Diagnostic concordance of remote WSI-based thyroid cytology compared to conventional cytological diagnosis: a pilot study
Abstract Background Whole slide imaging (WSI) has emerged as a key component of digital pathology, enabling remote diagnostics and standardized evaluation. However, its diagnostic concordance in thyroid cytopathology, particularly under real-world conditions, remains insufficiently validated. Methods In this retrospective pilot study, thirteen thyroid cases were evaluated using both conventional light microscopy and WSI. One pathologist performed glass slide evaluation, while two independent pathologists assessed the corresponding digital slides remotely. All evaluations were conducted in a blinded manner. Diagnostic agreement was analyzed using percentage concordance and Cohen’s kappa coefficient. Results The overall concordance rate between conventional microscopy and WSI was 69.2%. Cohen’s kappa values ranged from 0.58 to 0.63, indicating moderate to substantial agreement. Full concordance was observed in 7 cases, predominantly among malignant diagnoses. Partial concordance occurred in 5 cases, mainly involving borderline categories. Complete discordance was identified in 1 case, exclusively within an indeterminate cytological category. No clinically significant misclassification between benign and malignant diagnoses was observed. Conclusion WSI demonstrated moderate diagnostic agreement with conventional microscopy in well-defined thyroid cytopathology categories. However, variability remains more pronounced in indeterminate cases, reflecting inherent diagnostic challenges. These findings support the preliminary feasibility of WSI in real-world and remote diagnostic workflows, although larger, multicenter studies are required to confirm these results.
Authors
- Özben Yalçın (ORCID: https://orcid.org/0000-0002-0019-1922)
- Şeyma Büyücek (ORCID: https://orcid.org/0000-0002-9106-6595)
- Hatice Elmas (ORCID: https://orcid.org/0000-0002-9796-9197)
- Abdullah Sahin (ORCID: https://orcid.org/0000-0003-0196-2319)
- Merve Dogan Ayan (ORCID: https://orcid.org/0000-0002-5673-9061)
Publication Details
- Journal
- Diagnostic Pathology
- Published
- 2026-09-22
- DOI
- https://doi.org/10.1186/s13000-026-01820-9
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00