Neural network-based screening (NNS) of cervical smears can be performed as
a so-called "hybrid screening method, " in which parts of the cases are ad
ditionally studied by light,microscope, and " NNS, in which the cytological
it can also be used as "pure diagnosis is based only on the digital images
, generated by the NNS system. A random enriched sample of 985 cases, in a
previous study diagnosed by hybrid NNS, was drawn to be screened by pure NN
S, This study population comprised 192 women with (pre)neoplasia of the cer
vix and 793 negative cases. With pure NNS, more cases were recognized as se
verely abnormal; with hybrid NNS, more cases were cytologically diagnosed a
s low-grade. For a threshold value greater than or equal to HSIL. (high-gra
de squamous intraepithelial lesions), the areas under the receiver operatin
g characteristic (ROC) curves (AUC) were 81% (95% CI, 75-88%) for pure NNS
vs. 78% (95% CI, 75-81%) for hybrid NNS. For low-grade squamous intraepithe
lial lesions (LSIL), the A UC was significantly higher for hybrid NNS (81%;
95% CI, 77-85%) than for pure NNS (75%; 95% CI, 70-80%), Pure NNS provides
optimized prediction of HSIL cases or negative outcome, For the detection
of LSIL, light microscopy has additional value. (C) 2001 Wiley-Liss. Inc.