Skip Navigation
Skip to contents

Clin Endosc : Clinical Endoscopy

Search
Advanced Search
OPEN ACCESS
mobile search button mobile menu button

Articles

Page Path
HOME > Clin Endosc > Volume 59(1); 2026 > Article
Review Recent advancement in size measurement during endoscopy
Hye Kyung Jeonorcid, Gwang Ha Kimorcid
Clinical Endoscopy 2026;59(1):1-8.
DOI: https://doi.org/10.5946/ce.2025.070
Published online: May 23, 2025

Department of Internal Medicine, Pusan National University School of Medicine and Biomedical Research Institute, Pusan National University Hospital, Busan, Korea

Correspondence: Gwang Ha Kim Department of Internal Medicine, Pusan National University School of Medicine and Biomedical Research Institute, Pusan National University Hospital, 179 Gudeok-ro, Seo-gu, Busan 49241, Korea E-mail: doc0224@pusan.ac.kr
• Received: March 4, 2025   • Revised: March 22, 2025   • Accepted: March 24, 2025

© 2026 Korean Society of Gastrointestinal Endoscopy

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 6,177 Views
  • 554 Download
  • 2 Web of Science
  • 3 Crossref
  • 3 Scopus
next
Full Article

Download PDF Download PDF

  • Accurate lesion size measurement is essential in endoscopic practice as it influences treatment strategies, surveillance decisions, and clinical outcomes, especially in colorectal polyps. Traditional measurement techniques, including visual estimation and biopsy forceps, have significant interobserver variability and procedural inefficiencies. Recent advancements in digital measurement technologies, including virtual scale endoscopy (VSE) and artificial intelligence (AI)-assisted virtual rulers, have addressed these limitations. VSE projects a virtual scale onto endoscopic images, enhancing measurement precision and reducing variability. Several studies have demonstrated its superior accuracy compared with conventional methods; however, limitations such as increased procedure time and operator training requirements persist. AI-assisted virtual rulers utilize deep learning algorithms to automate lesion size estimation, significantly improving reproducibility and diagnostic reliability. Although these technologies offer promising improvements, challenges remain, including real-time integration, standardization, and regulatory approval. Future research should focus on refining AI models, expanding validation studies, and optimizing their usability in routine practice. A hybrid approach that combines AI automation with real-time digital tools may enhance the precision and efficiency of endoscopic lesion assessment, ultimately improving patient outcomes.
  • Keywords: Artificial intelligence; Endoscopy; Polyps; Size measurement
Endoscopy plays a crucial role in the early detection and treatment of gastrointestinal (GI) lesions. The lesion size is critical in determining the need for resection, polyp surveillance intervals, and treatment selection. The risk of colorectal polyps with advanced pathologies increases with their size. Colorectal polyps ≥10 mm in size have a higher risk of metachronous advanced neoplasia and subsequent colorectal cancer.1 Therefore, accurate size measurement of colorectal polyps is important for adequate choice of resection technique and surveillance.2,3 Another vital aspect of accurate polyp size measurement results in the implementation of resect-and-discard and diagnose-and-leave strategies at the 5-mm threshold.4 However, conventional size estimation methods are subject to interobserver variability and lack standardization.
As the field of endoscopy continues to advance, the integration of innovative measurement technologies provides an opportunity to improve procedural efficiency and diagnostic accuracy. In recent years, research in this area has expanded, with numerous studies evaluating the effectiveness of new measurement techniques, such as virtual scale endoscopy (VSE), fiber laser-based endoscopic virtual rulers (VRs), and artificial intelligence (AI), compared with traditional methods (Table 1). This review evaluates recent research on these emerging technologies, their impact on clinical endoscopic practice, and potential future improvements in real-world applications.
Visual estimation
Visual estimation is the most commonly used method for measuring lesions in the GI tract. However, it is highly subjective, and its accuracy varies significantly among endoscopists. Several studies have shown that reliance on visual estimation leads to misclassification of colorectal polyp sizes, and accuracies range from 54% to 65%, potentially affecting clinical decision-making.5-9 One major drawback of visual estimation is its dependence on the experience of the endoscopist; the intraclass correlation coefficient for polyps >5 mm is low (0.13).10 While experienced endoscopists may have a better grasp of lesion size assessment, studies have shown that even among experts, discrepancies in measurement are common.11 Additionally, visual estimation does not provide a standardized approach to measurement, leading to inconsistencies in clinical practice. With the increasing need for precision in endoscopic procedures, relying solely on visual estimation has become less acceptable, emphasizing the importance of implementing more reliable measurement tools.
Biopsy forceps and endoscopic rulers
Biopsy forceps and endoscopic rulers provide a reference for estimating the lesion size. Although these tools improve accuracy compared with visual estimation, their utility is limited by procedural inefficiencies and difficulty in positioning relative to the lesion.12,13 The additional time required to use these methods, combined with their dependence on user experience, restricts their practicality in routine clinical practice. Additionally, inconsistencies in how these tools are applied can lead to discrepancies in lesion measurements, further complicating treatment decisions.
Recent studies have also highlighted the variability in measurements when using biopsy forceps.12,14,15 In some cases, the technique requires multiple attempts to achieve an accurate assessment, which can increase the procedure time and cause patient discomfort. Moreover, endoscopic rulers are not always practical, particularly when measuring lesions in hard-to-reach anatomical locations. Consequently, researchers have explored more advanced alternatives that leverage digital measurements and AI to enhance accuracy and efficiency.
Virtual scale endoscope in polyps
Nakatani et al.16 first developed a three-dimensional (3D) endoscopic system integrating multiple laser beams for real-time measurement. The system employs stereoscopic imaging, where multiple angles of the lesion are captured to construct a 3D model. Therefore, it corrects barrel distortions and enhances measurement accuracy by generating a 3D coordinate mapping of the lesion. Calibration techniques were used to adjust for lens distortion and improve the measurement precision. 3D imaging provides depth perception, which is particularly useful for assessing complex lesions.
VSE is a new endoscopic technique that helps estimate the size of lesions in the GI tract. VSE operates by projecting a virtual scale onto an endoscopic image to estimate lesion size in real time. The virtual scale function is operated using a combination of a new endoscope (EC-760S-A/M; Fujifilm Corp.) and virtual scale software (EW10-VM01; Fujifilm Corp.) (Fig. 1). Yoshioka et al.17 reported a preliminary study comparing polyp size estimation by VSE using biopsy forceps in 33 pseudo-polyps. The correction estimation rate by VSE was higher than that using biopsy forceps (60.6% vs. 39.4%, p=0.0028). In a prospective study by Shimoda et al.,5 10 endoscopists measured the size of simulated polyps in colon phantoms by conventional visual estimation and VSE estimation. The relative accuracy in relation to true polyp size was significantly higher for VSE estimation than for visual estimation (84.0% vs. 62.5%, p<0.001). VSE estimation had significantly less interobserver and intraobserver variations in polyp size estimation than visual estimation (p=0.034 and p=0.017, respectively). However, the mean of required times was significantly longer for estimation by VSE than by visual estimation (6.4 min vs. 2.9 min, p<0.001).
Djinbachian et al.18 conducted a prospective randomized trial to compare a novel VSE using a laser with biopsy forceps and endoscopic rulers in measuring simulated colorectal polyps. The study involved six endoscopists who performed 60 measurements at a 1:1:1 ratio using different tools. VSE showed a significantly higher accuracy (82.7%; 95% confidence interval [CI], 80.8%–84.8%) than biopsy forceps (78.9%; 95% CI, 76.2%–81.5%; p=0.02) and endoscopic rulers (78.4%; 95% CI, 76.0%–80.8%; p=0.006). The coefficient for estimating interobserver agreement among all endoscopists for VSE was significantly higher than that for biopsy forceps and endoscopic rulers (0.96 vs. 0.93, p<0.001). Additionally, VSE reduced misclassification rates, particularly for polyps >5 mm and >20 mm in size.
Minakata et al.19 recently reported a prospective multicenter study on the usefulness of VSE for 20 early lesions ≤20 mm in size each in the esophagus, stomach, and colon. The normalized difference (ND) in the VSE measurement and visual measurement of the lesion on the endoscopic resection specimen was 0.3%±8.8% and –1.7%±29.3%, respectively. The variability was significantly smaller in the ND of the VSE measurement than that in the visual measurement (standard deviation, 8.8% vs. 29.3%; p<0.001). No differences were observed in VSE measurements according to organ or morphology.
VSE allows clinicians to quickly assess lesion dimensions without requiring additional equipment. The ability to instantly visualize accurate lesion measurements enhances diagnostic precision and procedural efficiency. Moreover, standardization of lesion size measurement through VSE contributes to consistency among endoscopists, thereby reducing interobserver variability. This technology continues to evolve with advancements in display resolution and real-time computational speed to enhance its usability in various endoscopic procedures. However, the virtual scale is displayed only on the right side of the laser, which has a potential limitation in horizontal measurements in the vertical plane. The VSE measurement error also becomes larger as the scope is displaced at an angle to the target.17 Moreover, as with other endoscopic procedures, VSE may require training before clinical application; controlling an endoscope with a wide-angle lens to aim at a point is not easy for endoscopists.
AI-assisted VRs in polyps
AI has been proposed as an alternative method for measuring polyp sizes.20 Abdelrahim et al.21 reported a preliminary study on two computer vision techniques for binary classification of polyp size (≤5 mm vs. >5 mm) using the structure from motion (SfM, a photogrammetric imaging technique) and convolutional neural networks (CNN). The overall diagnostic accuracy of the SfM system was superior to endoscopists’ judgment in the pig colon model in polyp size classification as ≤5 mm or >5 mm (85.2% vs. 59.5%). The CNN demonstrated an accuracy of 80% in distinguishing polyps >5 mm from diminutive polyps in ten videos of human polyps.
Kwak et al.22 developed a bifurcation-to-bifurcation distance measuring method using the W-Net model for vessel segmentation in colonoscopy images and then compared its measurement with those obtained by eight endoscopists. This system measured the distance between the leaves of vessels on the colorectal mucosa as a reference for calculating polyp size. Endoscopists tended to underestimate the polyp size even when using the biopsy forceps method, especially when the polyps were >10 mm in size. This AI-assisted technique was highly accurate and reliable for measuring the size of colon polyps (concordance correlation coefficient [CCC], 0.961; 95% CI, 0.926–0.979), irrespective of polyp size, clearly outperforming the visual estimation and open biopsy forceps methods used by endoscopists.
Kuai et al.23 introduced an endoscopic VR that employs fiber laser technology combined with AI to achieve real-time lesion size measurement. The study included ex vivo and in vivo validation, where the VR demonstrated high accuracy compared with conventional rulers. Fiber laser-based technology enables non-contact measurement, eliminating the potential risks associated with physical instruments while enhancing measurement reliability.
Wang et al.24 developed a deep learning-based system (ENDOANGEL-CPS) to estimate colorectal polyp size in real time. This system consisted of two main models: a depth estimation model and a polyp segmentation model. The CCC between AI-based system estimation and in vitro-measured values were 0.89 and 0.93 in simulated colon images and multicenter videos of 157 polyps, respectively. The CCC of the AI-based system was significantly higher than that of all endoscopists (0.89 vs. 0.41, p<0.001), and the accuracy of the AI-based system for classifying polyp size into three groups (≤5 mm, 6–9 mm, ≥10 mm) was significantly higher than that of the endoscopists (89.9% vs. 54.7%, p<0.001).
AI-assisted VRs in esophageal varices
The bleeding risk of esophageal varices (EVs) is associated with variceal pressure; patients with large EVs have higher variceal pressure than those without.25 Therefore, the diameter of EVs is not only a predictor of EV bleeding but also an important reference for selecting the appropriate treatment modiality.26,27 It is also a crucial factor in evaluating the efficacy of EV treatment.27 Visual measurement during endoscopy is the most commonly used method for estimating the diameter of EVs; however, the measurement of EV diameter varies greatly among endoscopists. The EV manometer was developed and then used to measure the diameter of EVs.25 However, this device has important limitations, such as high cost and an additional bleeding risk during examination.28 Accordingly, VR, a type of AI-assisted software that enables noninvasive measurement of the diameter of EVs by endoscopy, has been developed. Jin et al.29 and Fang et al.30 evaluated AI-assisted VR using algorithms such as Gaussian filters, Canny edge detectors, and Hough circles for measuring EVs, and their findings indicated that VR significantly reduced measurement time and improved diagnostic accuracy compared with traditional EV manometers.
In a retrospective analysis of 345 patients with cirrhosis, The VR measurement of EV diameter was moderately correlated with endoscopist measurements (κ=0.591, p<0.001).30 The VR group with EV diameter >1 cm had a lower rebleeding rate after endoscopic treatment than the endoscopist group (3.8% vs. 11.3%, p=0.048). No significant differences were observed between the VR and endoscopist groups in terms of efficacy and safety outcomes for patients with an EV diameter of ≤1 cm.
In another recent clinical study including 128 patients diagnosed with portal hypertension and gastroesophageal varices, AI-based EV diameter measurement was positively correlated with portal pressure gradient (PPG) (r=0.521, p<0.001).31 In addition, the accuracy of EV diameter for diagnosing PPG function was 81.4% (95% CI, 72.0%–90.8%), indicating that when the EV diameter is >1.1 cm, PPG levels are likely to be >20 mmHg. Therefore, AI-assisted VR assigns precise numerical values to the variceal diameters, thereby improving standardization across operators. Additionally, AI models incorporate predictive analytics based on patient data and provide risk assessments for variceal bleeding. This integration of AI into EV measurement enhances decision-making and stratifies patients based on EV size to predict bleeding risk.
The evolution of endoscopic size measurement techniques has significantly improved the accuracy and standardization of lesion assessment. However, each method has advantages and limitations, necessitating careful consideration in clinical applications (Table 2).
VSE has emerged as a promising solution that utilizes real-time projection of a measurement scale onto an endoscopic image. Compared with traditional methods, VSE significantly reduces interobserver variability and enhances measurement reliability. Additionally, it allows for immediate assessment of lesion size without the need for additional tools. However, VSE has notable drawbacks, including the need for extended procedure time and specialized operator training. Its accuracy is also influenced by the scope angle, as measurement errors increase when the endoscope is displaced. Moreover, the current VSE system projects the measurement scale in only one direction, limiting its application to lesions with complex orientations. Further research is required to optimize VSE usability and expand its application to different GI pathologies.
AI-assisted VRs leverage deep learning algorithms to automate lesion size measurements, offering superior accuracy and consistency compared with conventional methods. These AI-driven models can process endoscopic images in real time, reduce endoscopist subjectivity, and improve diagnostic precision. Additionally, AI-assisted systems facilitate the standardization of lesion assessment and enhance reproducibility among different endoscopists. However, the effectiveness of AI-based measurements depends on the quality and diversity of training datasets. Variability in lesion morphology, lighting conditions, and endoscopic imaging parameters can affect AI performance, necessitating the ongoing refinement of algorithm adaptability. Moreover, the integration of AI systems into routine endoscopic practice requires seamless real-time processing without disrupting workflow efficiency. Ensuring regulatory approval and clinical validation through large-scale studies are crucial for its broader adoption.
Future studies should refine these emerging techniques to enhance their clinical applicability. AI models should incorporate adaptive learning algorithms to improve their performance across diverse patient populations. The integration of AI and VSE into real-time endoscopic workflows must be optimized to reduce procedural time without compromising accuracy. Most studies have focused on measuring colorectal polyp size. Correct size measurement is also crucial in decision-making for early GI neoplasms and subepithelial tumors.32-36 Therefore, further research on these lesions is needed. Standardization and regulatory approval are critical for ensuring widespread adoption. Large-scale multicenter clinical trials are essential to validate these advancements and facilitate their integration into routine endoscopic practice. A hybrid approach combining AI-driven automation with real-time digital measurement tools may shape the future of endoscopic lesion assessment and improve diagnostic precision and procedural efficiency.
Fig. 1.
Virtual scale endoscope. (A) Tip of virtual scale endoscope. (B, C) The software detects the position of the laser point in the endoscopic field of view and displays a virtual scale on the right side. To compare the lesion size using the virtual scale, the endoscopist positions the laser point on the left edge of the lesion. (D) Real measurement of a colorectal polyp. Provided by Fujifilm Corp. with permission.
ce-2025-070f1.jpg
Table 1.
Summary of recent studies on emerging technologies in endoscopic size measurement
Study Year Method Results
VSE
 Nakatani et al. 2007 Three-dimensional measurement endoscopic system with virtual rulers A novel endoscopic system equipped with four laser beam sources provided virtual rulers over endoscopic images, allowing physicians to measure lesions accurately.
 Yoshioka et al. 2021 VSE for real-time polyp measurement The virtual scale function of gastrointestinal endoscopy reduced measurement errors to ≤0.7 mm. Compared with biopsy forceps, VSE provided significantly more precise real-time polyp size estimations (p<0.001).
 Shimoda et al. 2022 Prospective study comparing VSE and visual estimation of polyp size VSE demonstrated an 84.0% accuracy in polyp size estimation compared with 62.5% for visual estimation. However, VSE required significantly longer measurement times (6.4 min vs. 2.9 min, p<0.001).
 Djinbachian et al. 2023 Randomized trial on comparison of VSE, biopsy forceps, and endoscopic ruler in polyp measurement VSE had a higher accuracy (82.7%) than biopsy forceps (78.9%) and endoscopic ruler (78.4%). VSE also reduced the rate of misclassification of polyps at clinically relevant size thresholds (5 mm, 10 mm, and 20 mm).
 Minakata et al. 2024 Prospective study on VSE accuracy for measuring early gastrointestinal lesions VSE measurement had significantly lower variability than visual estimation (p<0.001). VSE was particularly effective for accurately measuring lesions of various sizes and morphologies in real clinical settings.
AI-assisted VR in polyps
 Abdelrahim et al. 2022 Computer vision techniques for polyp classification SfM had an 85.2% accuracy, far surpassing the 59.5% accuracy of endoscopists in polyp size classification as ≤5 mm or >5 mm. The CNN showed an accuracy of 80% for distinguishing polyps >5 mm from diminutive polyps in ten videos of human polyps.
 Kwak et al. 2022 AI-based polyp size measurement using W-Net model The AI-based method significantly improved accuracy in polyp size estimation compared with visual estimation and biopsy forceps. The concordance correlation coefficient was 0.961, indicating a high level of agreement with true polyp size measurements.
 Wang et al. 2024 Deep learning-based real-time system (ENDOANGEL-CPS) for polyp size estimation The ENDOANGEL-CPS system had a much higher accuracy in estimating polyp size than human endoscopists (89.9% vs. 54.7%, p<0.001). It significantly reduced inappropriate surveillance recommendations (1.5% vs. 16.6% for endoscopists, p<0.001).
AI-assisted VR in esophageal varices
 Jin et al. 2023 AI-based VR for measuring esophageal varices diameter VR provided equivalent accuracy to esophageal varix manometer but with significantly reduced measurement time (31 s vs. 159 s, p<0.01). VR can reduce unnecessary interventions, decrease the risk of complications, and serve as a cost-effective tool for esophageal varix assessment.
 Fang et al. 2024 VR-based diameter measurement in esophageal variceal endoscopic therapy In a multicenter study of 345 patients with cirrhosis, VR-based esophageal variceal diameter measurements significantly reduced rebleeding rates after endoscopic therapy for varices >1 cm (3.8% vs. 11.3%, p=0.048).
 Mao et al. 2024 AI-based VR for measuring the diameter of esophageal varices A strong correlation was found between AI-measured esophageal varix diameter and the portal pressure gradient, with r=0.521 (p<0.001).

VSE, virtual scale endoscopy; AI, artificial intelligence; VR, virtual ruler; SfM, structure from motion; CNN, convolutional neural network.

Table 2.
Comparison of endoscopic size measurement techniques
Advantage Disadvantage
Visual estimation Simple, quick to perform Subjective, experience-dependent
No additional equipment needed Lacks standardization
High interobserver variability
Biopsy forceps/endoscopic rulers More accurate than visual estimation Time-consuming
Can be used in real time Requires precise positioning, making it difficult for some anatomical locations
Virtual scale endoscopy Real-time lesion size measurement Needs special equipment
Reduces interobserver variability Requires training
Highly accurate Prolonged procedural time
Accuracy affected by scope angle and lesion orientation
AI-assisted virtual rulers Automated, highly accurate Variability of performance
Real-time processing Requires AI integration
Highly accurate Regulatory approval and clinical validation are necessary

AI, artificial intelligence.

  • 1. He X, Hang D, Wu K, et al. Long-term risk of colorectal cancer after removal of conventional adenomas and serrated polyps. Gastroenterology 2020;158:852–861.ArticlePubMedPMC
  • 2. Moond V, Loganathan P, Malik S, et al. Cold snare polypectomy versus cold endoscopic mucosal resection for small colorectal polyps: a meta-analysis of randomized controlled trials. Clin Endosc 2024;57:747–758.ArticlePubMedPMCPDF
  • 3. Kim J, Gweon TG, Kwak MS, et al. Survey of the actual practices used for endoscopic removal of colon polyps in Korea: a comparison with the current guidelines. Gut Liver 2025;19:77–86.ArticlePubMedPMC
  • 4. ASGE Technology Committee, Abu Dayyeh BK, Thosani N, et al. ASGE technology committee systematic review and meta-analysis assessing the ASGE PIVI thresholds for adopting real-time endoscopic assessment of the histology of diminutive colorectal polyps. Gastrointest Endosc 2015;81:502–516.ArticlePubMed
  • 5. Shimoda R, Akutagawa T, Tomonaga M, et al. Estimating colorectal polyp size with a virtual scale endoscope and visual estimation during colonoscopy: prospective, preliminary comparison of accuracy. Dig Endosc 2022;34:1471–1477.ArticlePubMedPDF
  • 6. Kim JH, Park SJ, Lee JH, et al. Is forceps more useful than visualization for measurement of colon polyp size? World J Gastroenterol 2016;22:3220–3226.ArticlePubMedPMC
  • 7. Sakata S, McIvor F, Klein K, et al. Measurement of polyp size at colonoscopy: a proof-of-concept simulation study to address technology bias. Gut 2018;67:206–208.ArticlePubMed
  • 8. Utsumi T, Horimatsu T, Nishikawa Y, et al. Short educational video to improve the accuracy of colorectal polyp size estimation: multicenter prospective study. Dig Endosc 2020;32:1074–1081.ArticlePubMedPDF
  • 9. Beukema KR, Simmering JA, Brusse-Keizer M, et al. Factors influencing endoscopic estimation of colon polyp size in a colon model. Clin Endosc 2022;55:540–548.ArticlePDF
  • 10. Elwir S, Shaukat A, Shaw M, et al. Variability in, and factors associated with, sizing of polyps by endoscopists at a large community practice. Endosc Int Open 2017;5:E742–E745.ArticlePubMedPMC
  • 11. Chaptini L, Chaaya A, Depalma F, et al. Variation in polyp size estimation among endoscopists and impact on surveillance intervals. Gastrointest Endosc 2014;80:652–659.ArticlePubMed
  • 12. Izzy M, Virk MA, Saund A, et al. Accuracy of endoscopists' estimate of polyp size: a continuous dilemma. World J Gastrointest Endosc 2015;7:824–829.ArticlePubMedPMC
  • 13. Turner JK, Wright M, Morgan M, et al. A prospective study of the accuracy and concordance between in-situ and postfixation measurements of colorectal polyp size and their potential impact upon surveillance. Eur J Gastroenterol Hepatol 2013;25:562–567.ArticlePubMed
  • 14. Moug SJ, Vernall N, Saldanha J, et al. Endoscopists' estimation of size should not determine surveillance of colonic polyps. Colorectal Dis 2010;12:646–650.ArticlePubMed
  • 15. Eichenseer PJ, Dhanekula R, Jakate S, et al. Endoscopic mis-sizing of polyps changes colorectal cancer surveillance recommendations. Dis Colon Rectum 2013;56:315–321.ArticlePubMed
  • 16. Nakatani H, Abe K, Miyakawa A, et al. Three-dimensional measurement endoscope system with virtual rulers. J Biomed Opt 2007;12:051803.ArticlePubMed
  • 17. Yoshioka M, Sakaguchi Y, Utsunomiya D, et al. Virtual scale function of gastrointestinal endoscopy for accurate polyp size estimation in real-time: a preliminary study. J Biomed Opt 2021;26:096002.ArticlePubMedPMC
  • 18. Djinbachian R, Taghiakbari M, Haumesser C, et al. Comparing size measurement of colorectal polyps using a novel virtual scale endoscope, endoscopic ruler or forceps: a preclinical randomized trial. Endosc Int Open 2023;11:E128–E135.ArticlePubMedPMC
  • 19. Minakata N, Ikematsu H, Kiyomi F, et al. Usefulness of virtual scale endoscope for early gastrointestinal lesions. DEN Open 2024;5:e386.ArticlePubMedPMC
  • 20. Rey JF. As how artificial intelligence is revolutionizing endoscopy. Clin Endosc 2024;57:302–308.ArticlePubMedPMCPDF
  • 21. Abdelrahim M, Saiga H, Maeda N, et al. Automated sizing of colorectal polyps using computer vision. Gut 2022;71:7–9.ArticlePubMedPMC
  • 22. Kwak MS, Cha JM, Jeon JW, et al. Artificial intelligence-based measurement outperforms current methods for colorectal polyp size measurement. Dig Endosc 2022;34:1188–1195.ArticlePubMedPDF
  • 23. Kuai Y, Zhou S, Sun B, et al. Use of an endoscopic virtual ruler based on the fiber laser principle and artificial intelligence technology. Endoscopy 2025;57:86–87.ArticlePubMedPMC
  • 24. Wang J, Li Y, Chen B, et al. A real-time deep learning-based system for colorectal polyp size estimation by white-light endoscopy: development and multicenter prospective validation. Endoscopy 2024;56:260–270.ArticlePubMed
  • 25. Scheurlen C, Roleff A, Neubrand M, et al. Noninvasive endoscopic determination of intravariceal pressure in patients with portal hypertension: clinical experience with a new balloon technique. Endoscopy 1998;30:326–332.ArticlePubMed
  • 26. de Franchis R; Baveno V Faculty. Revising consensus in portal hypertension: report of the Baveno V consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 2010;53:762–768.ArticlePubMed
  • 27. Singh S, Chandan S, Vinayek R, et al. Comprehensive approach to esophageal variceal bleeding: from prevention to treatment. World J Gastroenterol 2024;30:4602–4608.ArticlePubMedPMC
  • 28. Kong DR, Xu JM, Zhang L, et al. Computerized endoscopic balloon manometry to detect esophageal variceal pressure. Endoscopy 2009;41:415–420.ArticlePubMed
  • 29. Jin J, Dong B, Ye C, et al. A noninvasive technology using artificial intelligence to measure the diameter of esophageal varices under endoscopy. Surg Laparosc Endosc Percutan Tech 2023;33:282–285.ArticlePubMed
  • 30. Fang Z, Bai Y, Mao Y, et al. Role of virtual ruler-based diameter measurement in endoscopic therapy for cirrhotic esophageal varices: a retrospective multicenter study. Can J Gastroenterol Hepatol 2024;2024:8823825.ArticlePubMedPMC
  • 31. Mao Y, Fang Z, He Y, et al. Correlation between the diameter of esophageal varices measured using a virtual ruler under endoscopy and portal pressure gradient. Front Med (Lausanne) 2024;11:1443581.ArticlePubMedPMC
  • 32. Ishihara R. Surveillance for metachronous cancers after endoscopic resection of esophageal squamous cell carcinoma. Clin Endosc 2024;57:559–570.ArticlePubMedPMCPDF
  • 33. Park CH, Yang DH, Kim JW, et al. Clinical practice guideline for endoscopic resection of early gastrointestinal cancer. Clin Endosc 2020;53:142–166.ArticlePubMedPMCPDF
  • 34. Jeon HK, Kim GH. Endoscopic resection for superficial non-ampullary duodenal epithelial tumors. Gut Liver 2025;19:19–30.ArticlePubMedPMC
  • 35. Choe Y, Cho YK, Kim GH, et al. Prevalence, natural progression, and clinical practices of upper gastrointestinal subepithelial lesions in Korea: a multicenter study. Clin Endosc 2023;56:744–753.ArticlePubMedPMCPDF
  • 36. Panzuto F, Parodi MC, Esposito G, et al. Endoscopic management of gastric, duodenal and rectal NETs: position paper from the Italian Association for Neuroendocrine Tumors (Itanet), Italian Society of Gastroenterology (SIGE), Italian Society of Digestive Endoscopy (SIED). Dig Liver Dis 2024;56:589–600.ArticlePubMed

Figure & Data

REFERENCES

    Citations

    Citations to this article as recorded by  
    • Accuracy of Virtual Scale Endoscopy in colorectal polyp size measurement: a Grading of Recommendations Assessment, Development and Evaluation–assessed pairwise and network meta-analysis
      Mohamed S. Elgendy, Mohamed Rifai, Amira M. Taha, Islam Rajab, Abdulrahman Maged, Mohamed A. Elgamasy, Hosam I. Taha, Mohamed Abuelazm, Babu P. Mohan, Douglas G. Adler
      Gastrointestinal Endoscopy.2026; 103(5): 904.     CrossRef
    • Lyon endoscopic submucosal dissection score: a preprocedure prediction model for operating time in colorectal endoscopic submucosal dissection
      Elena De Cristofaro, Jean Grimaldi, Diana Giannarelli, Roupen Djinbachian, Jérémie Jacques, Timothée Wallenhorst, Clara Yzet, Louis-Jean Masgnaux, Florian Rostain, Alexandru Lupu, Jérôme Rivory, Mathieu Pioche
      Gastrointestinal Endoscopy.2026;[Epub]     CrossRef
    • Expert Endoscopist Agreement for Size Measurement of Large (> 2 cm) Colorectal Laterally Spreading Tumors: A Prospective Video-Based Study
      Roupen Djinbachian, Jérémie Jacques, Victoire Michal, Ludovico Alfarone, Robert Bechara, Nicholas G. Burgess, Mariana Figueiredo, Yusuke Fujiyoshi, Lucile Heroin, Michal F. Kaminski, Eric Lam, Philippe Leclercq, Isabelle Lienhart-Chambon, Alexandru Lupu,
      Digestive Diseases and Sciences.2026;[Epub]     CrossRef

    • PubReader PubReader
    • ePub LinkePub Link
    • Cite
      CITE
      export Copy Download
      Close
      Download Citation
      Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.
      Format:
      • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
      • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
      Include:
      • Citation for the content below
      Recent advancement in size measurement during endoscopy
      Clin Endosc. 2026;59(1):1-8.   Published online May 23, 2025
      DOI: https://doi.org/10.5946/ce.2025.070
      Close
    • XML DownloadXML Download
    Figure
    • 0
    Related articles
    Recent advancement in size measurement during endoscopy
    Image
    Fig. 1. Virtual scale endoscope. (A) Tip of virtual scale endoscope. (B, C) The software detects the position of the laser point in the endoscopic field of view and displays a virtual scale on the right side. To compare the lesion size using the virtual scale, the endoscopist positions the laser point on the left edge of the lesion. (D) Real measurement of a colorectal polyp. Provided by Fujifilm Corp. with permission.

    Fig. 1.

    Recent advancement in size measurement during endoscopy
    Study Year Method Results
    VSE
     Nakatani et al. 2007 Three-dimensional measurement endoscopic system with virtual rulers A novel endoscopic system equipped with four laser beam sources provided virtual rulers over endoscopic images, allowing physicians to measure lesions accurately.
     Yoshioka et al. 2021 VSE for real-time polyp measurement The virtual scale function of gastrointestinal endoscopy reduced measurement errors to ≤0.7 mm. Compared with biopsy forceps, VSE provided significantly more precise real-time polyp size estimations (p<0.001).
     Shimoda et al. 2022 Prospective study comparing VSE and visual estimation of polyp size VSE demonstrated an 84.0% accuracy in polyp size estimation compared with 62.5% for visual estimation. However, VSE required significantly longer measurement times (6.4 min vs. 2.9 min, p<0.001).
     Djinbachian et al. 2023 Randomized trial on comparison of VSE, biopsy forceps, and endoscopic ruler in polyp measurement VSE had a higher accuracy (82.7%) than biopsy forceps (78.9%) and endoscopic ruler (78.4%). VSE also reduced the rate of misclassification of polyps at clinically relevant size thresholds (5 mm, 10 mm, and 20 mm).
     Minakata et al. 2024 Prospective study on VSE accuracy for measuring early gastrointestinal lesions VSE measurement had significantly lower variability than visual estimation (p<0.001). VSE was particularly effective for accurately measuring lesions of various sizes and morphologies in real clinical settings.
    AI-assisted VR in polyps
     Abdelrahim et al. 2022 Computer vision techniques for polyp classification SfM had an 85.2% accuracy, far surpassing the 59.5% accuracy of endoscopists in polyp size classification as ≤5 mm or >5 mm. The CNN showed an accuracy of 80% for distinguishing polyps >5 mm from diminutive polyps in ten videos of human polyps.
     Kwak et al. 2022 AI-based polyp size measurement using W-Net model The AI-based method significantly improved accuracy in polyp size estimation compared with visual estimation and biopsy forceps. The concordance correlation coefficient was 0.961, indicating a high level of agreement with true polyp size measurements.
     Wang et al. 2024 Deep learning-based real-time system (ENDOANGEL-CPS) for polyp size estimation The ENDOANGEL-CPS system had a much higher accuracy in estimating polyp size than human endoscopists (89.9% vs. 54.7%, p<0.001). It significantly reduced inappropriate surveillance recommendations (1.5% vs. 16.6% for endoscopists, p<0.001).
    AI-assisted VR in esophageal varices
     Jin et al. 2023 AI-based VR for measuring esophageal varices diameter VR provided equivalent accuracy to esophageal varix manometer but with significantly reduced measurement time (31 s vs. 159 s, p<0.01). VR can reduce unnecessary interventions, decrease the risk of complications, and serve as a cost-effective tool for esophageal varix assessment.
     Fang et al. 2024 VR-based diameter measurement in esophageal variceal endoscopic therapy In a multicenter study of 345 patients with cirrhosis, VR-based esophageal variceal diameter measurements significantly reduced rebleeding rates after endoscopic therapy for varices >1 cm (3.8% vs. 11.3%, p=0.048).
     Mao et al. 2024 AI-based VR for measuring the diameter of esophageal varices A strong correlation was found between AI-measured esophageal varix diameter and the portal pressure gradient, with r=0.521 (p<0.001).
    Advantage Disadvantage
    Visual estimation Simple, quick to perform Subjective, experience-dependent
    No additional equipment needed Lacks standardization
    High interobserver variability
    Biopsy forceps/endoscopic rulers More accurate than visual estimation Time-consuming
    Can be used in real time Requires precise positioning, making it difficult for some anatomical locations
    Virtual scale endoscopy Real-time lesion size measurement Needs special equipment
    Reduces interobserver variability Requires training
    Highly accurate Prolonged procedural time
    Accuracy affected by scope angle and lesion orientation
    AI-assisted virtual rulers Automated, highly accurate Variability of performance
    Real-time processing Requires AI integration
    Highly accurate Regulatory approval and clinical validation are necessary
    Table 1. Summary of recent studies on emerging technologies in endoscopic size measurement

    VSE, virtual scale endoscopy; AI, artificial intelligence; VR, virtual ruler; SfM, structure from motion; CNN, convolutional neural network.

    Table 2. Comparison of endoscopic size measurement techniques

    AI, artificial intelligence.

    Table 1.

    Table 2.

    Clin Endosc : Clinical Endoscopy Twitter Facebook
    © Korean Society of Gastrointestinal Endoscopy.
    Close layer
    TOP