2022-RA-757-ESGO Bioimpedance screening for cervical cancer

Introduction/Background The pathological changes at different severity stages of cervical intraepithelial neoplasia (CIN) can give rise to the differences in electrical properties tissues, which lay foundation for screening CIN using non-invasive and cost-effective technique bioimpedance as well paves way easy availability screening, especially developing countries thereby reduction occurrence cancer. Methodology complex impedance measurements collected by a multi-electrode probe from clinically diagnosed 117 normal subjects, 18 subjects with low-grade squamous lesion (LSIL), 19 high-grade (HSIL) have been fitted Cole-Cole model, is used characterize biological tissues. geometrical features extracted model later exploited classify conditions tissues receiver operating characteristic (ROC) curve based on three algorithms: logistic regression, support vector machine, Naive Bayes. Results Regardless algorithms being used, area under (AUC) obtained apparently higher than that real part only, ROC areas between LSIL are much more improved when exploited, shown figure 2. Furthermore, due significant discrepancy HSIL cervix, greater AUC be observed separation irrespective used. Conclusion Compared utilization only measurements, our results present superior performance combination improving effectiveness classification during according analysis.