Back to Search
Start Over
Classification of thymic epithelial neoplasms is still a challenge to thoracic pathologists: a reproducibility study using digital microscopy.
- Source :
-
Archives of pathology & laboratory medicine [Arch Pathol Lab Med] 2014 May; Vol. 138 (5), pp. 658-63. - Publication Year :
- 2014
-
Abstract
- Context: Thymic epithelial tumors are rare, constituting interpretive challenges for pathologists. Digital imaging can be useful as an educational tool for these rare tumors.<br />Objectives: To evaluate the diagnostic reproducibility of thymic tumors among thoracic pathologists.<br />Design: Twenty cases of thymoma or thymic carcinoma were scanned into the Aperio system. The images were sent to pathologists with expertise in thoracic pathology at 6 different centers, who were asked to classify the tumors according to the 2004 World Heath Organization classification and to diagnose invasion. Interobserver agreement was evaluated. After discussion of the first 20 cases, a second set of 10 cases was evaluated.<br />Results: There was agreement for the diagnosis of thymoma and thymic carcinoma in 70% of cases (n = 14); in the remaining 6 cases, there was disagreement for cases of B3 thymoma (n = 5) and type A thymoma (n = 1) and thymic carcinoma. The overall κ was 0.39. When invasion was evaluated, the overall κ was 0.45. In the second round of the study, after discussion of diagnostic criteria, the interobserver agreement for the diagnosis of thymoma versus thymic carcinoma was 0.67 and that for the determination of invasion was 0.57-suggesting interpretative improvement.<br />Conclusion: The reproducibility of diagnosis of thymic epithelial tumors, using digital imaging, is comparable to that in previous studies using glass slides. Digital imaging is a good tool for remote consultation and for educational purposes. This technology could be used to train pathologists with low-level experience in thymic epithelial tumors and to foster collaborative work in the field.
- Subjects :
- Computers
Diagnosis, Computer-Assisted instrumentation
Diagnosis, Differential
Diagnostic Imaging instrumentation
Humans
Microscopy instrumentation
Neoplasms, Glandular and Epithelial diagnosis
Observer Variation
Pathology, Clinical standards
Reproducibility of Results
Thymoma diagnosis
Thymus Neoplasms diagnosis
Diagnosis, Computer-Assisted methods
Diagnostic Imaging methods
Microscopy methods
Neoplasms, Glandular and Epithelial classification
Thymoma classification
Thymus Neoplasms classification
Subjects
Details
- Language :
- English
- ISSN :
- 1543-2165
- Volume :
- 138
- Issue :
- 5
- Database :
- MEDLINE
- Journal :
- Archives of pathology & laboratory medicine
- Publication Type :
- Academic Journal
- Accession number :
- 24786123
- Full Text :
- https://doi.org/10.5858/arpa.2013-0028-OA