An effective diagnosis of cervical cancer neoplasia by extracting the diagnostic features using CRF

Cervical cancer is one of the most common forms of cancer in the woman worldwide. Most cases of cervical cancer can be prevented if it is detected earlier through various screening programs. This paper provides various methods for the automated diagnosis of cervical cancer neoplasia. The techniques that are investigated to create a fully automated system to locate precancerous and cancerous regions in an image of a cervix generated by the digital colposcope is considered. The image regions corresponding to different tissue types are identified for the extraction of domain-specific anatomical features. Domain-specific diagnostic features are used in a probabilistic manner using Conditional Random Fields (CRF). The abnormal areas in colposcopic images are located exactly. Thus this automated diagnosis of cervical cancer method is more useful for the developing countries since they have low-resource settings and poor financial condition.