Please use this identifier to cite or link to this item: http://hdl.handle.net/2381/42976
Title: A novel method for unsupervised scanner-invariance with DCAE model
Authors: Moyes, Andrew
Zhang, Kun
Wang, Liping
Ji, Ming
Crookes, Danny
Zhou, Huiyu
First Published: 6-Sep-2018
Presented at: 29th British Machine Vision Conference (BMVC) Newcastle
Start Date: 3-Sep-2015
End Date: 6-Sep-2015
Citation: 29th British Machine Vision Conference (BMVC), 2018
Abstract: Automated analysis of histopathology whole-slide images is impeded by the scannerdependent variance introduced in the slide scanning process. This work presents a novel dual-channel auto-encoder based model with a multi-component loss which learns a scanner-invariant representation of histopathology images. The learned representation can be used for a number of histopathology-related applications where images are captured from different scanners such as nuclei detection and cancer segmentation. The approach is validated on a set of lung tissue sub-images extracted from whole slide images. This method achieves a 50% improvement in SSIM score on tissue masks derived from the learned representation compared to related methods. To the best of the author’s knowledge, this is the first work which explicitly learns a scanner-invariant representation of histopathology images from multiple domains simultaneously without labelled data or expensive preprocessing techniques.
Links: http://bmvc2018.org/contents/papers/0936.pdf
http://hdl.handle.net/2381/42976
Version: Post-print
Status: Peer-reviewed
Type: Conference Paper
Rights: Copyright 2018. The copyright of this document resides with its authors. It may be distributed unchanged freely in print or electronic forms.
Appears in Collections:Conference Papers & Presentations, Dept. of Computer Science

Files in This Item:
File Description SizeFormat 
a_novel_method_for_unsupervised_scanner_invariance_using_a_dual_channel_auto_encoder_encoder_model_camera_ready.pdfPost-review (final submitted author manuscript)2.24 MBAdobe PDFView/Open


Items in LRA are protected by copyright, with all rights reserved, unless otherwise indicated.