University of Oulu

Akram S.U., Qaiser T., Graham S., Kannala J., Heikkilä J., Rajpoot N. (2018) Leveraging Unlabeled Whole-Slide-Images for Mitosis Detection. In: Stoyanov D. et al. (eds) Computational Pathology and Ophthalmic Medical Image Analysis. OMIA 2018, COMPAY 2018. Lecture Notes in Computer Science, vol 11039. Springer, Cham

Leveraging unlabeled whole-slide-images for mitosis detection

Saved in:
Author: Akram, Saad Ullah1,2; Qaiser, Talha2; Graham, Simon2;
Organizations: 1Center for Machine Vision and Signal Analysis (CMVS), University of Oulu, Oulu, Finland
2Tissue Image Analytics (TIA) Lab, University of Warwick, Coventry, UK
3Department of Computer Science, Aalto University, Espoo, Finland
4The Alan Turing Institute, London, UK
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 0.9 MB)
Persistent link:
Language: English
Published: Springer Nature, 2018
Publish Date: 2019-06-06


Mitosis count is an important biomarker for prognosis of various cancers. At present, pathologists typically perform manual counting on a few selected regions of interest in breast whole-slide-images (WSIs) of patient biopsies. This task is very time-consuming, tedious and subjective. Automated mitosis detection methods have made great advances in recent years. However, these methods require exhaustive labeling of a large number of selected regions of interest. This task is very expensive because expert pathologists are needed for reliable and accurate annotations. In this paper, we present a semi-supervised mitosis detection method which is designed to leverage a large number of unlabeled breast cancer WSIs. As a result, our method capitalizes on the growing number of digitized histology images, without relying on exhaustive annotations, subsequently improving mitosis detection. Our method first learns a mitosis detector from labeled data, uses this detector to mine additional mitosis samples from unlabeled WSIs, and then trains the final model using this larger and diverse set of mitosis samples. The use of unlabeled data improves F1-score by ∼5% compared to our best performing fully-supervised model on the TUPAC validation set. Our submission (single model) to TUPAC challenge ranks highly on the leaderboard with an F1-score of 0.64.

see all

Series: Lecture notes in computer science
ISSN: 0302-9743
ISSN-E: 1611-3349
ISSN-L: 0302-9743
ISBN: 978-3-030-00949-6
ISBN Print: 978-3-030-00948-9
Pages: 69 - 77
DOI: 10.1007/978-3-030-00949-6_9
Host publication: Computational Pathology and Ophthalmic Medical Image Analysis. First International Workshop, COMPAY 2018, and 5th International Workshop, OMIA 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16 - 20, 2018
Host publication editor: Stoyanov, Danail
Taylor, Zeike
Ciompi, Francesco
Xu, Yanwu
Conference: OMIA: International Workshop on Ophthalmic Medical Image Analysis. COMPAY: International Workshop on Computational Pathology
Type of Publication: A4 Article in conference proceedings
Field of Science: 213 Electronic, automation and communications engineering, electronics
217 Medical engineering
Copyright information: © Springer Nature Switzerland AG 2018. This is a post-peer-review, pre-copyedit version of an article published in OMIA 2018, COMPAY 2018: Computational Pathology and Ophthalmic Medical Image Analysis. The final authenticated version is available online at: