A deep learning-based segmentation system for rapid onsite cytologic pathology evaluation of pancreatic masses: A retrospective, multicenter, diagnostic study

Song Zhang; Yangfan Zhou; Dehua Tang; Muhan Ni; Jinyu Zheng; Guifang Xu; Chunyan Peng; Shanshan Shen; Qiang Zhan; Xiaoyun Wang; Duanmin Hu; Wu-Jun Li; Lei Wang; Ying Lv; Xiaoping Zou

doi:10.1016/j.ebiom.2022.104022

A deep learning-based segmentation system for rapid onsite cytologic pathology evaluation of pancreatic masses: A retrospective, multicenter, diagnostic study

EBioMedicine. 2022 Jun:80:104022. doi: 10.1016/j.ebiom.2022.104022. Epub 2022 May 2.

Authors

Song Zhang¹, Yangfan Zhou², Dehua Tang³, Muhan Ni³, Jinyu Zheng⁴, Guifang Xu¹, Chunyan Peng¹, Shanshan Shen¹, Qiang Zhan⁵, Xiaoyun Wang⁶, Duanmin Hu⁷, Wu-Jun Li⁸, Lei Wang⁹, Ying Lv¹⁰, Xiaoping Zou¹¹

Affiliations

¹ Department of Gastroenterology, Nanjing Drum Tower Hospital, Clinical College of Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, Jiangsu 210008, China; Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China.
² National Institute of Healthcare Data Science at Nanjing University, Nanjing, Jiangsu 210008, China; National Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu 210008, China.
³ Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China.
⁴ Department of Pathology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China.
⁵ Department of Gastroenterology, Wuxi People's Hospital Affiliated to Nanjing Medical University, Wuxi, Jiangsu 214023, China.
⁶ Department of Gastroenterology, Wuxi Second People's Hospital Affiliated to Nanjing Medical University, Wuxi, Jiangsu 214002, China.
⁷ Department of Gastroenterology, The Second Affiliated Hospital of Soochow University, Suzhou, Jiangsu 215004, China.
⁸ National Institute of Healthcare Data Science at Nanjing University, Nanjing, Jiangsu 210008, China; National Key Laboratory for Novel Software Technology, Department of Computer Science and Technology, Nanjing University, Nanjing, Jiangsu 210008, China; Center for Medical Big Data, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China. Electronic address: liwujun@nju.edu.cn.
⁹ Department of Gastroenterology, Nanjing Drum Tower Hospital, Clinical College of Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, Jiangsu 210008, China; Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China. Electronic address: leiwang9631@nju.edu.cn.
¹⁰ Department of Gastroenterology, Nanjing Drum Tower Hospital, Clinical College of Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, Jiangsu 210008, China; Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China. Electronic address: lvying@njglyy.com.
¹¹ Department of Gastroenterology, Nanjing Drum Tower Hospital, Clinical College of Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, Jiangsu 210008, China; Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210008, China. Electronic address: zouxp@nju.edu.cn.

Abstract

Background: We aimed to develop a deep learning-based segmentation system for rapid on-site cytopathology evaluation (ROSE) to improve the diagnostic efficiency of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) biopsy.

Methods: A retrospective, multicenter, diagnostic study was conducted using 5345 cytopathological slide images from 194 patients who underwent EUS-FNA. These patients were from Nanjing Drum Tower Hospital (109 patients), Wuxi People's Hospital (30 patients), Wuxi Second People's Hospital (25 patients), and The Second Affiliated Hospital of Soochow University (30 patients). A deep convolutional neural network (DCNN) system was developed to segment cell clusters and identify cancer cell clusters with cytopathological slide images. Internal testing, external testing, subgroup analysis, and human-machine competition were used to evaluate the performance of the system.

Findings: The DCNN system segmented stained cells from the background in cytopathological slides with an F1-score of 0·929 and 0·899-0·938 in internal and external testing, respectively. For cancer identification, the DCNN system identified images containing cancer clusters with AUCs of 0·958 and 0·948-0·976 in internal and external testing, respectively. The generalizable and robust performance of the DCNN system was validated in sensitivity analysis (AUC > 0·900) and was superior to that of trained endoscopists and comparable to cytopathologists on our testing datasets.

Interpretation: The DCNN system is feasible and robust for identifying sample adequacy and pancreatic cancer cell clusters. Prospective studies are warranted to evaluate the clinical significance of the system.

Funding: Jiangsu Natural Science Foundation; Nanjing Medical Science and Technology Development Funding; National Natural Science Foundation of China.

Keywords: Deep convolutional neural network; EUS-FNA; Pancreatic mass; Rapid on-site cytopathology evaluation.

Publication types

Multicenter Study

MeSH terms

Deep Learning*
Endoscopic Ultrasound-Guided Fine Needle Aspiration / methods
Humans
Pancreatic Neoplasms* / diagnostic imaging
Pancreatic Neoplasms* / pathology
Prospective Studies
Retrospective Studies