Magnifying Networks for Histopathological Images with Billions of Pixels
Amongst the other benefits conferred by the shift from traditional to digital pathology is the potential to use machine learning for diagnosis, prognosis, and personalization. A major challenge in the realization of this potential emerges from the extremely large size of digitized images, which are...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/14/5/524 |
| _version_ | 1797264672416595968 |
|---|---|
| author | Neofytos Dimitriou Ognjen Arandjelović David J. Harrison |
| author_facet | Neofytos Dimitriou Ognjen Arandjelović David J. Harrison |
| author_sort | Neofytos Dimitriou |
| collection | DOAJ |
| description | Amongst the other benefits conferred by the shift from traditional to digital pathology is the potential to use machine learning for diagnosis, prognosis, and personalization. A major challenge in the realization of this potential emerges from the extremely large size of digitized images, which are often in excess of 100,000 × 100,000 pixels. In this paper, we tackle this challenge head-on by diverging from the existing approaches in the literature—which rely on the splitting of the original images into small patches—and introducing magnifying networks (MagNets). By using an attention mechanism, MagNets identify the regions of the gigapixel image that benefit from an analysis on a finer scale. This process is repeated, resulting in an attention-driven coarse-to-fine analysis of only a small portion of the information contained in the original whole-slide images. Importantly, this is achieved using minimal ground truth annotation, namely, using only global, slide-level labels. The results from our tests on the publicly available Camelyon16 and Camelyon17 datasets demonstrate the effectiveness of MagNets—as well as the proposed optimization framework—in the task of whole-slide image classification. Importantly, MagNets process at least five times fewer patches from each whole-slide image than any of the existing end-to-end approaches. |
| first_indexed | 2024-04-25T00:32:37Z |
| format | Article |
| id | doaj.art-d1621a6c9f78452d960cf19d27812af6 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-04-25T00:32:37Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-d1621a6c9f78452d960cf19d27812af62024-03-12T16:42:03ZengMDPI AGDiagnostics2075-44182024-03-0114552410.3390/diagnostics14050524Magnifying Networks for Histopathological Images with Billions of PixelsNeofytos Dimitriou0Ognjen Arandjelović1David J. Harrison2Maritime Digitalisation Centre, Cyprus Marine and Maritime Institute, Larnaca 6300, CyprusSchool of Computer Science, University of St Andrews, St Andrews KY16 9SX, UKSchool of Medicine, University of St Andrews, St Andrews KY16 9TF, UKAmongst the other benefits conferred by the shift from traditional to digital pathology is the potential to use machine learning for diagnosis, prognosis, and personalization. A major challenge in the realization of this potential emerges from the extremely large size of digitized images, which are often in excess of 100,000 × 100,000 pixels. In this paper, we tackle this challenge head-on by diverging from the existing approaches in the literature—which rely on the splitting of the original images into small patches—and introducing magnifying networks (MagNets). By using an attention mechanism, MagNets identify the regions of the gigapixel image that benefit from an analysis on a finer scale. This process is repeated, resulting in an attention-driven coarse-to-fine analysis of only a small portion of the information contained in the original whole-slide images. Importantly, this is achieved using minimal ground truth annotation, namely, using only global, slide-level labels. The results from our tests on the publicly available Camelyon16 and Camelyon17 datasets demonstrate the effectiveness of MagNets—as well as the proposed optimization framework—in the task of whole-slide image classification. Importantly, MagNets process at least five times fewer patches from each whole-slide image than any of the existing end-to-end approaches.https://www.mdpi.com/2075-4418/14/5/524histologyhistopathologydeep learningwhole slide imagedigital pathologygigapixel images |
| spellingShingle | Neofytos Dimitriou Ognjen Arandjelović David J. Harrison Magnifying Networks for Histopathological Images with Billions of Pixels Diagnostics histology histopathology deep learning whole slide image digital pathology gigapixel images |
| title | Magnifying Networks for Histopathological Images with Billions of Pixels |
| title_full | Magnifying Networks for Histopathological Images with Billions of Pixels |
| title_fullStr | Magnifying Networks for Histopathological Images with Billions of Pixels |
| title_full_unstemmed | Magnifying Networks for Histopathological Images with Billions of Pixels |
| title_short | Magnifying Networks for Histopathological Images with Billions of Pixels |
| title_sort | magnifying networks for histopathological images with billions of pixels |
| topic | histology histopathology deep learning whole slide image digital pathology gigapixel images |
| url | https://www.mdpi.com/2075-4418/14/5/524 |
| work_keys_str_mv | AT neofytosdimitriou magnifyingnetworksforhistopathologicalimageswithbillionsofpixels AT ognjenarandjelovic magnifyingnetworksforhistopathologicalimageswithbillionsofpixels AT davidjharrison magnifyingnetworksforhistopathologicalimageswithbillionsofpixels |