_N, NRA/PRA, and ODAMP by the phoropter and the IPVF were evaluated using the intraclass correlation coefficient (ICC) and within-subject standard deviation (Sw) for repeatability [11, 12]. Besides, the mean test values of the two instruments were evaluated using the Bland–Altman plot for agreement. The 95% limit of agreement (LoA) was defined as the mean ± 1.96 standard deviation of the differences [13, 14]. Medcalc version 19.2.6 (MedCalc Software bv, Ostend, Belgium) was used for power and sample size calculations. A sample size of 80 subjects was estimated to detect a difference between measurements of AMP in the two instruments, based on expected mean of differences 1.0 D and expected standard deviation of differences 0.397 D.

Results

Over 80 healthy participants aged 18–25 years were enrolled. The phoria test was performed for 25 male and 57 female participants. The NRA/PRA test was performed for 20 male and 60 female participants. The AMP test was performed for 23 male and 58 female participants.

Repeatability in the Measurements of Phoria, NRA/PRA, and ODAMP

Table 1 displays the results of the IPVF instrument, and Table 2 displays the results of the phoropter. The ICCs of the repeatability of the three consecutive measurements using the IPVF instrument were: Phoria_D, 0.958; Phoria_N, 0.964; NRA, 0.872; PRA, 0.937; and ODAMP, 0.929. The Sw of the repeatability of the three consecutive measurements using the IPVF instrument were: Phoria_D, 0.084; Phoria_N, 0.101; NRA, 0.014; PRA, 0.032; and ODAMP, 0.048. The ICCs of the repeatability of the three consecutive measurements using the phoropter were: Phoria_D, 0.978; Phoria_N, 0.983; NRA, 0.914; PRA, 0.732; and ODAMP, 0.952. The Sw of the repeatability of the three consecutive measurements using the phoropter were: Phoria_D, 0.047; Phoria_N, 0.079; NRA, 0.011; PRA, 0.065; and ODAMP, 0.036.

Table 1 The intraclass correlation coefficient (ICC) value, 95% confidence interval of ICC, the within-subject standard deviation (Sw), and the standard deviation (SD) for the new instrument
Full size table
Table 2 The intraclass correlation coefficient (ICC) value, 95% confidence interval of ICC, the within-subject standard deviation (Sw), and the standard deviation (SD) for phoropters
Full size table

Agreement in the Measurements of Phoria, NRA/PRA, and ODAMP

The Bland–Altman plots were used to evaluate the agreement between the two instruments (Figs. 2, 3, 4, 5 and 6). The mean difference of the mean test values of phoria, NRA/PRA, and AMP between the two instruments were close to zero. Almost all dots in the Bland–Altman plots were plotted with the accepted threshold of 95% LoAs between the two instruments, and the 95% LoAs were all narrow.

Fig. 2
figure 2

Bland–Altman plot of horizontal phoria at distance to evaluate the agreement between the intelligent inspection platform and the phoropter

Full size image
Fig. 3
figure 3

Bland–Altman plot of horizontal phoria at near to evaluate the agreement between the intelligent inspection platform and the phoropter

Full size image
Fig. 4
figure 4

Bland–Altman plot of negative relative accommodation to evaluate the agreement between the intelligent inspection platform and the phoropter

Full size image
Fig. 5
figure 5

Bland–Altman plot of positive relative accommodation to evaluate the agreement between the intelligent inspection platform and the phoropter

Full size image
Fig. 6
figure 6

Bland–Altman plot of the accommodative amplitude of the right eye to evaluate the agreement between the intelligent inspection platform and the phoropter

Full size image

Discussion

The IPVF is a newly developed visual function measurement and diagnostic instrument that can automatically measure visual function. However, no research has been conducted on its accuracy. In the present study, we compared the platform with the conventional equipment phoropter and concluded that the new instrument performed well in both repeatability and agreement tests.

A previous study with a similar age range (mean age: 24.3 years) as this study compared the repeatability of the AMP test using the minus lens methods and found that the coefficient of repeatability (COR) was ± 1.43 D [7]. Antona et al. [8]. also assessed the repeatability of the same methods of measuring AMP. Their results indicated that the COR was ± 2.52 D, drawing a conclusion with high repeatability. The current study demonstrated resembles results with the ICC almost close to 1. In addition, researchers have focused on the degree of repeatability of AMP using different methods (push-up, push-down, and minus lens methods) and discovered different conclusions [15, 16]. Therefore, further research is still needed to determine the repeatability of the IPVF for AMP measurement using push-up and push-down methods. Previous studies on the repeatability of phoria using von Graefe, Thorington, and modified Thorington methods showed a high repeatability, regardless of the test method used [17,18,19]. In the current study, high repeatability (ICC > 0.75) among the three consecutive measurements was achieved for all tests performed using the new instrument, with ICCs of Phoria_D and Phoria_N over 0.950 (0.958 and 0.964, respectively). When the phoropter was used, high repeatability was generated in Phoria_D, Phoria_N, NRA, and ODAMP, with ICCs over 0.90 (0.91–0.98), and a lower but acceptable repeatability in PRA (ICC = 0.73). The data described above indicate that the repeatability of all included tests using the new instrument was equal to or slightly better than that using the phoropter.

The measurement of visual function using the phoropter can be affected by the following factors: proficiency and operating standards of examiners, consistency and effectiveness of examination language, comprehensive ability of the participant, accuracy of participant feedback, the participant’s ability to accurately comprehend the examiner’s feedback, and the examiner’s ability to accurately calculate and record the test result. However, the standard operation and language protocol of the new instrument ensured that each test had the same tone of speaking, speed of voice prompt, and lens switch. In addition, in the program setting, each test was independent; therefore, it was not affected by the previous test. Instead of manual operation, the automatic program and instrument ensured a standard measurement procedure, consistent test language, accurate reception of the participant feedback, and precise calculation and recording of the test result. This might be the reason for the better repeatability of the PRA test by the platform than that of the phoropter.

The Bland–Altman plot is a commonly used and widely accepted method to test the agreement between two measurement methods [20]. Therefore, it was adopted to test the agreement between the new instrument and the phoropter. The maximum absolute 95% LoAs of Phoria_D, Phoria_N, NRA, PRA, and ODAMP were 5.682 Δ, 4.750 Δ, 0.564 D, 1.772 D, and 2.073 D, respectively (Table 3). The 95% LoAs were all narrow, indicating high agreement between the two instruments. Rami Aboumourad et al. used dynamic retinoscopy to obtain objective measures of AMP and compared them with the subjective push-up technique. Since the difference less than 2 D means no systematic bias, they concluded that the two techniques can be interchangeable in clinical use (range of 95% LoA: −1.87 to 1.92 D)[21]. In addition, compared with the study on assessing the agreement of four different methods used to determine the accommodative response [22], in which the 95% LoAs in the Bland–Altman plot in the accommodative response test was 0.87–1.74 D, the 95% LoA between the new instrument and the phoropter in the current study was similar. Because of the automaticity of the integrated platform, during the test the participant is only required to follow simple steps of operation, mostly pressing buttons on the control pad. The instrument provides clear and easy-to-follow instructions, and relatively reliable measurement results can be easily obtained. Phoropters are manually operated by an optometrist throughout the process. Multiple participants can be tested simultaneously, under the supervision of only one optometrist. Thus, the intelligent integrated platform may be more convenient.

Table 3 The lower and upper limit and the range of 95% limit of agreement (LoA) in Bland–Altman plots and the mean difference (MD) between the measurement done by the new instrument and the phoropter
Full size table

This study had some limitations. In addition to examiner-related factors, the factors related to participants should be considered as well, because the participants’ comprehension and ability to provide accurate feedback, which depends on their age, education level, etc., also affect the accuracy of test results. The age range in this study was 18–25 years, and the interaction language was Mandarin. Whether the validity and reliability of the test results using the new instrument can be extended to a wider age range and to people who are accustomed to using other dialects requires further investigation. In the current study, a measurement of the time required for the examination was not done.

Conclusions

The intelligent integrated platform, in comparison with the phoropter, performed well in both repeatability and agreement tests. The platform is a new type of intelligent visual function inspection instrument with good reliability, and it could be an alternative for clinicians to obtain visual function measurements with improved efficiency and fewer subjective errors.