Automated Detection and Classification of Cervical Cancer Using Pap Smear Microscopic Images: A Comprehensive Review and Future Perspectives
Abstract
Computer aided categorization of smear images are considered as an open problem for a last few decades. Cervical cancer is the main cause of mortality among women population worldwide and more prevalent in under developed countries. This can be successfully treated and cured if detected at its early phase. Computerized image analysis methods are primarily of great interest as it provides significant benefit for clinicians with reliable and timely analysis of the sample. Dedicated image analysis algorithms provide mathematical description of the region of interest which provide a great support to Pathologist for decision making. In this review, we have performed an outline of the state of art techniques expressed in prominent publications on computer assisted diagnostic system for cancer detection. By utilizing the domain aspects of cervical cancer, suitable methods and techniques are explored and presented. This review also presents a knowledge to assess the methodology used in the literature and emphasized some of the inadequacies and weaknesses in the reviewed methods. The study accentuated the future directions pertinent to the development of cost effective automated disease classification system that should be a significant benefit for the countries that has limited resources and treatment services.
