A Novel Fully-Automated Deep Learning Pipeline for the Effective Detection of Cervical Cancer

In the process of computer-assisted cervical cancer detection, one of the most critical challenges is the detection of aberrant cervical cells. In this research, we devise an artificial intelligence (AI) framework solution that enables to inevitably screen for undetermined significance cells in order to help expedite the successive medical assessment of the subject areas. This solution is founded on the premise that AI can be thought of as a form of learning. The system can be broken down into the following significant stages: In the initial step, cervical cell segmentation is carried out, which really accountable for effectively recovering cell images from the Entire Slide Image (ESI); Next there is a cell classification component known as Compact VGG that is built on a compacted visual geometric group (VGG) network. This component is the most important part of the framework and is what is utilised to construct the model. For evaluating the efficacy of the given developed framework, cervical cancer image dataset is used, which was extracted from public repository. Also, performance evaluation metrices is determined, which proves the proposed model produces the reliable outcome based on the results produced.