Article Text

Download PDFPDF
PTU-029 Nice Classification for Optical Diagnosis of Colonic Lesions is Equally Applicable When Using Blue Laser Imaging
  1. R Rameshshanker1,2,
  2. A Wilson1,
  3. S Thomas-Gibson1,
  4. N Suzuki1,
  5. P Tekkis2,
  6. BP Saunders1
  1. 1Wolfson Endoscopy Unit, St Mark’s Hospital, Harrow
  2. 2Imperial College, London, UK

Abstract

Introduction Optical diagnosis of colonic polyps offers the potential for significant improvements in colonoscopy efficiency and cost saving. BLI is a new, push button, image enhancement endoscopy technology using two laser light sources to accentuate mucosal surface and vascular pattern of polyps. It’s visual appearance is similar to NBI. This study assessed whether the validated NICE classification system for colonic polyps (NBI) was applicable for BLI images of polyps.

Objectives

  • To assess whether NICE classification could be used to interpret BLI images

  • To assess the diagnostic performance of BLI imaging modalities in characterising colonic polyps

Methods High quality still images of 25 polyps were retrieved from an endoscopy image database. For each polyp we examined 4 different images: BLI, BLI with magnification (BLI MAG), BLI bright (BLI BR) and BLI BR with magnification (BLI BR MAG) (total of 100 images). Four experienced endoscopists were asked to make an optical diagnosis based on NICE (NBI International Colo rectal Endoscopic classification) for each BLI modality images.

Results A total of 100 polyp images were reviewed. Reviewers were able to identify a minimum of one NICE feature in all cases (100%) and all three NICE features (colour, vessel and surface pattern) in 90% of the images. There was no significant difference in the accuracy of predicting polyp histology between BLI BR and BLI BR MAG (89.6% vs 92%, p = 0.47) or BLI and BLI MAG (90.5% vs 92%, p = 0.8). However, the confidence levels were significantly increased when presented with magnified images (BLI BR vs BLI BR MAG, 55.4% vs 91.3%, p < 0.0001 and BLI vs BLI MAG 55.3% vs 91.3%).

Conclusion Our preliminary data highlights that the NICE classification could be used efficiently to interpret BLI images. The confidence levels in making an optical diagnosis were significantly increased with BLI magnification. Further studies are needed to validate these observations.

Disclosure of Interest None Declared

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.