Evaluation of Three Algorithms for the Segmentation of Overlapping Cervical Cells

In this paper, we introduce and evaluate the systems submitted to the first Overlapping Cervical Cytology Image Segmentation Challenge, held in conjunction with the IEEE International Symposium on Biomedical Imaging 2014. This challenge was organized to encourage the development and benchmarking of techniques capable of segmenting individual cells from overlapping cellular clumps in cervical cytology images, which is a prerequisite for the development of the next generation of computer-aided diagnosis systems for cervical cancer. In particular, these automated systems must detect and accurately segment both the nucleus and cytoplasm of each cell, even when they are clumped together and, hence, partially occluded. However, this is an unsolved problem due to the poor contrast of cytoplasm boundaries, the large variation in size and shape of cells, and the presence of debris and the large degree of cellular overlap. The challenge initially utilized a database of 16 high-resolution (×40 magnification) images of complex cellular fields of view, in which the isolated real cells were used to construct a database of 945 cervical cytology images synthesized with a varying number of cells and degree of overlap, in order to provide full access of the segmentation ground truth. These synthetic images were used to provide a reliable and comprehensive framework for quantitative evaluation on this segmentation problem. Results from the submitted methods demonstrate that all the methods are effective in the segmentation of clumps containing at most three cells, with overlap coefficients up to 0.3. This highlights the intrinsic difficulty of this challenge and provides motivation for significant future improvement.

I. Introduction

Cervical cancer is a common occurring condition primarily caused by the infection of some types of human papillomavirus. According to a report by WHO published in 2012 [1], cervical cancer is the second most common gynecological cancer in less developed regions. In Australia, current reports estimate 885 new cases of cervical cancer to be diagnosed in 2015, and 250 deaths due to this disease [2]. Currently, Pap smear test [3] is an important routine screening in the early detection of this type of cancer. In this screening process, a clinician collects a sample of cells from the uterine cervix, which are then stained using the Papanicolaou technique to enable visual inspection on a microscope, where the appearance of each cell provides features that indicate the stages of cervical cancer. The development of automated cell deposition techniques, such as monolayer preparations, has facilitated both manual and automated slide analysis techniques by removing a large portion of blood, mucus, and other debris, reducing cellular overlap and producing specimens that are more likely to occur in a single focal plane. However, the manual analysis of cell abnormalities on Pap-stained specimens is a time-consuming and error-prone procedure, where sensitivity is affected by the number of cells inspected, the overlap among these cells, the poor contrast of the cell cytoplasm, and the presence of mucus, blood, and inflammatory cells [4] (see examples in Fig. 1).