Original Research ArticleA novel automation-assisted cervical cancer reading method based on convolutional neural network
Introduction
Cervical cancer is one of the most common causes of cancer death for women worldwide and is most frequently in developing countries [1]. Papanicolaou test (abbreviated as Pap test) or cervical cytology is now a mainstay cervical cancer screening method to detect potentially pre-cancerous and cancerous process in the cervix, which has demonstrated reduction in cervical cancer incidence and mortality in developed countries [2], [3]. Such method is performed by a visual examination of cytopathological analysis under the microscope of the glass slide and finally giving a diagnosis report according to the descriptive diagnosis method of the Bethesda system (TBS) [4]. However, manual analysis of microscope images is time-consuming, labor-intensive and error-prone as a handful of abnormal cells among millions of cells within a single slide has to be identified by a trained professional [5].
Therefore, automation-assisted screening based on cervical cytology has become a necessity. Since the first system was developed in 1950s, extensive research has attempted to exploit automation-assisted reading systems based on automatic image analysis techniques (e.g., [6], [7], [8]) which led to a couple of commercial systems emerged, such as the BD FocalPoint Slide Profiler [9] and ThinPrep [10] which received approval from the US Food and Drug Administration (FDA). While automation-assisted reading systems can increase productivity by reducing the time needed to read slides, their current performance and costs are not recommended for application in primary cervical screening [11], [12]. To this end, lots of automation-assisted methods based on cervical cell image analysis have been proposed [5], [13], [14], [15], [16]. Most of them follow the multi-stage pipeline, i.e., first identifying the candidate regions based on segmentation, then extracting hand-crafted features based on the characteristics of nuclei and cytoplasm for classification, as shown in Fig. 1(a).
While most of these studies have achieved available performance whether in cell segmentation or cell classification, there still have some challenges to use them in clinical automation-assisted reading. First, the current automation-assisted reading approaches has not been sufficiently cost-effective to promote to the cervical cell screening in clinical due to the tedious image patches preprocessing and screening process for cyto-technicians and doctors. The majority of existing research have done on the Herlve dataset [17] which only contains single-cell images with a size of 200 × 100 pixels approximately and was produced carefully by trained professionals. As shown in Fig. 2, images from Herlve dataset are all clear with no overlapping and impurity. In fact, the slide image with around 2160 million pixels obtained by the Whole Slide Imaging (WSI) technology [18] has complex cellular situation, such as cell overlapping, noise and impurity. Thus, one original cervical cell slide should be cropped into a huge number of single-cell image patches by using sliding windows or region proposal generation methods based on low-level image features, which lead to low efficiency. Second, it is difficult to make segmentation of the cytoplasm and nuclei absolutely due to the high degree of cell overlapping, the poor contrast of the cell cytoplasm and the presence of mucus, noise and impurity. Third, it is worth considering that whether the hand-crafted features can represent complex identification information or not since the richer semantic information sensitive to recognition may actually exist in hidden upper-level features of cervical cell images [19]. In addition, medical images are too complex and variable to get a perfect annotation as ground truth, which leads to noisy label unavoidably. However, the previous studies usually deal with cervical cell segmentation and classification without considering the existence of noisy labels.
To cope with these problems, we propose to utilize CNN-based object detection to automatically extract and learn task-specific features, and achieve the cervical cells recognition efficiently on multi-cell images with cell overlapping and clusters, as showing in Fig. 1(b). Our method directly operates on multi-cell image with size of 4000 × 3000 automatically extract more complex discriminative features and can obtain an image-level classification results. Moreover, it not only achieves cervical cell image-level classification but also provides more detailed location and category reference information of abnormal cells. In detail, we exploit YOLOv3 [20] as our cervical cell object detection baseline model due to the efficiency, accuracy and flexibility. In order to improve the classification performance of hard examples which are four highly similar categories, we cascade a further task-specifical classifier. Furthermore, we weaken the influence of noisy labels by smoothing their distribution.
Our contributions are summarized as follows: (1) Unlike the previous method, we treat the cervical cell recognition as object detection which automatically detect cervical cells directly on multi-cell images. It is more efficient as we can extract features automatically without manual intervention and careful design for all stages. Our method not only achieves cervical cell image-level classification but also provides more detailed location and category information of abnormal cells simultaneously. (2) We propose a simple and effective scheme, cascading a further task-specific classifier to improve the performance of hard example recognition. (3) We investigate the existence of noisy labels on cervical cell dataset and propose an approach to weaken the influence of them by smoothing their distribution.
Section snippets
Cervical cell recognition
From the analysis of the existing work, extensive research [21], [22], [23], [24], [25] has been devoted to the field of automatic recognition of cervical cancer and have obtained good results. The previous cervical cell recognition can be classified into two types according to the number of cells in the image: recognition based on single-cell image and multi-cell image.
Early methods proposed to achieve the automatic segmentation and classification of abnormal cervical cells based on isolated
Methodology
The pipeline of proposed method includes cervical cell detection and hard example classifier, as shown in Fig. 3. In this section, we firstly describe the pipeline of proposed end-to-end CNN-based object detection. And then, we introduce several improvements such as hard example classification, smoothing noisy label regularization, to make it more appropriate for our cervical cell recognition.
Image dataset
As there is no standard clinical cervical cells dataset with multi cells available publicly, we establish our own dataset captured by digital camera Ximea MC124CG-SY-UB with 12 million pixels situated on the microscope Olympus BX40 with 20× objective. Each pixel has a size of 3.45 μm2. For one cervical cell slide, we can capture about 1800 images. The specimens were prepared by liquid-based cytology with Feulgen staining. The dataset used in this paper is consisted of 12,909 cervical images with
Conclusions
In this paper, we utilize object detection method to achieve the automation-assisted cervical cell reading system. Different from the multi-stage traditional approaches, which rely on the accuracy of segmentation and the efficiency of hand-crafted features, our method extract high-level features automatically and detect cervical cells directly. We exploit YOLOv3 as a base model to detect 10 categories and then cascading a further hard example classifier to refine the 4 categories: ASC-US,
Authors’ contribution
Yao Xiang: conceptualization, methodology. Wanxin Sun: data curation, writing – original draft preparation. Changli Pan: software, validation. Meng Yan: methodology. Zhihua Yin: visualization, software. Yixiong Liang: conceptualization, writing – reviewing and editing.
Funding statement
This work was partially supported by the National Natural Science Foundation of China under Grant No. 61602522, and the Fundamental Research Funds of the Central Universities of Central South University [No.2018zzts595].
References (68)
- et al.
A review of image analysis and machine learning techniques for automated cervical cancer screening from pap-smear images
Comput Methods Programs Biomed
(2018) - et al.
Automated classification of pap smear images to detect cervical dysplasia
Comput Methods Programs Biomed
(2017) Automated screening of cervical cytology specimens
Human Pathol
(1996)- et al.
Automation-assisted versus manual reading of cervical cytology (mavaric): a randomised controlled trial
Lancet Oncol
(2011) - et al.
Unsupervised segmentation and classification of cervical cell images
Pattern Recogn
(2012) - et al.
Hybrid ensemble learning technique for screening of cervical cancer using papanicolaou smear image analysis
Pers. Med. Universe
(2015) - et al.
Automatic cervical cell segmentation and classification in pap smears
Comput Methods Programs Biomed
(2014) - et al.
A multiscale algorithm for nuclei extraction in pap smear images
Expert Syst Appl
(2016) - et al.
Nucleus and cytoplast contour detector of cervical smear image
Pattern Recogn Lett
(2008) - et al.
Segmentation of cervical cell nuclei in high-resolution microscopic images: a new algorithm and a web-based software framework
Comput Methods Programs Biomed
(2012)
Cytoplasm and nucleus segmentation in cervical smear images using radiating Gvf snake
Pattern Recogn
Evaluation of three algorithms for the segmentation of overlapping cervical cells
IEEE J Biomed Health Informatics
Cervical cancer classification from pap-smears using an enhanced fuzzy c-means algorithm
Informatics Med Unlocked
Segmentation of cytoplasm and nuclei of abnormal cells in cervical cytology using global and local graph cuts
Comput Med Imaging Graph
Automatic cell nuclei segmentation and classification of cervical pap smear images
Biomed Signal Process Control
Object detection based on deep learning for urine sediment examination
Biocybern Biomed Eng
Multi-modal framework for automatic detection of diagnostically important regions in nonalcoholic fatty liver ultrasonic images
Biocybern Biomed Eng
An overview of prevention and early detection of cervical cancers
Indian J Med Paediatr Oncol: Off J Indian Soc Med Paediatr Oncol
The Bethesda system for reporting cervical cytology: definitions, criteria, and explanatory notes
A pap-smear analysis tool (pat) for detection of cervical cancer from pap-smear images
Biomed Eng Online
Evaluation of the papnet cytologic screening system for quality control of cervical smears
Am J Clin Pathol
The autopap system for primary screening in cervical cytology. comparing the results of a prospective, intended-use study with routine manual practice
Acta Cytol
The Becton Dickinson focalpoint GS imaging system: clinical trials demonstrate significantly improved sensitivity for the detection of important cervical lesions
Am J Clin Pathol
Assisted primary screening using the automated thinprep imaging system
Am J Clin Pathol
Screening for cervical cancer using automated analysis of pap-smears
Comput Math Methods Med
Automation-assisted cervical cancer screening in manual liquid-based cytology with hematoxylin and eosin staining
Cytometry Part A
Various techniques for classification and segmentation of cervical cell images – a review
Int J Comput Appl
Pap-smear benchmark data for pattern classification
Nature inspired Smart Information Systems (NiSIS 2005)
Whole slide imaging
Am J Dermatopathol
Deeppap: deep convolutional networks for cervical cell classification
IEEE J Biomed Health Inform
Yolov3: an incremental improvement
Cervical cancer diagnosis based on random forest
Int J Perform Eng
Automation of detection of cervical cancer using convolutional neural networks
Crit Rev Biomed Eng
Development of algorithms for automated detection of cervical pre-cancers with a low-cost, point-of-care, pocket colposcope
IEEE Trans Biomed Eng
Cited by (56)
Deep learning for computational cytology: A survey
2023, Medical Image AnalysisCitation Excerpt :For example, Amorim et al. (2020) stained cervical cells by AgNOR to show cell proliferation, differentiation, and malignant transformation. Xiang et al. (2020) stained cervical specimens of LBC by Feulgen, and Marzahl et al. (2020) stained pulmonary hemosiderophages cytology using Perlss’ Prussian Blue and Turnbull’s Blue. Stain Normalization.
Multi-class nucleus detection and classification using deep convolutional neural network with enhanced high dimensional dissimilarity translation model on cervical cells
2022, Biocybernetics and Biomedical EngineeringCervical Cancer Classification From Pap Smear Images Using Deep Convolutional Neural Network Models
2024, Interdisciplinary Sciences – Computational Life SciencesClassification of single-cell cervical pap smear images using EfficientNet
2023, Expert Systems