Deep neural network trained on gigapixel images improves lymph node metastasis detection in clinical settings

Shih-Chiang Huang, Chi-Chung Chen, Jui Lan, Tsan-Yu Hsieh, Huei-Chieh Chuang, Meng-Yao Chien, Tao-Sheng Ou, Kuang-Hua Chen, Ren-Chin Wu, Yu-Jen Liu, Chi-Tung Cheng, Yu-Jen Huang, Liang-Wei Tao, An-Fong Hwu, I-Chieh Lin, Shih-Hao Hung, Chao-Yuan Yeh () and Tse-Ching Chen ()
Additional contact information
Shih-Chiang Huang: Chang Gung University, College of Medicine
Chi-Chung Chen: aetherAI Co., Ltd.
Jui Lan: Chang Gung University, College of Medicine
Tsan-Yu Hsieh: Chang Gung University, College of Medicine
Huei-Chieh Chuang: Chang Gung University, College of Medicine
Meng-Yao Chien: aetherAI Co., Ltd.
Tao-Sheng Ou: aetherAI Co., Ltd.
Kuang-Hua Chen: Chang Gung University, College of Medicine
Ren-Chin Wu: Chang Gung University, College of Medicine
Yu-Jen Liu: Chang Gung University, College of Medicine
Chi-Tung Cheng: Chang Gung University, College of Medicine
Yu-Jen Huang: National Taiwan University
Liang-Wei Tao: National Taiwan University
An-Fong Hwu: National Taiwan University
I-Chieh Lin: Chang Gung University, College of Medicine
Shih-Hao Hung: National Taiwan University
Chao-Yuan Yeh: aetherAI Co., Ltd.
Tse-Ching Chen: Chang Gung University, College of Medicine

Nature Communications, 2022, vol. 13, issue 1, 1-14

Abstract: Abstract The pathological identification of lymph node (LN) metastasis is demanding and tedious. Although convolutional neural networks (CNNs) possess considerable potential in improving the process, the ultrahigh-resolution of whole slide images hinders the development of a clinically applicable solution. We design an artificial-intelligence-assisted LN assessment workflow to facilitate the routine counting of metastatic LNs. Unlike previous patch-based approaches, our proposed method trains CNNs by using 5-gigapixel images, obviating the need for lesion-level annotations. Trained on 5907 LN images, our algorithm identifies metastatic LNs in gastric cancer with a slide-level area under the receiver operating characteristic curve (AUC) of 0.9936. Clinical experiments reveal that the workflow significantly improves the sensitivity of micrometastasis identification (81.94% to 95.83%, P

Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-022-30746-1 Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:13:y:2022:i:1:d:10.1038_s41467-022-30746-1

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-022-30746-1

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().