Cytopathology

Biography

Biography: K Sujathan

Abstract

Cervical cancer is the fourth most common cancer among women with an estimated 528,000 new cases every year. Among the different screening methods, cytology based screening using PAP smear remains as the best method for the pre-selection of women with cervical intraepithelial lesions. However, this method pose a challenge in practical implementation as it is resource intensive requiring trained professionals skilled enough to identify a handful of abnormal cells among few hundred thousand cells. This motivates the need for automating the screening methodology. Since the 1960-ies several groups have attempted such automated screening systems leading also to a couple of commercial products. Still these have had limited impact on the screening situation in most of the world. C-DAC(T) together with RCC-T and Uppsala University has developed a semi-automated system which analyses digitized PAP slides prepared using LBC techniques and employs quantitative analysis using image processing and machine learning algorithms to screen out clearly normal smears and direct abnormal smears for human review. A low cost alternative to commercial LBC technique was also identified. The system filters out normal smears without human intervention while referring the suspicious cases for expert’s review. It was evaluated on 1006 smears with a sensitivity of 96% and specificity of 72% for high grade lesions, which is comparable to that of human experts. Since in screening programs, a big majority of cases are within normal limits, the system is able to drastically reduce the workload of Cytologists, thus able to screen a larger population even in low resource settings.