AN ANALYSIS OF THE VARIATION OF HUMAN INTERPRETATION - PAPNET(TM) A MINI-CHALLENGE

The Papnet(TM) was given a mini-challenge of 200 cervical smears loade d to 50% with varying degrees of abnormality as interpreted by the ori ginating laboratory. The range of abnormality extended from 'atypia' t o invasive cancer and a few 'glandular' lesions were included as were a few smears which had been reported as 'inadequate'. Three cytologist s (two cytopathologists and one cytotechnologist) read and analysed th e 128 monitor pictures per slide, selected by the Scanning Algorithm a nd Neural Network systems. These results were compared with a 'gold st andard' report on the glass slide produced by two cytopathologists. Th e analysis was done for each individual cytologist, for cases in which all three agreed, for a consenus between two of the three and for the 'best of three'. The latter gave an error rate of 4% false negative ( Papnet scan negative) and 10% 'false positive' (referred for glass sli de examination). Individual cytologists had higher error rates demonst rating that errors could be due to human interaction and not necessari ly to the Scanner. This also indicated that wide experience in interpr etation of monitor images is needed to achieve high quality results.