A New Cervical Cytology Dataset for Nucleus Detection and Image Classification (Cervix93) and Methods for Cervical Nucleus Detection

TL;DR

This paper introduces a new cervical cytology dataset with annotated nuclei and grades, along with baseline and deep learning methods that outperform existing approaches in nucleus detection and image classification.

Contribution

The paper provides a novel dataset for cervical cytology analysis and compares baseline and deep learning methods, demonstrating superior performance over existing techniques.

Findings

Deep learning method outperforms baseline and state-of-the-art methods

Dataset enables evaluation of nucleus detection and classification

Benchmarking code is publicly available

Abstract

Analyzing Pap cytology slides is an important tasks in detecting and grading precancerous and cancerous cervical cancer stages. Processing cytology images usually involve segmenting nuclei and overlapping cells. We introduce a cervical cytology dataset that can be used to evaluate nucleus detection, as well as image classification methods in the cytology image processing area. This dataset contains 93 real image stacks with their grade labels and manually annotated nuclei within images. We also present two methods: a baseline method based on a previously proposed approach, and a deep learning method, and compare their results with other state-of-the-art methods. Both the baseline method and the deep learning method outperform other state-of-the-art methods by significant margins. Along with the dataset, we publicly make the evaluation code and the baseline method available to download for further benchmarking.