Diagnosis of infectious disease pathology on whole slide images: a multi‐platform evaluation with implications for digital laboratory practice

PURPOSE: Adoption of whole slide imaging (WSI) is accelerating across anatomic pathology practices, yet there is a lack of evidence guiding its use for infectious disease diagnoses. WSI systems usually operate at a maximum of 40X magnification without z-plane depth, appropriately raising concerns about detecting microorganisms. METHODS: We assembled a 15-case panel (with immunohistochemistry [IHC]/special stains where indicated) spanning bacterial, viral, parasitic and fungal infections of various organ systems, including five organism-negative 'controls'. An infectious disease pathologist independently reviewed all glass slides to confirm the diagnoses. Eight board-certified pathologists reviewed cases (blinded to diagnoses) scanned on one of four high-throughput scanner platforms (Philips SG300, Hamamatsu S360, Pramana (Spectral HT2) and Leica GT450 designated A-D (not in order), two pathologists/platform). As a control group, two additional pathologists reviewed the same cases on glass slides. Additionally, in an attempt to compare image quality among the four WSI platforms, 10 pathologists (blinded to the platform) rated image quality from side-by-side screenshots in 12 representative cases. RESULTS: Aggregate diagnostic scores were: Scanner A: 78%, Scanner D: 71%, Scanner C: 69% and Scanner B: 65%, (mean 71%) vs. 79% for glass-slide reviewers. Intra-platform variability between the two pathologists on the same scanner (up to 23% on platform D) exceeded inter-platform differences. WSI reviewers performed near-equivalently to glass-slide reviewers for large organisms (92% vs. 89%) but lower for small/low-burden organisms (36% vs. 63%). CONCLUSIONS: Diagnostic performance on WSI for infectious disease pathology in this exploratory study varied primarily with organism size and burden, not by platform-specific image quality differences. WSI is non-inferior to glass for large organisms, but glass slide review may be superior for small (and low burden) organisms.

Authors

Institutions

Publication Details

Journal
Histopathology
Published
2026-11-17
DOI
https://doi.org/10.1111/his.70317
Primary Topic
AI in cancer detection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Diagnosis of infectious disease pathology on whole slide images: a multi‐platform evaluation with implications for digital laboratory practice

Sara Hayama, Quinn Carroll Rainer, Hemant Varma, Monika Vyas et al.
Histopathology
AI in cancer detection
article

Diagnosis of infectious disease pathology on whole slide images: a multi‐platform evaluation with implications for digital laboratory practice

Sara Hayama, Quinn Carroll Rainer, Hemant Varma, Monika Vyas, Yaileen D. Guzmán‐Arocho, Vikram Deshpande, Yuho Ono, Rahul Jawale, Nandan Padmanabha, Pavel Kopach, Emily Anne Towery, Joshua F. Goldsmith, Emily Elizabeth Ryan, Paul A. VanderLaan, Kanchan Kantekure, Ramy Arnaout, Yue Sun, Osman H. Yilmaz, Christos Tsokos
article en

Abstract

PURPOSE: Adoption of whole slide imaging (WSI) is accelerating across anatomic pathology practices, yet there is a lack of evidence guiding its use for infectious disease diagnoses. WSI systems usually operate at a maximum of 40X magnification without z-plane depth, appropriately raising concerns about detecting microorganisms. METHODS: We assembled a 15-case panel (with immunohistochemistry [IHC]/special stains where indicated) spanning bacterial, viral, parasitic and fungal infections of various organ systems, including five organism-negative 'controls'. An infectious disease pathologist independently reviewed all glass slides to confirm the diagnoses. Eight board-certified pathologists reviewed cases (blinded to diagnoses) scanned on one of four high-throughput scanner platforms (Philips SG300, Hamamatsu S360, Pramana (Spectral HT2) and Leica GT450 designated A-D (not in order), two pathologists/platform). As a control group, two additional pathologists reviewed the same cases on glass slides. Additionally, in an attempt to compare image quality among the four WSI platforms, 10 pathologists (blinded to the platform) rated image quality from side-by-side screenshots in 12 representative cases. RESULTS: Aggregate diagnostic scores were: Scanner A: 78%, Scanner D: 71%, Scanner C: 69% and Scanner B: 65%, (mean 71%) vs. 79% for glass-slide reviewers. Intra-platform variability between the two pathologists on the same scanner (up to 23% on platform D) exceeded inter-platform differences. WSI reviewers performed near-equivalently to glass-slide reviewers for large organisms (92% vs. 89%) but lower for small/low-burden organisms (36% vs. 63%). CONCLUSIONS: Diagnostic performance on WSI for infectious disease pathology in this exploratory study varied primarily with organism size and burden, not by platform-specific image quality differences. WSI is non-inferior to glass for large organisms, but glass slide review may be superior for small (and low burden) organisms.

Histopathology
Beth Israel Deaconess Medical Center (US), Harvard University (US), University of Massachusetts Chan Medical School (US), University of California, San Francisco (US), Baystate Health (US), Portsmouth Regional Hospital (US)
AI in cancer detection
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon