Analysis of ocular biometry in Korean using swept-source optical coherence tomography Authors: Jae Ryong Song, Da Eun Yoon, Youngsub Eom, Jong Suk Song, Joon Young Hyon, Hyun Sun Jeon Published: 2025 ABSTRACT AbstractTo evaluate and compare ocular biometric parameters in both adults and children within the Korean population using swept-source optical coherence tomography (SS-OCT; IOL Master 700). Ocular biometry data were exported from SS-OCT between January 2018 and October 2021—total 6,760 eyes of 6,760 subjects (5,091 adults and 1,669 children under 18 years) which were successfully measured were included. Axial length (AL), anterior keratometry (K), total keratometry (TK), anterior chamber depth (ACD), lens thickness (LT), central corneal thickness (CCT), white to white distance (WTW) were analyzed according to sex, age, and AL groups. All parameters were significantly different between male and female (P < 0.05), with male presenting longer AL, flatter K, deeper ACD, and thicker CCT. A significant negative correlation was observed between AL and Km (r = -0.329, p < 0.001). In children (9.5 2.8 years), AL showed a steep increase until age 11, followed by a slower growth rate, and age had significantly correlated with all parameters except LT and had the greatest significant correlation with AL (r = 0.476, p < 0.001), Meanwhile, in adults (58.7 ± 14.7 years), age and LT demonstrated the greatest significant correlation (r = 0.622, p < 0.001). In conclusion, this study revealed significant differences in ocular biometry by sex and age in the Korean population. Males exhibited larger ocular dimensions than females. With age, AL growth slowed after age 11 in children, while LT increased and ACD decreased in adults. These findings underscore the importance of considering both sex and age in clinical assessments and provide valuable data for optimizing refractive management and monitoring myopia progression.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-33221-1.Keywords: Biometry, Axial length, Eye, Age factors, Sex characteristics, Tomography, Optical coherenceSubject terms: Optical materials and structures, Visual system, Eye diseases, Lens diseases Introduction With an aging society, cataract surgeries are increasingly common, making accurate ocular biometry essential for refractive outcomes1,2. Advances in technology now provide higher precision in these measurements. While ultrasound-based A-scan were once standard, their contact method made risks of corneal error and damage. The IOL master 500, using partial coherence interferometry (PCI), improved repeatability and accuracy, though severe lens opacities or poor visual acuity still impact results. Recently, IOL master 700 which uses swept source optic coherence tomography (OCT) was developed to enhance accuracy and reproducibility, even with dense cataracts3,4. In assessing refractive error, particularly for identifying myopia, axial length (AL) measurement provides a precise and non-invasive tool, useful for both children and adults. Young children with mild hyperopia are considered at risk for developing myopia5, but refractive error assessments often encounter inaccuracies due to over-accommodation when cycloplegia is not used6. Moreover, subjective refraction typically has an accuracy of within ± 0.50 diopters (D)7,8. However, recent technology using PCI and swept-source techniques has advanced AL measurements to a higher level of precision, reducing errors to few microns9,10. These methods allow for rapid and non-invasive measurements, which are simple for practitioners and young patients alike. Utilizing AL alone, or in conjunction with other ocular measures (e.g., corneal curvature), provides a valuable approach for identifying children at risk of myopia due to excessive ocular growth5,11. Ocular biometric parameters such as AL, anterior chamber depth(ACD) and anterior keratometry(K) have different distributions according to race, age and sex12. While several studies have examined the ocular biometry of Europeans13, study on Korean populations have primarily focused on adult patients with cataract, which reported on the correlation between age and other ocular biometric parameters obtained through SS-OCT14. However, studies on ocular biometric parameters in children and their differences between sexes are limited. Therefore, the purpose of this study is to investigate and compare the characteristics of ocular biometric parameters in both adults and children within the Korean population using a swept-source optical coherence tomography (SS-OCT; IOL Master 700). Results Demographics and ocular biometry distribution of study population In the age and sex distribution of the study population participating of this study, 1,669 eyes from children (mean age: 9.59 ± 2.84 years; 811 males, 858 females) and 5,091 eyes from adults (mean age: 58.74 ± 14.78 years; 2,524 males, 2,567 females) were included. There was no significant difference in sex proportions between two groups (chi square test, p = 0.485). The axial length of children showed a symmetrical normal distribution, while that of adults was right-skewed, which aligns with previous study15,16. Additionally, the AL of adults was longer than that of children (24.60 ± 1.85 mm vs. 24.45 ± 1.59 mm, p = 0.003; Fig. 1). In both adults and children, the medium AL group constituted the majority (children: 72.8%, adults: 73.7%), while the proportion of individuals with a long AL was 15.9% in children and 20.0% in adults. Axial length distribution in children and adults. This histogram illustrates the frequency distribution of axial length measurements in each age group. Children exhibited symmetrical distribution, while adults showed a right-skewed distribution. The adult group had a significantly longer mean AL than the child group (24.60 mm vs. 24.45 mm, p = 0.003). When examining the relationship between AL and K, 52.7% fell into the medium AL and medium K groups which indicates that most cases had an average AL and K values. Instances of flat K in the short AL group (comprising 25 eyes, 0.4%) and steep K in the long AL group (comprising 122 eyes, 1.8%) were uncommon (supplementary Table 1), These findings align with a significant negative correlation between AL and Km (r = -0.329, p < 0.001), indicating that as AL increases, Km decreases. The prevalence of patients with TCA ≤ 0.5 D, 0.5D ~ 1.0D, and ≥ 1.0 D were 21.1%, 31.4%, and 45.5%, respectively (Supplementary Fig. 1). Ocular biometry in children (N = 1,669 eyes) All ocular biometric measurements showed significant differences between males and females in children. Males had longer AL (male vs. female: 24.76 ± 1.56 mm vs. 24.16 ± 1.54 mm, p < 0.001), deeper ACD (3.78 ± 0.26 mm vs. 3.67 ± 0.24 mm, p < 0.001), greater CCT (555.78 ± 33.82 μm vs. 551.79 ± 33.29 μm, p = 0.015), and larger WTW (12.26 ± 0.39 mm vs. 12.10 ± 0.39 mm, p < 0.001). Moreover, males showed flatter Km (43.11 ± 1.41 D vs. 43.74 ± 1.43 D, p < 0.001), flatter TKm (43.11 ± 1.41 D vs. 43.75 ± 1.43 D, p < 0.001), and thinner LT (3.37 ± 0.18 mm vs. 3.40 ± 0.17 mm, p = 0.002). These results highlight significant sex-based differences in ocular biometric parameters, with males having larger structural dimensions and flatter corneal curvatures, while females exhibited steeper keratometry readings and thicker LT (Table 1). Table 1Comparison of ocular biometric parameters between sexes in children.Biometric parameterTotal(N = 1669)Male(N = 811)Female(N = 858)*P valueAL (mm)24.45 ± 1.5924.76 ± 1.5624.16 ± 1.54< 0.001Km(D)43.43 ± 1.4543.11 ± 1.4143.74 ± 1.43< 0.001Kflat (D)42.55 ± 1.4242.24 ± 1.3942.85 ± 1.39< 0.001Ksteep (D)44.32 ± 1.6343.97 ± 1.5844.64 ± 1.61< 0.001TKm (D)43.44 ± 1.4543.11 ± 1.4143.75 ± 1.43< 0.001TKflat (D)42.58 ± 1.4242.27 ± 1.4042.87 ± 1.39< 0.001TKsteep (D)44.30 ± 1.6343.95 ± 1.5044.63 ± 1.61< 0.001ACD (mm)3.72 ± 0.253.78 ± 0.263.67 ± 0.24< 0.001LT (mm)3.385 ± 0.173.37 ± 0.183.40 ± 0.170.002CCT (㎛)553.73 ± 33.60555.78 ± 33.82551.79 ± 33.290.015WTW (mm)12.18 ± 0.40212.26 ± 0.3912.10 ± 0.39< 0.001* Data are presented as mean ± SD.* Independent T test.AL, axial length; K, keratometry; TK, total keratometry; ACD, anterior chamber depth; LT, lens thickness; CCT, central corneal thickness; WTW, white to white. Comparison of ocular biometric parameters between sexes in children. * Data are presented as mean ± SD. * Independent T test. AL, axial length; K, keratometry; TK, total keratometry; ACD, anterior chamber depth; LT, lens thickness; CCT, central corneal thickness; WTW, white to white. AL had a strong positive correlation with age and ACD (r = 0.491, r = 0.518, respectively, P < 0.001). A significant negative correlation was observed between ACD and LT (r = -0.438, P < 0.01), and a negative correlation was observed between WTW and corneal curvature, specifically with Km (R = -0.508, P < 0.001) and TKm (R = -0.489, P < 0.001). (Fig. 2-A) The most rapid growth in AL occurs up to age 11, where the mean AL increases from 23.22 mm (95% CI: 23.02–23.43) in children under 6 years to 25.27 mm (95% CI: 25.05–25.48) at age 11. Beyond this age, the rate of increase slows, with mean AL reaching 25.75 mm (95% CI: 25.38–26.12) at age 15 and 26.02 mm (95% CI: 25.22–26.82) by age 17 (Table 2; Fig. 3). Although a slight decrease in mean AL was observed at age 12 compared to age 11 (25.09 mm vs. 25.27 mm), this primarily reflected the female subgroup (24.46 mm vs. 25.06 mm). The male subgroup demonstrated continued AL increase (25.49 mm to 25.64 mm), and the 95% confidence intervals between ages 11 and 12 overlapped substantially in females (24.77–25.35 mm vs. 24.09–24.84 mm), indicating that this fluctuation likely represents statistical variation rather than a true biological phenomenon. Correlations between ocular biometry parameters in children (A) and adults (B). The heatmap displays Pearson correlation coefficients between age, axial length (AL), mean keratometry (Km), steep keratometry (K1), flat keratometry (K2), mean total keratometry (TKm), steep total keratometry (TK1), flat total keratometry (TK2), anterior chamber depth (ACD), lens thickness (LT), central corneal thickness (CCT), and white-to-white distance (WTW). Positive correlations are represented by red, and negative correlations by blue, with the intensity of the color indicating the strength of the correlation. Table 2Axial length (AL) distribution by age and sex in children.Age(years)Number (M/F)Total, Mean[95% CI]Male, Mean(95% CI)Female, Mean(95% CI)Total1669 (811/858)24.45 (24.37–24.53)24.76 (24.65–24.86)24.16 (24.06–24.26)< 6217 (113/104)23.22 (23.02–23.43)23.51 (23.26–23.75)22.92 (22.60-23.24)7223 (103/120)23.54 (23.37–23.71)23.90 (23.65–24.16)23.23 (23.02–23.44)8230 (117/113)24.15 (23.99–24.31)24.42 (24.19–24.64)23.87 (23.66–24.09)9297 (132/165)24.35 (24.20–24.50)24.59 (24.35–24.82)24.17 (23.97–24.36)10152 (62/90)24.80 (24.58–25.03)25.05 (24.74–25.36)24.63 (24.32–24.95)11126 (60/66)25.27 (25.05–25.48)25.49 (25.18–25.81)25.06 (24.77–25.35)12136 (74/62)25.09 (24.83–25.35)25.64 (25.28–25.95)24.46 (24.09–24.84)1392 (50/42)25.29 (24.98–25.61)25.49 (24.99–25.99)25.06 (24.74–25.42)1475 (38/37)25.54 (25.24–25.84)25.69 (25.29–26.08)25.39 (24.92–25.86)1567 (32/35)25.75 (25.38–26.12)26.24 (25.76–26.71)25.30 (24.78–25.83)1635 (20/15)25.82 (25.36–26.27)26.31 (25.76–26.85)25.16 (24.47–25.85)1719 (10/9)26.02 (25.22–26.82)26.35 (25.31–27.39)25.65 (24.21–27.09)Axial length (mean, mm) with 95% confidence intervals (CIs) is presented for all subjects, as well as separately for males and females across different age groups (years). Axial length (AL) distribution by age and sex in children. Axial length (mean, mm) with 95% confidence intervals (CIs) is presented for all subjects, as well as separately for males and females across different age groups (years). Age-related changes in axial length by sex in Korean children (ages 2–17 years). Mean axial length (AL) with 95% confidence intervals (shaded areas) across ages 2–17 years, stratified by sex. Males (blue circles) consistently showed longer AL than females (red squares) at all ages. Both sexes exhibited rapid AL growth until age 11, followed by growth deceleration. Ocular biometry in adults (N = 5,091 eyes) In adults, all ocular biometric measurements showed statistically significant differences between males and females. Males had longer AL (24.92 ± 1.79 mm vs. 24.30 ± 1.86 mm, p < 0.001), deeper ACD (3.33 ± 0.41 mm vs. 3.14 ± 0.42 mm, p < 0.001), greater CCT (551.30 ± 34.66 μm vs. 543.17 ± 33.71 μm, p < 0.001), and larger WTW (11.95 ± 0.41 mm vs. 11.73 ± 0.40 mm, p < 0.001). They also had flatter Km (43.44 ± 1.50 D vs. 44.09 ± 1.49 D, p < 0.001), lower TKm (43.43 ± 1.51 D vs. 44.05 ± 1.50 D, p < 0.001), and thinner LT (4.40 ± 0.47 mm vs. 4.44 ± 0.45 mm, p = 0.001).(Table 3) These findings align with the patterns observed in children, with males having larger and flatter ocular structures, while females exhibited smaller, steeper, and more crowded ocular dimensions. Table 3Comparison of ocular biometric parameters between sexes in adults.BiometricparametersTotal(N = 5091)Male(N = 2524)Female(N = 2567)*P valueAL (mm)24.60 ± 1.8524.92 ± 1.7924.30 ± 1.86< 0.001Km (D)43.77 ± 1.5343.44 ± 1.5044.09 ± 1.49< 0.001Kflat (D)43.28 ± 1.5342.95 ± 1.5043.61 ± 1.49< 0.001Ksteep (D)44.25 ± 1.6043.92 ± 1.5844.57 ± 1.56< 0.001TKm (D)43.74 ± 1.5443.43 ± 1.5144.05 ± 1.50< 0.001TKflat (D)43.25 ± 1.5442.93 ± 1.5143.57 ± 1.51< 0.001TKsteep (D)44.24 ± 1.6143.93 ± 1.5944.54 ± 1.58< 0.001ACD (mm)3.24 ± 0.433.33 ± 0.413.14 ± 0.42< 0.001LT (mm)4.42 ± 0.464.40 ± 0.474.44 ± 0.450.001CCT (㎛)547.20 ± 34.41551.30 ± 34.66543.17 ± 33.71< 0.001WTW (mm)11.84 ± 0.4211.95 ± 0.4111.73 ± 0.40< 0.001* Data are presented as mean ± SD.* Independent T test.AL, axial length; K, keratometry; TK, total keratometry; ACD, anterior chamber depth; LT, lens thickness; CCT, central corneal thickness; WTW, white to white. Comparison of ocular biometric parameters between sexes in adults. * Data are presented as mean ± SD. * Independent T test. AL, axial length; K, keratometry; TK, total keratometry; ACD, anterior chamber depth; LT, lens thickness; CCT, central corneal thickness; WTW, white to white. There was a negative correlation between age and AL (r = -0.396, P < 0.001), which is the opposite of the positive correlation seen in children. It’s also significant that age has a strong positive correlation with LT (r = 0.680, P < 0.001) and a negative correlation with ACD (r = -0.496, P < 0.001). There is a substantial negative correlation between LT and ACD(r = − 0.713, P < 0.001). (Fig. 2-B) Discussion This study examines the differences in ocular biometric parameters between sexes in children and adults, as well as age-related changes in children within the Korean population using swept-source optical coherence tomography (SS-OCT; IOL Master 700). Significant sex-specific differences were observed, with males generally exhibiting longer AL, deeper ACD, and flatter corneal curvature, while females demonstrated steeper keratometry and thicker LT in both age groups. In children, AL showed a rapid increase until approximately age 11, followed by a slower growth rate thereafter, reflecting dynamic growth patterns during childhood. Additionally, in adults, LT increased, and ACD decreased with age, highlighting structural changes in the eye associated with aging. These findings underscore the importance of considering both sex and age when assessing ocular biometry for refractive management and risk evaluation. Our research revealed statistically significant differences between sexes across all assessed biometric parameters. Specifically, in our sample, females exhibited smaller ocular dimensions than males across all these biometric aspects. It is a widely recognized fact that females generally exhibit shorter stature and lower body weight compared to males, which extends to ocular development as well17. These findings are consistent with a previous study on adolescent twin pairs, which explored sex differences in ocular biometric measurements18. They included opposite-sex and same-sex twin pairs, revealing that boys generally exhibited greater AL and WTW measurements, but lower keratometry readings in both steep and flat axes compared to girls. Moreover, Wang L et al. confirms that female sex, older age, shorter AL, shallow ACD, and larger LT are significant risk factors for angle closure19. Given the statistically significant differences in ocular biometry between sexes, clinicians must consider sex-based biometric differences in clinical settings. The average AL in adults in our study was found to be 24.60 mm, which is longer than values reported in studies from other countries; 23.95 mm in Israel, 23.55 mm in Singapore, 23.44 mm in Australia, 23.43 mm in Germany, 23.69 mm in the USA and 24.32 mm in China13,15,16,20–22. Considering high prevalence of myopia in Chinese populations, it is notable that AL in this study exceeds that of Chinese. This finding highlights the need for careful monitoring of AL elongation in Korean population to manage myopia progression and related complications. A previous study on the Korean population reported a mean AL of 24.22 ± 1.87 mm, comparable to our findings14. The slightly longer AL in our study may be due to demographic differences, as our population included a higher proportion of males (49.6% in our study vs. 38% in Yoon et al.) and younger participants (mean age: 58.7 years vs. 66.1 years). Despite these minor differences, both studies confirm that the Korean population tends to have relatively longer AL compared to other populations. Considering the strong association between increased AL and sight-threatening conditions such as myopic maculopathy, retinal detachment, and glaucoma23, these findings highlight the importance of regular AL monitoring to manage myopia progression and prevent complications. In children, AL exhibits a growth pattern characterized by early accelerated elongation followed by a gradual deceleration, consistent with previous studies11,24. Longitudinal research, where AL was measured annually using A-scan ultrasonography, has also demonstrated a strong correlation between axial elongation and myopia progression, with AL stabilizing around 16 years of age on average11. This emphasizes the importance of monitoring axial elongation as a key marker for managing myopia progression during childhood. Our findings are consistent with a large-scale study in Chinese children, which developed age-specific and sex-specific percentile curves for AL and axial length/corneal radius (AL/CR)24. This study demonstrated that AL continued to increase with age, and a higher AL/CR ratio correlated strongly with a more myopic refractive error. The percentile charts also revealed that changes in AL/CR ratios were most significant at younger ages, highlighting the need for careful monitoring of eye growth during childhood​. Furthermore, establishing normative values for axial length based on age is essential for identifying rapid myopia progression and guiding timely, personalized interventions in children. Rose LVT et al. demonstrated that AL growth rate can serve as an indicator for rapid progression of myopia in children25. Rapid AL growth, such as an increase exceeding 0.1 mm per six months, can serve as a critical indicator of fast progression, prompting proactive management strategies. This approach highlights the importance of baseline measurements and growth rate of axial length monitoring in tailoring treatment plans. By utilizing age-specific normative values, clinicians can effectively categorize patients and implement timely interventions to mitigate myopia progression. In adults, lens thickness increases significantly with age, while ACD decreases, highlighting a strong inverse relationship between LT and ACD. A study combining A scan ultrasonography and Scheimpflug photography found an average increase of 24 μm per year in LT among subjects aged 16 to 65 years, indicating a strong correlation between age and changes in LT26. Additionally, a retrospective study of 5,273 Korean cataract patients demonstrated significant correlations between age and ocular biometric parameters, including a positive correlation with LT (r = 0.484) and negative correlations with ACD (r = -0.357) and AL (r = -0.344), which is consistent with our findings14. The aging of the lens is associated with a loss of elasticity; the lens gradually becomes rigid and loses its ability to accommodate, which further increases the anterior-posterior diameter of the lens27. Considering that eyes at increased risk for primary angle-closure have decreased ACD and a large lens, age-related changes in lens thickness and ACD should be carefully accounted for when selecting the IOL power calculation formula and determining the optimal timing for surgery in aged patients28. Despite these insights, this study has limitations inherent to its cross-sectional design, as it does note account for time-dependent changes in optical biometry. Additionally, because this study was conducted at a single institution, the findings may not fully reflect the characteristics of the broader Korean population. Furthermore, our study includes children who visited the clinic for myopia treatment, which may lead to longer AL compared to the general population, which limits this study’s generalizability. Moreover, the absence of intraocular pressure (IOP) is also a limitation of this study, as IOL Master 700 does not measure IOP and such data were not systematically available in our retrospective biometry database. Previous studies have reported associations between IOP and ocular biometry29. Future prospective studies incorporating systematic IOP measurements would provide valuable insights into the relationship between IOP and ocular biometric parameters in the Korean population, particularly for glaucoma risk assessment and myopia progression monitoring. Nonetheless, this study is meaningful in that it provides data on optical biometry measurements specifically for pediatric patients. In conclusion, this study analyzed ocular biometry in Koreans using SS-OCT, revealing notable differences by sex and age. We found significant differences in ocular measurements between sexes, with males exhibiting larger ocular dimensions. Age-based trends included increasing LT and decreasing AL in adults, while AL increased with age in children. These findings highlight the need to consider age and sex in ocular assessments and may assist in optimizing intraocular lens selection and refractive outcomes in cataract surgery, as well as guiding future research on myopia progression in this population. This research offers essential biometric data to support personalized eye care in the Korean population. Methods This is a retrospective single center cross-sectional analysis of consecutive eyes for which ocular biometric parameters were measured with an SS-OCT. We exported the raw data from an SS-OCT (IOL Master 700), which were obtained between January 2018 and October 2021 at Seoul National University Bundang Hospital. All subjects who underwent IOLMaster 700 biometry at our institution during the study period and obtained reliable measurements were included in the analysis. Among a total of 26,790 eyes of 14,373 patients whose ocular biometry analysis was performed, one eye per patient was included in this study. When biometric measurements were taken from both eyes, one eye was randomly selected using a computer-generated random number sequence (using the RAND function in Microsoft Excel) to avoid potential correlation issues from including both eyes of the same individual. For patients with measurements from only one eye, that eye was included in the analysis. The exclusion criteria were as follows: silicone oil filled eye, history of refractive surgery, ocular biometry that were not successfully measured, and eyes that have had cataract surgery(pseudophakia). Finally, total 6,760 eyes of 6760 patients were included. The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board (IRB) (IRB N: B-2405-898-002) of Seoul National University Bundang Hospital. Informed consent was waived owing to the retrospective nature of this study by the IRB of Seoul National University Bundang Hospital. Personal identifiable information was removed from all cases during the study to protect privacy. Sex and date of birth were recorded before biometric measurement. Ocular biometry measurements were performed using an IOL master 700 SS-OCT optical biometer. Among the exported raw data from IOL master 700, AL, standard keratometry (K), total keratometry (TK), anterior chamber depth (ACD), lens thickness (LT), central corneal thickness (CCT) and white to white distance (WTW) were reviewed for analysis. K was calculated as 1000 × (1.3375–1)/Ranterior. Mean keratometry (Km) was defined by (Kflat + Ksteep)/2. Mean total keratometry (TKm) was defined by (TKflat + TKsteep)/2. Astigmatism in K and TK was defined as the difference between the steep and flat keratometric values. To assess and compare the optical biometry between children and adults, the subjects were classified into two age groups; children (2–17 years), and adults (18–96 years). Moreover, to assess the correlation between AL and other parameters, eyes were classified into three AL subgroups according to AL; short AL (< 22.5 mm), medium AL (22.5 ~ 26.0 mm), long AL (> 26.0 mm)30,31. Eyes were also classified into three K subgroups according to Km; Flat Km (< 42.0 D), medium Km (42.0 D ~ 45.0 D), steep Km (> 45.0 D)32,33. The normality of the data was assessed using the Shapiro-Wilk test, and homoscedasticity was assessed using Levene’s test. Independent T test was used for comparing optical biometry between sexes. Correlations between all ocular biometric parameters were examined by Pearson correlation coefficients. Data were entered into an electronic spreadsheet (Excel; Microsoft), and SPSS Statistics software (version 23; IBM Corporation, Armonk, NY, USA) was used to perform statistical analyses. Results were expressed as the mean ± standard deviation (SD) and P values less than 0.05 were considered statistically significant. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Author contributions All authors meet the authorship criteria as defined by Scientific Reports, including substantial contributions to the conception and design of the work, acquisition and analysis of data, drafting and revising the manuscript, and final approval of the version to be published. H.S.J. conceived and designed the study. J.R.S. performed the statistical analysis, drafted the manuscript, and contributed to data collection. D.E.Y. collected the data. Y.S.E. and J.S.S. revised the manuscript and approved the final version. J.Y.H. and H.S.J. interpreted the results. All authors agree to be personally accountable for their own contributions and to ensure the integrity of the entire work. Funding This work was supported by the SNUBH Research Fund (No. 09-2019-0002 [H.S.J.]) The funding organization had no role in the design or conduct of this research. Data availability The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.