Validation of the Cut-Plane Method for Accurate Measurement of Side-Branch Ostium Area in Coronary Bifurcation Lesions: A 3D-OCT Analysis

Abstract

Objective

This study aimed to validate the effectiveness, accuracy, and feasibility of the cut-plane method for measuring the side branch (SB) ostium area in three-dimensional optical coherence tomography (3D-OCT) pullbacks performed in the main branch (MB).

Methods

A total of 109 sets of OCT pullbacks from the MB and SB of coronary artery bifurcation lesions were analyzed using Vivolight OCT software. Measurements of the SB ostium area from the MB and SB pullbacks were analyzed. Measurements of the SB ostium area from the actual SB pullback were used as a reference. 3D cut-plane analysis was used to estimate the correlations and mean errors with the reference measurements.

Results

Thirty-four sets of OCT images from the C7XR system and 75 sets from the Cornaris^TM system were analyzed using Vivolight software. There was a strong correlation between the reference measurements of the SB ostium area and the measurements obtained through 3D cut-plane analysis in the overall dataset (r = 0.925). This correlation was observed consistently with both the C7XR system (r = 0.955) and Cornaris^TM system (r = 0.900). Similar results were found in subset analyses of true and nontrue bifurcations (r = 0.936; r = 0.898, respectively) and in left main (LM) or non-LM bifurcation subsets (r = 0.932; r = 0.873, respectively).

Conclusions

There were strong correlations between measurements of the SB ostium area by 3D-OCT and the reference measurements, and thus may be a reliable and accurate alternative to direct OCT pullback examinations of the SB.

Abbreviations

CBL, Coronary bifurcation lesion

ICCs, Intra-class correlation coefficients

LM, Left main

MB, Main branch

OCT, Optical coherence tomography

PCI, Percutaneous coronary intervention

SB, Side branch

3D, Three-dimensional

INTRODUCTION

Coronary bifurcation lesions (CBLs) present significant challenges in percutaneous coronary intervention (PCI), with low procedural success rates and a high risk of complications accounting for approximately 15-20% of all treated cases.^1,2 The European Bifurcation Club has consistently emphasized the need to improve and standardize PCI for CBL.³ While the provisional approach is considered the gold standard for most bifurcation lesions, accurate lesion stratification is crucial in selecting the optimal stenting technique.^4,5 Occlusion of the side branch (SB) during coronary bifurcation treatment is associated with adverse clinical outcomes, highlighting the importance of precise pre- and post-operative assessments of the SB ostium for risk stratification and prognosis prediction in bifurcation lesion management. In current clinical practice, coronary angiography and qualitative comparative analysis have limitations in providing accurate information about the SB ostium due to overlap and foreshortening caused by X-ray system constraints.⁶ While conventional two-dimensional intravascular imaging allows for reliable evaluation of vessel geometry and lesion dimensions beyond the coronary bifurcation anatomy,⁷ additional intravascular imaging is necessary to visualize the SB and assess lesion characteristics. However, this adds to the complexity, time requirements, and potential complications of the procedure.

Advancements in image processing and computer hardware have enhanced the convenience and accuracy of comprehensive assessments of plaque distribution in bifurcation lesions. The utilization of three-dimensional (3D) reconstruction, based on the high-resolution capabilities of optical coherence tomography (OCT), offers potential advantages in various stages of bifurcation interventions. It not only enables assessment of SB compromise after coronary bifurcation stenting, but also ensures optimal guide wire recrossing position after main-branch (MB) stenting.^8-10 Karanasos et al. developed a novel approach using two-dimensional OCT that allows for a more precise assessment of the SB ostium from MB pullbacks, thus reducing the need for SB instrumentation.¹¹ The cut-plane analysis approach for SB ostial assessments from a main-vessel pullback is of significant clinical importance. However, the limited tissue penetration of OCT has been shown to result in the exclusion of a substantial amount of data in previous studies due to out-of-view artifacts, thereby limiting its clinical applicability.

Vivolight OCT software, developed by Vivolight Corporation in Shenzhen, China, is an innovative OCT imaging and analysis system that utilizes a series of diagnostic algorithms to convert and interpret OCT data into 3D images. This software overcomes the limitations associated with out-of-view artifacts in conventional two-dimensional OCT analysis, thereby allowing for the analysis of this data in 3D conditions and increasing the clinical applicability of the cut-plane analysis approach for SB ostium. The objective of this study was to validate the efficacy, accuracy, and feasibility of 3D cut-plane analysis for SB ostium from MB pullbacks using Vivolight OCT software.

METHODS

Study population

A retrospective analysis was conducted on consecutive patients who underwent OCT examinations for CBL at a large tertiary hospital between January 2020 and March 2022. Patients were included in the study if they had good quality frequency-domain OCT pullbacks from both the MB and SB of the coronary bifurcation, based on the European Bifurcation Club’s CBL definition and Medina classification.² Five patients with poor image quality (e.g., incomplete flushing) or artifacts (e.g., guidewire shadow) impeding visualization of the SB ostium were excluded. In total, 100 patients with 109 bifurcation lesions were enrolled in the study.

Ethical considerations

The study protocol was approved by the Institutional Review Board of our institution, and it adhered to the principles outlined in the Declaration of Helsinki. Informed consent was obtained from all patients for the use of their imaging data in clinical research.

OCT image acquisition

OCT images were acquired using two different systems: the C7XR system (St Jude/LightLab Imaging, Inc., Westford, MA, USA) with a Dragonfly catheter (St Jude) (n = 35), and the Cornaris^TM system (Vivolight Corporation, Shenzhen, China) with a Pathfinder164 catheter (Vivolight Corporation) (n = 74). The OCT catheter was advanced over the guide wire and positioned distal to the selected bifurcation site in either the MB or SB. Automated OCT pullback was performed at a speed of 20 mm/s while injecting contrast medium (Iodixanol 370, Visipaque^TM, GE HealthCare, Ireland) at a rate of 3-4 mL/s through the guiding catheter. The same procedure was repeated for the other branch. The acquired images were digitally stored for offline analysis using Vivolight OCT software (Vivolight Imaging Systems, Shenzhen, China).

OCT image analysis

OCT measurements were conducted using Vivolight OCT software (Vivolight Imaging Systems, Shenzhen, China) by an experienced observer (QH Jin) at our institution’s dedicated core laboratory. The analysis involved measuring the area of the SB ostium and the MB area at the carina level. The SB ostium was identified as the initial frame in which the carina became visible after the SB pullback procedure. The automated detection algorithm provided by the Vivolight OCT software was used to outline the contour of the SB ostium, with manual adjustments made as necessary.¹²

SB pullback: SB reference ostium area

The SB ostium was identified during the SB pullback procedure, and the area of the ostium was quantified using the Vivolight OCT offline software for both the C7XR and Cornaris^TM systems.¹²

MB pullback: SB ostium area for the 3D cut-plane measurement

The SB ostium area of interest was evaluated using manual 3D-OCT cut-plane analysis. The digital files containing the images obtained from the MB pullbacks were imported into the Vivolight OCT software. Subsequently, the image data were reconstructed through 3D rendering, as shown in Figure 1. The SB ostium within the images was manually identified and analyzed using the ‘cut-plane’ model.¹¹ The detailed steps of the ‘cut-plane’ analysis are presented in Figure 2.

Figure 1.

Schematic diagram of three-dimensional (3D) side-branch model. Review a set of data using the Vivolight optical coherence tomography (OCT) software, which displays the cross-section view and L view by default (A); Enabling the side branch analysis mode in the 3D function of the software, which allows quantitative measurement and analysis of side branches in 3D (B).

Figure 2.

OCT image patterns and study illustrations. Schematic diagrams of two types of OCT data acquisition for the same SB. The green color represents the OCT pullback from the SB for OCT data acquisition, while the blue color represents the OCT pullback from the MB for OCT data acquisition. The green dotted ellipse indicates the cross-sectional image of the OCT of the SB obtained by pulling back from the SB, and the red dotted ellipse indicates the OCT of the SB obtained by using the cut-plane method of pulling back from the MB. The red ellipse dotted line represents the schematic diagram of the SB OCT cross-sectional image obtained from the MB pullback using the cut-plane method. The measurement of SB ostium area as a gold standard is performed using SB OCT cross-sectional images obtained from SB pullback (A). Schematic diagrams B, C, E, and F illustrate the measurement of SB ostium area using the cut-plane method based on the pullback OCT image of the MB. In diagram B, the L view is adjusted until the bifurcation carina angle is clearly visible and minimal, and then the position of the white line (i.e., the direction of the cut-plane) is manually adjusted to be perpendicular to the centerline of the SB. Diagram C shows the position of the cut plane in 3D view, assisting in determining the perpendicularity of the white line to the centerline of the SB. Based on the white line in diagram B, the 2D cut plane image is obtained in diagram E, where the SB profile can be manually adjusted to obtain the SB ostium area measurement results. Diagram F visualizes the structure of the SB profile in the 2D cut-plane image in 3D view. MB, main branch; OCT, optical coherence tomography; SB, side branch; 2D, two-dimensional; 3D, three-dimensional.

Statistical analysis

Statistical analysis was performed using SPSS version 25.0 (IBM, Chicago, IL, USA). A two-tailed p value of < 0.05 was considered statistically significant. Categorical variables were presented as frequencies and percentages and compared using the chi-square test or Fisher’s exact test, as appropriate. Continuous variables were reported as mean ± standard deviation and compared using the Student’s t-test. The paired t-test was used to assess differences in absolute measurement errors. Spearman correlation coefficient analysis was used to evaluate correlations between SB ostium measurements obtained through SB pullback and those obtained through 3D cut-plane analysis. The significance of differences in correlation coefficients was assessed using the z-test after Fisher’s r-to-z transformation. Bland-Altman plots were used to assess the measurement error for each method in relation to the reference measurements. Intra- and interobserver variability were evaluated using intraclass correlation coefficients (ICCs). Subgroup analysis was performed to investigate the impact of SB ostium diameter and the presence of lesions at the SB ostium on the accuracy of SB ostium area recognition after 3D reconstruction. Subgroups were defined based on the presence of left main (LM) bifurcation lesions and true bifurcation lesions.

Variability assessment

Interobserver and intraobserver variability were evaluated in two distinct cohorts, each consisting of 20 patients. Intraobserver variability was reevaluated in one cohort by the same observer (JG Cui) after a one-month interval. Interobserver variability was assessed in another cohort by two experienced interventional cardiologists (JG Cui and QH Jin).

RESULTS

Baseline and angiographic results

Data were collected from a cohort of 100 patients, encompassing a total of 109 bifurcations. Table 1 summarizes the baseline characteristics of the patients, and Table 2 provides detailed angiographic features. Among the participants, 78% were male, with a mean age of 61.3 ± 9.1 years. The majority of bifurcation lesions were located in the LM branch (29.4%) and the left anterior descending and diagonal branches (61.5%). According to the Medina classification, there were 65 true bifurcation lesions (59.6%) and 44 nontrue bifurcation lesions (40.4%). There were no significant differences in bifurcation location and the percentage of true bifurcation between the C7XR and Cornaris^TM systems.

Table 1. Baseline characteristics.

Number of patients/lesions	100/109
C7XR system	35/35
CornarisTM system	65/74
Sex (male), n (%)	78 (78.0)
Age (years)	61.3 ± 9.1
Clinical presentation, n (%)
Stable angina	14 (14.0)
Unstable angina	78 (78.0)
Myocardial infarction	8 (8.0)
Coronary risk factors, n (%)
Diabetes mellitus	33 (33.0)
Dyslipidemia	16 (16.0)
Hypertension	57 (57.0)
Smoking	46 (46.0)

All values are presented as n (%) or mean ± standard deviation.

Table 2. Angiographic characteristics.

Angiographic characteristics	Total (109)	C7XR system (35)	CornarisTM system (74)	p
Imaged bifurcation, n (%)				0.119
Left main	32 (29.4)	14 (40)	18 (24.3)
LAD diagonal	67 (61.5)	19 (54.3)	48 (64.9)
LCX-OM	6 (5.5)	0 (0)	6 (8.1)
RCA-PDA	4 (3.7)	2 (5.7)	2 (2.7)
Medina classic, n (%)				0.637
True bifurcation	65 (59.6)	22 (62.9)	43 (58.1)
0,1,1	14 (12.8)	6 (17.1)	8 (10.8)
1,0,1	5 (4.6)	2 (5.7)	3 (4.1)
1,1,1	46 (42.2)	14 (40.0)	32 (43.2)
Nontrue bifurcation	44 (40.4)	13 (37.1)	31 (41.9)
0,0,1	4 (3.7)	1 (2.9)	3 (4.1)
1,1,0	14 (12.8)	5 (14.3)	9 (12.2)
0,1,0	21 (19.3)	6 (17.1)	15 (20.3)
1,0,0	5 (4.6)	1 (2.9)	4 (5.4)

All values are presented as n (%).

LAD, left anterior descending artery; LCX, left circumflex artery; OM, obtuse marginal branch; PDA, posterior descending artery; RCA, right coronary artery.

OCT analysis

Comparisons of measurements between 3D cut-plane analysis and pullback from the SB demonstrated no significant difference in the ostium area for the total sample, or for the C7XR system and Cornaris^TM system (Figure 3). The overall correlation coefficient between the SB bifurcation ostium lumen area obtained from SB pullback and 3D cut-plane analysis from MB pullback was 0.925 [95% confidence interval (CI): 0.873 to 0.962]. In subgroup analysis of the C7XR system and Cornaris^TM system, the correlation coefficients were 0.955 (95% CI: 0.928 to 0.991) and 0.900 (95% CI: 0.833 to 0.954), respectively (Table 3). There was no significant difference in the correlation coefficient between the two systems (0.955 vs. 0.900, p = 0.052). Similar results were observed in subset analysis of true and nontrue bifurcations (0.936 vs. 0.898, p = 0.226), as well as in analysis of LM vs. non-LM bifurcations (0.932 vs. 0.873, p = 0.134).

Figure 3.

Box plots for pairwise comparison of 3D cut-plane analysis and pullback from SB. Pairwise comparison of the SB ostium area measurements for total (A) and for C7XR system (B) and (C) for Cornaris^TM system via 3D cut-plane analysis and pullback from the SB. SB, side branch; 3D, three-dimensional.

Table 3. Optical coherence tomography measurements and corresponding errors.

SB ostium lumen area (mm2)	SB pullback	MB pullback: 3D cut-plane analysis
	Reference measurement (mean ± SD)	Measurement (mean ± SD)	Error (95% limits of agreement)	Pearson correlation coefficient (95% CI)
Total (109)	3.23 ± 2.13	3.27 ± 2.26	0.032 (-0.132 to 0.196)	0.925 (0.873 to 0.962)
C7XR system (35)	3.80 ± 2.32	3.93 ± 2.53	0.135 (-0.125 to 0.395)	0.955 (0.928 to 0.991)
CornarisTM system (74)	2.97 ± 1.99	2.95 ± 2.07	-0.017 (-0.227 to 0.194)	0.900 (0.833 to 0.954)
True bifurcation (65)	2.71 ± 1.85	2.68 ± 1.99	-0.034 (-0.210 to 0.141)	0.936 (0.878 to 0.975)
Nontrue bifurcation (44)	4.01 ± 2.30	4.14 ± 2.17	0.129 (-0.192 to 0.450)	0.898 (0.801 to 0.964)
LM (32)	4.77 ± 2.70	4.77 ± 2.87	0.002 (-0.373 to 0.377)	0.932 (0.869 to 0.975)
Non-LM (77)	2.60 ± 1.43	2.64 ± 1.61	0.446 (-0.133 to 0.223)	0.873 (0.791 to 0.944)

Values are presented as the mean ± SD or mean (95% limits of agreement).

CI, confidence intervals; LM, left main; MB, main branch; SB, side branch; SD, standard deviation; 3D, three dimensional.

The correlations and Bland-Altman plots (Figure 4) demonstrated that 3D cut-plane analysis of the SB ostium area, both in the total sample and in the C7XR system and Cornaris^TM system subgroups, exhibited higher reliability characterized by excellent fit, lower mean error, and narrower limits of agreement.

Figure 4.

Correlation and Bland-Altman plots of the SB ostium area by 3D cut-plane analysis and pullback from SB. Area measurements from SB pullback are used as the reference. Correlation of the SB ostium area by 3D cut-plane analysis and pullback from SB of total (A) and of C7XR system (B) and of Cornaris^TM system (C). Bland-Altman plots of the SB ostium area by 3D cut-plane analysis and pullback from SB of total (D) and of C7XR system (E) and of Cornaris^TM system (F). Red dotted lines correspond to measurement error, and blue dotted lines correspond to 95% confidence intervals. MB, main branch; SB, side branch; 3D, three-dimensional.

Table 3 presents the absolute values of the measurements in the ostium area. There was no significant difference between the area measurement obtained by pullback from the SB and that by 3D cut-plane analysis (3.23 ± 2.13 mm² vs. 3.27 ± 2.26 mm², p = 0.700). The reference measurements of the SB ostium area obtained by 3D cut-plane analysis from MB pullback were comparable to those obtained by the C7XR system (3.80 ± 2.32 mm² vs. 3.93 ± 2.53 mm², p = 0.299) and the Cornaris^TM system (2.97 ± 1.99 mm² vs. 2.95 ± 2.07 mm², p = 0.874). Furthermore, the SB ostium area measurements for true and nontrue bifurcations obtained by pullback from the SB and 3D cut-plane analysis were similar (2.71 ± 1.85 mm² vs. 2.68 ± 1.99 mm², p = 0.702; 4.01 ± 2.30 mm² vs. 4.14 ± 2.17 mm², p = 0.423, respectively). The differences between the two methods were also not statistically significant in the LM and non-LM subgroups (4.77 ± 2.70 mm² vs. 4.77 ± 2.87 mm², p = 0.993; 2.60 ± 1.43 mm² vs. 2.64 ± 1.61 mm², p = 0.620, respectively).

Variability analysis

The results of intra- and interobserver agreements were very high for area measurements [ICCs: 0.990 (0.976 to 0.996) and 0.993 (0.983 to 0.997), respectively] (Table 4).

Table 4. Interobserver and intraobserver variability.

	Observer 1	Observer 2	Difference (Observer 1-Observer 2)	ICC (95% CI)	Observer 1-Measurement 1	Observer 1-Measurement 2	Difference (Measurement 1-Measurement 2)	ICC (95% CI)
SB ostium lumen area (mm2)	2.86 ± 1.69	2.80 ± 1.7	0.06 (-0.07 to 0.19)	0.993 (0.983 to 0.997)	2.94 ± 1.35	2.85 ± 1.44	0.09 (-0.03 to 0.22)	0.990 (0.976 to 0.996)

Values are presented as the mean ± standard deviation or mean (95% limits of agreement).

CI, confidence intervals; ICC, intraclass correlation coefficient; SB, side branch.

DISCUSSION

The main findings of this study are as follows. First, the use of Vivolight OCT software to analyze the SB ostium area, along with 3D rendering using a cut-plane model, showed a high level of reliability and agreement with the reference measurements obtained from SB pullbacks. Second, the 3D cut-plane analysis method had a relatively low mean error in measuring the SB ostium area. Third, the 3D cut-plane analysis method demonstrated excellent repeatability in assessing the ostium areas of the branches. Fourth, the Vivolight OCT software was found to be compatible with images acquired from both the C7XR and Cornaris^TM systems. Overall, our results revealed that the 3D cut-plane analysis method shows promise as an alternative to direct OCT examinations for measuring the ostium area in the SB. However, further research is needed to validate these findings and explore their clinical implications.

The anatomy and morphology of bifurcation lesions are important factors in making technical decisions and achieving successful bifurcation PCI.¹⁰ The use of a routine crossover stenting strategy for bifurcation lesions has been associated with shorter procedure and fluoroscopy times, as well as lower rates of procedure-related complications.¹³ However, it is important to note that the SB ostium may be affected after the index procedure, leading to worsened stenosis or complete blockage. Therefore, accurate assessment of bifurcation lesions, especially the corresponding SB ostium, is crucial for determining the appropriate subsequent treatment strategy.

The V-RESOLVE score has been validated as a promising tool for predicting SB occlusion in PCI. It includes six angiographic predictors, one of which is the bifurcation angle.¹⁴ However, it is challenging to achieve the anatomically defined optimal bifurcation viewing angle based on coronary angiography in approximately half of cases due to limitations imposed by current X-ray systems.⁶ This often leads to over- or underassessment of the SB ostium, caused by overlapping and foreshortening effects.

Assessing the severity of bifurcation lesions solely through angiography can be difficult. While conventional intravascular imaging allows for the identification of detailed cross-sectional information about the vessel wall, it does not provide precise morphological details about the bifurcation lesions.

While measuring the SB ostium area through a pullback from the SB allows for direct and accurate measurements, it involves additional procedures that may increase the risk of complications. Additionally, it is not feasible to conduct pullback measurements in small-diameter SB or bifurcation lesions with large angles. However, with the advancements in intravascular imaging techniques, particularly the use of 3D rendering based on OCT, visualization of the SB ostium has become more accessible. This advancement greatly facilitates the evaluation of bifurcation lesions and helps in selecting interventional treatment strategies.^9,10,15-17 In this study, the absolute area of the SB ostium was evaluated using 3D cut-plane analysis with Vivolight OCT software, both in the overall analysis and with the C7XR and Cornaris^TM systems. The findings showed excellent agreement with the reference measurements obtained through direct SB pullback, with minimal mean error associated with the 3D cut-plane analysis. The 3D cut-plane model consistently exhibited exceptional agreement with the reference measurement method for evaluating the branch ostium area in true and nontrue bifurcation lesions, as well as subgroups categorized by LM and non-LM bifurcations. These results highlight the clinical applicability of the 3D cut-plane model. The high agreement among observers in this study emphasizes the accuracy and reliability of the measurement technique used to assess SB ostium area. These findings enhance the overall reliability and validity of the study results, further supporting the clinical utility and applicability of the measurement method in evaluating bifurcation lesions.

In a previous study, the efficacy and accuracy of a two-dimensional cut-plane model in evaluating bifurcation ostial area were demonstrated.¹¹ However, patients with out-of-view artifacts caused by guide wire positioning, which hindered complete visualization of the SB ostium, were excluded. This exclusion could potentially limit the clinical applicability of 3D-OCT for assessing bifurcation lesions.¹¹ In the present study, patients with incomplete visualization of the SB ostium due to guide wire positioning were included, excluding only imaging sets with guide wire shadowing or poor image quality that hindered 3D rendering and contouring of the SB ostium. The robustness and reliability of 3D-OCT cut-plane analysis was validated using Vivolight OCT software, which demonstrated high correlation and excellent stability. The real-time online 3D reconstruction capability of Vivolight OCT software enables efficient 3D rendering of the target bifurcation and accurate contouring of the SB ostium, facilitating precise analysis of SB ostium area and morphology. To ensure data integrity, we used raw data from the C7XR and Cornaris^TM systems for 3D rendering and analysis with Vivolight OCT software. The compatibility of Vivolight OCT software with images obtained from the C7XR and Cornaris^TM systems greatly enhances its clinical applicability compared to previous studies.

As technology continues to advance, the use of 3D-OCT-based image reconstruction and analysis is becoming increasingly important in evaluating bifurcation lesions and selecting appropriate interventional strategies. Accurate measurement and analysis enable a better assessment of lesion severity, SB morphology and size, and the selection of optimal treatment approaches. It is crucial to adhere to strict indications and operational guidelines to ensure the safety and efficacy of 3D-OCT. Additionally, further research is needed to determine the optimal utilization and benefits of 3D-OCT in clinical practice, as well as its potential advantages and prospects when combined with other imaging techniques.

Limitations

Several limitations should be acknowledged in this study. First, it is important to note that this study was conducted retrospectively with a relatively small sample size and potential operator preference, which may introduce selection bias. This bias is indirectly supported by the overrepresentation of lesions in the LM and left anterior descending-diagonal bifurcation locations. Second, the manual modifications made to the cut-plane model after 3D rendering using Vivolight OCT software may have introduced variability in the measurements among different analysts. However, it is worth noting that there was high agreement among observers in area measurements, suggesting minimal variability. Third, the Vivolight OCT software used in this study was unable to provide information on the maximum and minimum diameters of the ostium, limiting its comprehensive assessment. Further improvements are needed in this regard. Fourth, due to the retrospective design, a time analysis for the SB ostium area measurements could not be obtained. As a result, a direct comparison of the time required for each measurement method, which is an important indicator of clinical applicability, could not be conducted. Finally, all of the imaging data were acquired before PCI or stent implantation, and caution should be exercised when considering the clinical applicability of the findings to patients requiring PCI. Future research should explore the feasibility of using 3D cut-plane analysis for assessing SB ostium after stenting.

CONCLUSIONS

The assessment of ostium area using Vivolight OCT software and 3D cut-plane analysis showed a strong correlation and excellent agreement with reference measurements obtained from SB pullbacks. Vivolight OCT software was compatible with images from both C7XR and Cornaris^TM systems. With its real-time online analysis capabilities, this approach has the potential to be an alternative to direct branch OCT examinations. Further studies are needed to validate and explore the clinical usefulness of this method.

DECLARATION OF CONFLICT OF INTEREST

The authors declare that they have no known competing financial interests or personal relationships with manufacturer companies that could have appeared to influence the work reported in this paper.

FUNDING

None.

AUTHOR CONTRIBUTIONS

Jianguo Cui: Conceptualization, Methodology, Formal analysis Writing - Original draft preparation. Qinhua Jin: Supervision, Validation, Writing - Reviewing and Editing. Xun Wu: Resources, Software, Methodology. Yundai Chen: Writing - Reviewing and Editing.