Global context-aware cervical cell detection with soft scale anchor matching. (June 2021)

• Record Type: Journal Article
• Authors: Liang, Yixiong; Pan, Changli; Sun, Wanxin; Liu, Qing; Du, Yun
• Is Part Of: Computer methods and programs in biomedicine. Volume 204(2021)
• Publication Date: 2021-06
• Subjects: Cervical cancer -- Object detection -- Convolutional neural network -- Global context -- Ground truth assignment; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2021.106061
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: An improved YOLOv3 is proposed for cervical cancer screening. Totally avoid the dependence on cell segmentation. Global context information is introduced into the network architecture. Ground truths are assigned to anchors more reasonable. Abstract: Background and Objective: Computer-aided cervical cancer screening based on an automated recognition of cervical cells has the potential to significantly reduce error rate and increase productivity compared to manual screening. Traditional methods often rely on the success of accurate cell segmentation and discriminative hand-crafted features extraction. Recently, detector based on convolutional neural network is applied to reduce the dependency on hand-crafted features and eliminate the necessary segmentation. However, these methods tend to yield too much false positive predictions. Methods: This paper proposes a global context-aware framework to deal with this problem, which integrates global context information by an image-level classification branch and a weighted loss. And the prediction of this branch is merged into cell detection for filtering false positive predictions. Furthermore, a new ground truth assignment strategy in the feature pyramid called soft scale anchor matching is proposed, which matches ground truths with anchors across scales softly. This strategy searches the most appropriate representation of ground truths in each layer and add more positive samples with different scales, which facilitate the feature learning. Results: Our proposed methods finally get 5.7 % increase in mean average precision and 18.5 % increase in specificity with sacrifice of 2.6 % delay in inference time. Conclusions: Our proposed methods which totally avoid the dependence on segmentation of cervical cells, show the great potential to reduce the workload for pathologists in automation-assisted cervical cancer screening.