Hoque, Md. Z., Keskinarkaus, A., Nyberg, P., Mattila, T., & Seppänen, T. (2022). Whole slide image registration via multi-stained feature matching. Computers in Biology and Medicine, 144, 105301. https://doi.org/10.1016/j.compbiomed.2022.105301
Whole slide image registration via multi-stained feature matching
|Author:||Hoque, Md. Ziaul1,2; Keskinarkaus, Anja1,2; Nyberg, Pia3,4;|
1Physiological Signal Analysis Group, Center for Machine Vision and Signal Analysis, University of Oulu, Finland
2Faculty of Information Technology and Electrical Engineering, University of Oulu, Finland
3Biobank Borealis of Northern Finland, Oulu University Hospital, Finland
4Translational & Cancer Research Unit, Medical Research Center Oulu, Faculty of Medicine, University of Oulu, Finland
5Department of Pathology, Oulu University Hospital, Finland
|Online Access:||PDF Full Text (PDF, 17.4 MB)|
|Persistent link:|| http://urn.fi/urn:nbn:fi-fe2022050432650
|Publish Date:|| 2022-08-02
In the recent decade, medical image registration and fusion process has emerged as an effective application to follow up diseases and decide the necessary therapies based on the conditions of patient. For many of the considerable diagnostic analyses, it is common practice to assess two or more different histological slides or images from one tissue sample. A specific area analysis of two image modalities requires an overlay of the images to distinguish positions in the sample that are organized at a similar coordinate in both images. In particular cases, there are two common challenges in digital pathology: first, dissimilar appearances of images resulting due to staining variances and artifacts; second, large image size. In this paper, we develop algorithm to overcome the fact that scanners from different manufacturers have variations in the images. We propose whole slide image registration algorithm where adaptive smoothing is employed to smooth the stained image. A modified scale-invariant feature transform is applied to extract common information and a joint distance helps to match keypoints correctly by eliminating position transformation error. Finally, the registered image is obtained by utilizing correct correspondences and the interpolation of color intensities. We validate our proposal using different images acquired from surgical resection samples of lung cancer (adenocarcinoma). Extensive feature matching with apparently increasing correct correspondences and registration performance on several images demonstrate the superiority of our method over state-of-the-art methods. Our method potentially improves the matching accuracy that might be beneficial for computer-aided diagnosis in biobank applications.
Computers in biology and medicine
|Type of Publication:||
A1 Journal article – refereed
|Field of Science:||
113 Computer and information sciences
213 Electronic, automation and communications engineering, electronics
The research work of this paper was conducted with the Physiological Signal Analysis group at Center for Machine Vision and Signal Analysis (CMVS) in the Faculty of Information Technology and Electrical Engineering (ITEE) at the University of Oulu. This research has been financially supported by the Academy of Finland 6 Genesis Flagship (Grant 318927) and Academy of Finland Identifying trajectories of healthy aging via integration of birth cohorts and biobank data (Grant 309112). We are thankful to MD, Ph.D., board-certified pathologist Johanna Mäkinen for selecting the lung HE-slides.
|Academy of Finland Grant Number:||
318927 (Academy of Finland Funding decision)
309112 (Academy of Finland Funding decision)
© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).