Automatic cervical cell segmentation and classification in Pap smears
Computer Methods and Programs in Biomedicine
| Hi-index | 0.00 |
This work presents a digital imaging system able to assist physicians to track cervical cancer. The goal is to automatically extract the region where the cervical cancer starts to occur. This region called Acetowite Epithelium Zone (AEZ) is highlighted by physicians using a diluate acetic acid on the surface of the cervix. Recent digital images processing works suggested to extract this region with methods based on active contour models. These methods are performant, they require manually selecting an area in the AEZ region, but have the disadvantage to be very sensitive to the size and localization of the selected area. To remove these inconveniences, we propose to use the Active Contours Without Edges method. This method is based on the intensity of the overall pixels of the selected region, therefore, it can extract the region contour even if additive noises are present and/or the contour is not well defined. Experimental results show the robustness of this method to delimit Acetowite Epithelium Zone.