Whole slide imaging for teleconsultation and clinical use
| Autor: | Bryan J. Dangott, Anil V. Parwani |
|---|---|
| Rok vydání: | 2010 |
| Předmět: |
medicine.medical_specialty
Pathology Third party business.industry Concordance Digital pathology Magnification Digital slide Health Informatics Gold standard (test) lcsh:Computer applications to medicine. Medical informatics Computer Science Applications Pathology and Forensic Medicine Glass slide lcsh:Pathology Commentary Medicine lcsh:R858-859.7 Medical physics Medical diagnosis business lcsh:RB1-214 |
| Zdroj: | Journal of Pathology Informatics Journal of Pathology Informatics, Vol 1, Iss 1, Pp 7-7 (2010) |
| ISSN: | 2153-3539 |
| Popis: | A thorough comparison of digital slide diagnosis with glass slide diagnosis is a critical hurdle for clinical use of whole slide scanners. The study by Wilbur et al. titled “Whole-slide imaging digital pathology as a platform for teleconsultation” examined the issue by specifically selecting difficult cases representing a broad spectrum of pathology from a variety of organ systems. The cases were characteristic of material that would often be referred for consultation. They then directly compared the Whole Slide Imaging Interpretation (WSII) with the Glass Slide Interpretation (GSI). A reference diagnosis was rendered on glass slides prior to sending the case for WSII and GSI consultations. Both WSII and GSI were performed at a third party lab by separate pathologists who were given limited clinical information. Thus, the consulting pathologists were forced to make a diagnosis on morphology alone without access to gross examinations, reference diagnosis, or referring physicians. A Zeiss Mirax Desk Scanner was used to create the WSI. However, the scanning magnification was not specified. The authors found an overall concordance of 91% between WSII and GSI, which is in line with other studies comparing these modalities.[1] They further analyzed the results by categorizing diagnoses into neoplastic and non-neoplastic groups. The neoplastic group performed slightly better with 93% concordance in comparison to the non-neoplastic group with 88% concordance, but the difference was not statistically significant. It is important to note that the authors also found three cases (5.7%) where the original glass slide reference diagnosis was discordant with the consultant GSI. Therefore, a built-in glass to glass comparison was performed in the study. It was determined that in these three cases the consultant GSI was most likely to be correct and the original reference diagnosis was incorrect. Assuming that the consultant GSI is the gold standard, WSII had a 90.6% concordance rate while the glass reference diagnosis had a 94.3% concordance rate. This correlates to a 3.7% difference between the consultant WSI concordance and the reference glass concordance to GSI. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|