Contrastive Transformer-Based Multiple Instance Learning for Weakly Supervised Polyp Frame Detection
Date
2022
Authors
Tian, Y.
Pang, G.
Liu, F.
Liu, Y.
Wang, C.
Chen, Y.
Verjans, J.
Carneiro, G.
Editors
Wang, L.
Dou, Q.
Fletcher, P.T.
Speidel, S.
Li, S.
Dou, Q.
Fletcher, P.T.
Speidel, S.
Li, S.
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Conference paper
Citation
Lecture Notes in Artificial Intelligence, 2022 / Wang, L., Dou, Q., Fletcher, P.T., Speidel, S., Li, S. (ed./s), vol.13433, pp.88-98
Statement of Responsibility
Yu Tian, Guansong Pang, Fengbei Liu, Yuyuan Liu, Chong Wang, Yuanhong Chen, Johan Verjans, Gustavo Carneiro
Conference Name
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) (18 Sep 2022 - 22 Sep 2022 : Singapore)
Abstract
Current polyp detection methods from colonoscopy videos use exclusively normal (i.e., healthy) training images, which i) ignore the importance of temporal information in consecutive video frames, and ii) lack knowledge about the polyps. Consequently, they often have high detection errors, especially on challenging polyp cases (e.g., small, flat, or partially visible polyps). In this work, we formulate polyp detection as a weakly-supervised anomaly detection task that uses video-level labelled training data to detect frame-level polyps. In particular, we propose a novel convolutional transformer-based multiple instance learning method designed to identify abnormal frames (i.e., frames with polyps) from anomalous videos (i.e., videos containing at least one frame with polyp). In our method, local and global temporal dependencies are seamlessly captured while we simultaneously optimise video and snippet-level anomaly scores. A contrastive snippet mining method is also proposed to enable an effective modelling of the challenging polyp cases. The resulting method achieves a detection accuracy that is substantially better than current state-of-the-art approaches on a new large-scale colonoscopy video dataset introduced in this work. Our code and dataset are available at https://github.com/tianyu0207/weakly-polyp.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG