Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography
In recent year, many deep learning have been playing an important role in the detection of cancers. This study aimed to real-timely differentiate a pancreatic cancer (PC) or a non-pancreatic cancer (NPC) lesion via endoscopic ultrasonography (EUS) image. A total of 1213 EUS images from 157 patients...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-10-01
|
Series: | Frontiers in Oncology |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2022.973652/full |
_version_ | 1811344850070536192 |
---|---|
author | Guo Tian Guo Tian Guo Tian Danxia Xu Danxia Xu Yinghua He Yinghua He Weilu Chai Weilu Chai Zhuang Deng Chao Cheng Xinyan Jin Guyue Wei Qiyu Zhao Qiyu Zhao Tianan Jiang Tianan Jiang Tianan Jiang |
author_facet | Guo Tian Guo Tian Guo Tian Danxia Xu Danxia Xu Yinghua He Yinghua He Weilu Chai Weilu Chai Zhuang Deng Chao Cheng Xinyan Jin Guyue Wei Qiyu Zhao Qiyu Zhao Tianan Jiang Tianan Jiang Tianan Jiang |
author_sort | Guo Tian |
collection | DOAJ |
description | In recent year, many deep learning have been playing an important role in the detection of cancers. This study aimed to real-timely differentiate a pancreatic cancer (PC) or a non-pancreatic cancer (NPC) lesion via endoscopic ultrasonography (EUS) image. A total of 1213 EUS images from 157 patients (99 male, 58 female) with pancreatic disease were used for training, validation and test groups. Before model training, regions of interest (ROIs) were manually drawn to mark the PC and NPC lesions using Labelimage software. Yolov5m was used as the algorithm model to automatically distinguish the presence of pancreatic lesion. After training the model based on EUS images using YOLOv5, the parameters achieved convergence within 300 rounds (GIoU Loss: 0.01532, Objectness Loss: 0.01247, precision: 0.713 and recall: 0.825). For the validation group, the mAP0.5 was 0.831, and mAP@.5:.95 was 0.512. In addition, the receiver operating characteristic (ROC) curve analysis showed this model seemed to have a trend of more AUC of 0.85 (0.665 to 0.956) than the area under the curve (AUC) of 0.838 (0.65 to 0.949) generated by physicians using EUS detection without puncture, although pairwise comparison of ROC curves showed that the AUC between the two groups was not significant (z= 0.15, p = 0.8804). This study suggested that the YOLOv5m would generate attractive results and allow for the real-time decision support for distinction of a PC or a NPC lesion. |
first_indexed | 2024-04-13T19:54:16Z |
format | Article |
id | doaj.art-11fa04873c1740f58148b76af1a4c4f0 |
institution | Directory Open Access Journal |
issn | 2234-943X |
language | English |
last_indexed | 2024-04-13T19:54:16Z |
publishDate | 2022-10-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Oncology |
spelling | doaj.art-11fa04873c1740f58148b76af1a4c4f02022-12-22T02:32:23ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-10-011210.3389/fonc.2022.973652973652Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonographyGuo Tian0Guo Tian1Guo Tian2Danxia Xu3Danxia Xu4Yinghua He5Yinghua He6Weilu Chai7Weilu Chai8Zhuang Deng9Chao Cheng10Xinyan Jin11Guyue Wei12Qiyu Zhao13Qiyu Zhao14Tianan Jiang15Tianan Jiang16Tianan Jiang17Department of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaKey Laboratory of Pulsed Power Translational Medicine of Zhejiang Province, Hangzhou, ChinaState Key Laboratory for Diagnosis and Treatment of Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaKey Laboratory of Pulsed Power Translational Medicine of Zhejiang Province, Hangzhou, ChinaDepartment of Clinical Pharmacy, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaZhejiang Provincial Key Laboratory for Drug Evaluation and Clinical Research, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaKey Laboratory of Pulsed Power Translational Medicine of Zhejiang Province, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaKey Laboratory of Pulsed Power Translational Medicine of Zhejiang Province, Hangzhou, ChinaDepartment of Ultrasound Medicine, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaKey Laboratory of Pulsed Power Translational Medicine of Zhejiang Province, Hangzhou, ChinaZhejiang University Cancer Center, Hangzhou, ChinaIn recent year, many deep learning have been playing an important role in the detection of cancers. This study aimed to real-timely differentiate a pancreatic cancer (PC) or a non-pancreatic cancer (NPC) lesion via endoscopic ultrasonography (EUS) image. A total of 1213 EUS images from 157 patients (99 male, 58 female) with pancreatic disease were used for training, validation and test groups. Before model training, regions of interest (ROIs) were manually drawn to mark the PC and NPC lesions using Labelimage software. Yolov5m was used as the algorithm model to automatically distinguish the presence of pancreatic lesion. After training the model based on EUS images using YOLOv5, the parameters achieved convergence within 300 rounds (GIoU Loss: 0.01532, Objectness Loss: 0.01247, precision: 0.713 and recall: 0.825). For the validation group, the mAP0.5 was 0.831, and mAP@.5:.95 was 0.512. In addition, the receiver operating characteristic (ROC) curve analysis showed this model seemed to have a trend of more AUC of 0.85 (0.665 to 0.956) than the area under the curve (AUC) of 0.838 (0.65 to 0.949) generated by physicians using EUS detection without puncture, although pairwise comparison of ROC curves showed that the AUC between the two groups was not significant (z= 0.15, p = 0.8804). This study suggested that the YOLOv5m would generate attractive results and allow for the real-time decision support for distinction of a PC or a NPC lesion.https://www.frontiersin.org/articles/10.3389/fonc.2022.973652/fulldeep learningendoscopic ultrasonographydiagnosisultrasonographypancreatic lesion |
spellingShingle | Guo Tian Guo Tian Guo Tian Danxia Xu Danxia Xu Yinghua He Yinghua He Weilu Chai Weilu Chai Zhuang Deng Chao Cheng Xinyan Jin Guyue Wei Qiyu Zhao Qiyu Zhao Tianan Jiang Tianan Jiang Tianan Jiang Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography Frontiers in Oncology deep learning endoscopic ultrasonography diagnosis ultrasonography pancreatic lesion |
title | Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
title_full | Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
title_fullStr | Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
title_full_unstemmed | Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
title_short | Deep learning for real-time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
title_sort | deep learning for real time auxiliary diagnosis of pancreatic cancer in endoscopic ultrasonography |
topic | deep learning endoscopic ultrasonography diagnosis ultrasonography pancreatic lesion |
url | https://www.frontiersin.org/articles/10.3389/fonc.2022.973652/full |
work_keys_str_mv | AT guotian deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT guotian deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT guotian deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT danxiaxu deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT danxiaxu deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT yinghuahe deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT yinghuahe deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT weiluchai deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT weiluchai deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT zhuangdeng deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT chaocheng deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT xinyanjin deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT guyuewei deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT qiyuzhao deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT qiyuzhao deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT tiananjiang deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT tiananjiang deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography AT tiananjiang deeplearningforrealtimeauxiliarydiagnosisofpancreaticcancerinendoscopicultrasonography |