View in EDS

Mitosis detection, fast and slow: Robust and efficient detection of mitotic figures.

Saved in:
Bibliographic Details
Title: Mitosis detection, fast and slow: Robust and efficient detection of mitotic figures.
Authors: Jahanifar, Mostafa1 (AUTHOR) mostafa.jahanifar@warwick.ac.uk, Shephard, Adam1 (AUTHOR), Zamanitajeddin, Neda1 (AUTHOR), Graham, Simon1,2 (AUTHOR), Raza, Shan E. Ahmed1 (AUTHOR), Minhas, Fayyaz1 (AUTHOR), Rajpoot, Nasir1,2 (AUTHOR) n.m.rajpoot@warwick.ac.uk
Source: Medical Image Analysis. May2024, Vol. 94, pN.PAG-N.PAG. 1p.
Subjects: Mitosis, Deep learning, Architectural design, Breast cancer
Abstract: Counting of mitotic figures is a fundamental step in grading and prognostication of several cancers. However, manual mitosis counting is tedious and time-consuming. In addition, variation in the appearance of mitotic figures causes a high degree of discordance among pathologists. With advances in deep learning models, several automatic mitosis detection algorithms have been proposed but they are sensitive to domain shift often seen in histology images. We propose a robust and efficient two-stage mitosis detection framework, which comprises mitosis candidate segmentation (Detecting Fast) and candidate refinement (Detecting Slow) stages. The proposed candidate segmentation model, termed EUNet , is fast and accurate due to its architectural design. EUNet can precisely segment candidates at a lower resolution to considerably speed up candidate detection. Candidates are then refined using a deeper classifier network, EfficientNet-B7, in the second stage. We make sure both stages are robust against domain shift by incorporating domain generalization methods. We demonstrate state-of-the-art performance and generalizability of the proposed model on the three largest publicly available mitosis datasets, winning the two mitosis domain generalization challenge contests (MIDOG21 and MIDOG22). Finally, we showcase the utility of the proposed algorithm by processing the TCGA breast cancer cohort (1,124 whole-slide images) to generate and release a repository of more than 620K potential mitotic figures (not exhaustively validated). • Novel two-stage method efficiently detects mitosis via deep learning models. • Self-supervised method pre-trains encoder and decoder of a segmentation model. • Exploring impacts of various domain generalization methods on mitosis segmentation. • AI results for TCGA-BRCA with 620K+ mitotic figures released, available online. • Outperforming other algorithms in mitosis detection and ranked 1st in MIDOG21/22. [ABSTRACT FROM AUTHOR]
Copyright of Medical Image Analysis is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:Counting of mitotic figures is a fundamental step in grading and prognostication of several cancers. However, manual mitosis counting is tedious and time-consuming. In addition, variation in the appearance of mitotic figures causes a high degree of discordance among pathologists. With advances in deep learning models, several automatic mitosis detection algorithms have been proposed but they are sensitive to domain shift often seen in histology images. We propose a robust and efficient two-stage mitosis detection framework, which comprises mitosis candidate segmentation (Detecting Fast) and candidate refinement (Detecting Slow) stages. The proposed candidate segmentation model, termed EUNet , is fast and accurate due to its architectural design. EUNet can precisely segment candidates at a lower resolution to considerably speed up candidate detection. Candidates are then refined using a deeper classifier network, EfficientNet-B7, in the second stage. We make sure both stages are robust against domain shift by incorporating domain generalization methods. We demonstrate state-of-the-art performance and generalizability of the proposed model on the three largest publicly available mitosis datasets, winning the two mitosis domain generalization challenge contests (MIDOG21 and MIDOG22). Finally, we showcase the utility of the proposed algorithm by processing the TCGA breast cancer cohort (1,124 whole-slide images) to generate and release a repository of more than 620K potential mitotic figures (not exhaustively validated). • Novel two-stage method efficiently detects mitosis via deep learning models. • Self-supervised method pre-trains encoder and decoder of a segmentation model. • Exploring impacts of various domain generalization methods on mitosis segmentation. • AI results for TCGA-BRCA with 620K+ mitotic figures released, available online. • Outperforming other algorithms in mitosis detection and ranked 1st in MIDOG21/22. [ABSTRACT FROM AUTHOR]
ISSN:13618415
DOI:10.1016/j.media.2024.103132