Related Experiment Video
Updated: May 9, 2026
Quantifying Microorganisms at Low Concentrations Using Digital Holographic Microscopy DHM
Published on: November 1, 2017
Assessment of microorganism detection in ThinPrep Papanicolaou tests utilizing the Hologic Genius Digital Diagnostics
Ali Seemaab1, Amy Colaizzi1, Jonee Matsko1
1Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA, United States.
Objective:
The Hologic Genius Digital Diagnostics System (HGDDS) analyzes ThinPrep Papanicolaou (Pap) tests (TPPTs) to assist in detecting cervical lesions. The aim of this study was to determine the sensitivity of the HGDDS in identifying commonly diagnosed microorganisms in Pap tests.
Methods:
A total of 305 TPPT cases were selected from Magee Women's Hospital, University of Pittsburgh, consisting of 244 cases with microorganism diagnoses (a total of 262 cases of Actinomyces, Candida spp, herpes simplex virus [HSV], and Trichomonas) and 61 cases without microorganisms. Slides were scanned and then subjected to artificial intelligence (AI) analysis using the HGDDS and subsequently reviewed on a digital workstation by a cytologist, followed by a resident and a cytopathologist who made the final diagnoses.
Results:
Diagnosis using the HGDDS demonstrated high sensitivity across all microorganisms (95.4%). Herpes simplex virus detection was comparatively lower (82.5%). Of the microorganisms, 85.2% were displayed in the first gallery of 30 images within row 5, 7.2% presented in the first gallery outside of row 5, and 7.6% presented in the hidden gallery of images. Among the 12 cases with missed diagnoses, 3 of 5 Candida spp and 3 of 7 HSV organisms were not presented within the 60 images selected by HGDDS. In another 6 cases, microorganisms were found within the 60 fields, but none were present in row 5.
Conclusions:
Very high sensitivity was observed for TPPTs across 3 of 4 common microorganisms on the HGDDS, although sensitivity was relatively lower for detecting HSV. Understanding morphologic patterns of various microorganisms in detection misses by the HGDDS may help guide the implementation of AI-assisted cervical cancer screening systems.
