Assessing the quality of cytopathology whole slide imaging for education from archived cases

J Am Soc Cytopathol. 2022 Sep-Oct;11(5):313-319. doi: 10.1016/j.jasc.2022.06.001. Epub 2022 Jun 8.

Abstract

Introduction: Many institutions have cytopathology case archives for education. Unfortunately, these slides deteriorate over time and have limited accessibility. Whole slide imaging (WSI) can overcome these limitations. However, suboptimal image quality and scanning effort are barriers.

Materials and methods: We selected 123 slides from cytopathology study sets for WSI scanning at 400x magnification without z-stacking. The Ventana DP 200 scanner and Virtuoso software were used. Slides were scanned in 2 rounds: the first round of slides was prepared for scanning with light cleaning, and the second round was performed only on slides that had unacceptable WSI quality after thorough cleaning. Slides were assessed with a 4-tier grading system created by the authors. Time to scan each slide was recorded.

Results: Within the first round, 96 of 123 (78%) slides scanned were determined to be of acceptable quality. After the second round of scanning, in total, 118 of 123 (95.9%) slides were determined to be of acceptable quality. The average time needed to scan each slide was 213 seconds.

Conclusions: The majority of slides scanned were of acceptable quality in the first round of scanning. After cleaning and rescanning, nearly every slide investigated was of acceptable quality. The primary objective is to provide other institutions that may be considering a similar project a benchmark so that they know what to expect in terms of slide scan success rate and the amount of time needed to digitize slides for educational archiving. This pilot study demonstrates the feasibility of using WSI for cytology education cases.

Keywords: Cytology education; Cytopathology education; Digital cytology; Digital cytopathology; Digital pathology; Whole slide imaging.

MeSH terms

  • Cytodiagnosis*
  • Humans
  • Microscopy*
  • Pilot Projects
  • Software