Hologic Genius Operator's Manual page 8

Digital Imager, including cables specified by the manufacturer. Otherwise, degradation of
the performance of this equipment could result.
•
Care should be taken to assure that slides are correctly oriented in the Digital Imager
slide carrier to prevent rejection by the system.
•
The Digital Imager should be placed on a flat, sturdy surface away from any vibrating
machinery to assure proper operation.
PERFORMANCE CHARACTERISTICS
OBJECTS OF INTEREST (OOI) STUDY
A laboratory study was conducted to demonstrate that the Genius Cervical AI algorithm
accurately selects OOIs. An OOI is a cell or cluster of cells on a slide preparation that most likely
contains clinically relevant information for diagnostic purposes. The study compared OOIs
selected by the GeniusCervical AI algorithm to the same samples imaged and reviewed by CTs
using the ThinPrep Imaging System (TIS-assisted review). The study evaluated the performance
of the Genius Cervical AI algorithm to present images suitable for diagnosing abnormal cervical
cases, for detecting the presence of common infectious organisms in a case, and for detecting
the presence of endocervical component (ECC) in a normal case. The study also measured
reproducibility of the Genius Digital Diagnostics System.
In the study, 260 ThinPrep slides were enrolled, made from individual residual ThinPrep Pap test
specimens, covering the full range of abnormal diagnostic categories as defined in The Bethesda
System for Reporting Cervical Cytology. The slides were imaged once on the ThinPrep Imaging
System, and the same slides were imaged three times on three different Genius Digital Imagers.
Slides were reviewed by CTs using the ThinPrep Imaging System (TIS-assisted review), and, after
a washout period, the same CT reviewed the nine runs of that same case on the Genius Digital
Diagnostics System. In each review on the Genius Digital Diagnostics System, the CT recorded
what the CT observed in every tile in the gallery for the case on the Review Station. The CT
reviews were conducted per standard laboratory procedure, recording the diagnostic result, the
presences or absence of endocervical component (ECC) and the presence of any infectious
organisms, such as trichomonas, candida, coccobacillus, for the TIS-assisted review.
The accuracy and reproducibility of the algorithm were measured by comparison to the TIS-
assisted diagnoses. The average and standard deviation across runs leading to the same
diagnosis or higher was the metric used.
OOI Study: Specimen Enrollment
Table 1 shows the nominal enrollment diagnoses (base on donor lab results) for the slides in the
study. In this study there was no independent truth standard, so the study did not measure
Genius™ Digital Diagnostics System Instructions for Use English AW-22963-001 Rev. 001 3-2021 4/13