Validation Study of Automated Whole Slide Imaging for Pathology Diagnosis
Author Information
Author(s): Gilbertson John R, Ho Jonhan, Anthony Leslie, Jukic Drazen M, Yagi Yukako, Parwani Anil V
Primary Institution: University of Pittsburgh Medical Center
Hypothesis
Can pathologists effectively diagnose complex cases using automated whole slide imaging?
Conclusion
The study found that current whole slide images provide sufficient information for reliable diagnostic decisions by pathologists.
Supporting Evidence
- In 17 of 25 cases, there were no discrepancies between the individual study pathologist reports.
- The WSI consensus diagnoses showed no significant discrepancies compared to original sign-out diagnoses.
- Pathologists rated the majority of whole slide images as either 'excellent' or 'diagnostic'.
- All 25 cases showed agreement between the WSI consensus diagnosis and the original sign-out report.
Takeaway
This study shows that doctors can use digital images of slides to make accurate diagnoses, even for complicated cases.
Methodology
Three pathologists independently diagnosed 25 cases using whole slide images, and their diagnoses were compared to original microscope-based reports.
Potential Biases
Potential bias due to the limited number of pathologists and their varying levels of experience with whole slide imaging.
Limitations
The study involved a small number of pathologists and cases, which may limit the generalizability of the findings.
Participant Demographics
Three pathologists with varying experience in pathology, including specific training in genitourinary and dermatopathology.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website