Abstract:
Cervical cancer is one of the most common causes of cancer death. Early detection of cancer can reduce mortality and there are many screening methods nowadays e.g. Pap smear, liquid-based cytology, and visual inspection by colposcopy. Screening by measuring bioimpedance is also another method that can detect the histological change of cells. In this paper, instead of using only bioimpedance to classify tissues for cancer, we propose to reconstruct the data of bioimpedance obtained from a measurement pattern as the image of bioimpedance distribution, based on Electrical Impedance Tomography (EIT). A nine-electrode planar array probe was investigated. The simulation result shows that simulated grade1 and 2/3 cervical intraepithelial neoplasia (CIN) objects can be located for most of the cases with the localization error of 40-48% of the object's diameter. When the objects were situated in the low sensitivity region, the object could not be located. Grade 1 and small-sized CIN were more difficult to locate. It is also noticeable that the amplitude of conductivity change in the images was smaller than the true amplitude by 1.1-9.4 times. This indicates that these images are more suitable to use as qualitative images than quantitative images. © 2021 ACM.