Automated and Interactive Lesion Detection and Segmentation in Uterine Cervix Images

Abstract

Abstract-This paper presents a procedure for automatic extraction and segmentation of a class-specific object (or region) by learning class-specific boundaries. We describe and evaluate the method with a specific focus on the detection of lesion regions in uterine cervix images. The watershed segmentation map of the input image is modeled using an MRF in which watershed regions correspond to binary random variables indicating whether the region is part of the lesion tissue or not. The local pairwise factors on the arcs of the watershed map indicate whether the arc is part of the object boundary. The factors are based on supervised learning of a visual word distribution. The final lesion region segmentation is obtained using a loopy belief propagation applied to the watershed arc-level MRF. Experimental results on real data show state-of-the-art segmentation results on this very challenging task that if necessary, can be interactively enhanced.

A system for detection of cervical precancerous in field emission scanning electron microscope images using texture features

“…The comparison performances of the cervical screening system and the several aforementioned approaches are tabulated in Table 3 in term of accuracy, sensitivity, and speci¯city. Four systems, namely system A, 25 B, 8 C, 26 and the proposed system are included for comparison.…”

“…Several other published computeraided screening systems using image processing techniques are based on thin prep, 6 colposcopy, 7 cervicography, 8 and°uorescent in situ hybridization (FISH). 9 The computer-aided screening system generally consists of three subsystems namely preprocessing system, features extraction and classi¯-cation systems.…”

A Double Layered Segmentation Algorithm for Cervical Cell Images based on GHFCM and ABC

“…In [11], an automatic nuclei segmentation technique for cervical images is proposed. The cervical image is represented as by a Markov Random Field (MRF), on the basis of the watershed segmentation map.…”

Scope of Improvement in Quality of Diagnostic Images for Pelvic region

“…As a result sufficient image enhancement cannot be achieved. Alush et al [4] introduced Water shed segmentation methodology which suffered from over segmentation but was able to characterize the regional intensity In [5], neural network has been deployed to understand the degrees of lesions. The neural network supports image enhancement, unless the data interpolation deviates from the actual data.…”