lancet-header

Preprints with The Lancet is part of SSRN´s First Look, a place where journals identify content of interest prior to publication. Authors have opted in at submission to The Lancet family of journals to post their preprints on Preprints with The Lancet. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early stage research papers that have not been peer-reviewed. The findings should not be used for clinical or public health decision making and should not be presented to a lay audience without highlighting that they are preliminary and have not been peer-reviewed. For more information on this collaboration, see the comments published in The Lancet about the trial period, and our decision to make this a permanent offering, or visit The Lancet´s FAQ page, and for any feedback please contact preprints@lancet.com.

Comprehensive AI Model Development for Gleason Grading: From Scanning, Cloud-Based Annotation to Pathologist-AI Interaction

33 Pages Posted: 25 Jul 2022

See all articles by Xinmi Huo

Xinmi Huo

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Kok Haur Ong

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Kah Weng Lau

National University of Singapore (NUS) - National University Hospital

Laurent Gole

University of Tartu - Institute of Molecular and Cell Biology

Char Loo Tan

National University of Singapore (NUS) - National University Hospital

Chongchong Zhang

Government of the People's Republic of China - Department of Pathology

Yonghui Zhang

Government of the People's Republic of China - Department of Pathology

Xiaohui Zhu

Southern Medical University - Department of Pathology

Longjie Li

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Hao Han

University of Tartu - Institute of Molecular and Cell Biology

David Young

University of Tartu - Institute of Molecular and Cell Biology

Haoda Lu

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Jun Xu

Nanjing University of Information Science & Technology (NUIST) - Jiangsu Key Laboratory of Big Data Analysis Technique; Nanjing University of Information Science & Technology (NUIST) - CICAEET

Wanyuan Chen

Hangzhou Medical College - Cancer Center

Stephan J. Sanders

University of California, San Diego (UCSD) - Department of Psychiatry

Lee Hwee Kuan

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Susan Swee-Shan Hue

National University of Singapore (NUS) - National University Hospital

Weimiao YU

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute

Soo Yong Tan

National University of Singapore (NUS) - National University Hospital

More...

Abstract

Background: AI-based solutions for automated Gleason grading have been developed to assist pathologists to make rapid and quantitative assessments, but the generalization across various scanners and updating AI models continuously using new annotated data from end users remains a key bottleneck in the field.

Methods: We proposed an comprehensive digital pathology workflow for AI-assisted Gleason grading, incorporating an image quality check software A!magQC, a cloud-based annotation platform A!HistoNotes and Pathologist-AI Interaction (PAI) strategy. In this study, 187 prostatectomy and 156 biopsy specimens (24,371 mm2 of annotations) were collected from National University Hospital Singapore. Images were first acquired from Akoya Biosciences scanner. In addition, a subset of prostatectomy specimens was scanned by 4 other scanners. Image normalization and color augmentation were applied to increase the generalizability of the model. Lastly, clinical experiments were conducted with 5 pathologists from Singapore and China to investigate the accuracy, efficiency, and consistency of Gleason grading assisted by our model. Furthermore, we performed semi-automatic annotation and attempted to update the existing model using new data, enabling the PAI.

Finding: For images scanned by Akoya Biosciences scanner, on average, our AI model achieves a sensitivity of 85%, specificity of 96% and F1 score of 78% on Gleason grading for prostatectomy specimens, and 96% sensitivity on tumor detection for biopsy specimens. For images scanned by other 4 scanners, the average F1 score increased from 67% to 75% after adopting our generalization solution. In the clinical experiments, our model was integrated into A!HistoNotes and accelerated the Gleason scoring by 43%.  Furthermore, it increased the annotation efficiency by 60% and was improved by incorporating semi-automatic annotations and incremental learning.

Interpretation: The proposed workflow helps to develop a generalised AI model for Gleason grading to assist pathologists in routine work. It can be applied to images scanned by a wide range of scanners. In addition to AI-assisted diagnosis, the model can be updated through the PAI efficiently.

Funding This project is funded by Exploit Technologies Gap-funded Project: “AI based H&E image analysis for prostate cancer staging” (EPTL/18-Gap029-R20H).

Declaration of Interest: All authors declare no competing interests.

Ethical Approval: Prostatectomy and biopsy formalin-fixed paraffin-embedded (FFPE) tissue specimens were collected from the Department of Pathology, National University of Singapore Hospital (NUHS) approved by the NHG Institutional Review Board (DSRB study reference number: 2018/01186).

Keywords: Digital Pathology, artificial intelligence, deep learning, Prostate Cancer, Pathologist-AI Interaction, Incremental Learning, Cloud Annotation, WSI Image Quality Control

Suggested Citation

Huo, Xinmi and Ong, Kok Haur and Lau, Kah Weng and Gole, Laurent and Tan, Char Loo and Zhang, Chongchong and Zhang, Yonghui and Zhu, Xiaohui and Li, Longjie and Han, Hao and Young, David and Lu, Haoda and Xu, Jun and Chen, Wanyuan and Sanders, Stephan J. and Kuan, Lee Hwee and Hue, Susan Swee-Shan and YU, Weimiao and Tan, Soo Yong, Comprehensive AI Model Development for Gleason Grading: From Scanning, Cloud-Based Annotation to Pathologist-AI Interaction. Available at SSRN: https://ssrn.com/abstract=4172090 or http://dx.doi.org/10.2139/ssrn.4172090

Xinmi Huo

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

Kok Haur Ong

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

30 Biopolis Street
#07-01 Matrix
Singapore, 138671
China

Kah Weng Lau

National University of Singapore (NUS) - National University Hospital ( email )

Level 9
1E Kent Ridge Road
119228
Singapore

Laurent Gole

University of Tartu - Institute of Molecular and Cell Biology ( email )

Tartu
Estonia

Char Loo Tan

National University of Singapore (NUS) - National University Hospital ( email )

Level 9
1E Kent Ridge Road
119228
Singapore

Chongchong Zhang

Government of the People's Republic of China - Department of Pathology ( email )

China

Yonghui Zhang

Government of the People's Republic of China - Department of Pathology ( email )

Xiaohui Zhu

Southern Medical University - Department of Pathology ( email )

Longjie Li

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

30 Biopolis Street
#07-01 Matrix
Singapore, 138671
China

Hao Han

University of Tartu - Institute of Molecular and Cell Biology ( email )

Tartu
Estonia

David Young

University of Tartu - Institute of Molecular and Cell Biology ( email )

Tartu
Estonia

Haoda Lu

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

Jun Xu

Nanjing University of Information Science & Technology (NUIST) - Jiangsu Key Laboratory of Big Data Analysis Technique ( email )

Nanjing, 210044
China

Nanjing University of Information Science & Technology (NUIST) - CICAEET

Nanjing, 210044
China

Wanyuan Chen

Hangzhou Medical College - Cancer Center ( email )

Stephan J. Sanders

University of California, San Diego (UCSD) - Department of Psychiatry ( email )

Lee Hwee Kuan

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

30 Biopolis Street
#07-01 Matrix
Singapore, 138671
China

Susan Swee-Shan Hue

National University of Singapore (NUS) - National University Hospital ( email )

Level 9
1E Kent Ridge Road
119228
Singapore

Weimiao YU (Contact Author)

Agency for Science, Technology and Research (A*STAR) - Bioinformatics Institute ( email )

30 Biopolis Street
#07-01 Matrix
Singapore, 138671
China

Soo Yong Tan

National University of Singapore (NUS) - National University Hospital ( email )

Level 9
1E Kent Ridge Road
119228
Singapore

Click here to go to TheLancet.com

Paper statistics

Downloads
155
Abstract Views
782
PlumX Metrics