Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT
BackgroundThe current clinical workflow for esophageal gross tumor volume (GTV) contouring relies on manual delineation with high labor costs and inter-user variability.PurposeTo validate the clinical applicability of a deep learning multimodality esophageal GTV contouring model, developed at one in...
Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-01-01
|
Series: | Frontiers in Oncology |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fonc.2021.785788/full |
_version_ | 1798027115758616576 |
---|---|
author | Xianghua Ye Dazhou Guo Chen-Kan Tseng Jia Ge Tsung-Min Hung Ping-Ching Pai Yanping Ren Lu Zheng Xinli Zhu Ling Peng Ying Chen Xiaohua Chen Chen-Yu Chou Danni Chen Jiaze Yu Yuzhen Chen Feiran Jiao Yi Xin Lingyun Huang Guotong Xie Jing Xiao Le Lu Senxiang Yan Dakai Jin Tsung-Ying Ho |
author_facet | Xianghua Ye Dazhou Guo Chen-Kan Tseng Jia Ge Tsung-Min Hung Ping-Ching Pai Yanping Ren Lu Zheng Xinli Zhu Ling Peng Ying Chen Xiaohua Chen Chen-Yu Chou Danni Chen Jiaze Yu Yuzhen Chen Feiran Jiao Yi Xin Lingyun Huang Guotong Xie Jing Xiao Le Lu Senxiang Yan Dakai Jin Tsung-Ying Ho |
author_sort | Xianghua Ye |
collection | DOAJ |
description | BackgroundThe current clinical workflow for esophageal gross tumor volume (GTV) contouring relies on manual delineation with high labor costs and inter-user variability.PurposeTo validate the clinical applicability of a deep learning multimodality esophageal GTV contouring model, developed at one institution whereas tested at multiple institutions.Materials and MethodsWe collected 606 patients with esophageal cancer retrospectively from four institutions. Among them, 252 patients from institution 1 contained both a treatment planning CT (pCT) and a pair of diagnostic FDG-PET/CT; 354 patients from three other institutions had only pCT scans under different staging protocols or lacking PET scanners. A two-streamed deep learning model for GTV segmentation was developed using pCT and PET/CT scans of a subset (148 patients) from institution 1. This built model had the flexibility of segmenting GTVs via only pCT or pCT+PET/CT combined when available. For independent evaluation, the remaining 104 patients from institution 1 behaved as an unseen internal testing, and 354 patients from the other three institutions were used for external testing. Degrees of manual revision were further evaluated by human experts to assess the contour-editing effort. Furthermore, the deep model’s performance was compared against four radiation oncologists in a multi-user study using 20 randomly chosen external patients. Contouring accuracy and time were recorded for the pre- and post-deep learning-assisted delineation process. |
first_indexed | 2024-04-11T18:46:17Z |
format | Article |
id | doaj.art-5f7ed8fc13024f389b266fe50ea57f9c |
institution | Directory Open Access Journal |
issn | 2234-943X |
language | English |
last_indexed | 2024-04-11T18:46:17Z |
publishDate | 2022-01-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Oncology |
spelling | doaj.art-5f7ed8fc13024f389b266fe50ea57f9c2022-12-22T04:08:45ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2022-01-011110.3389/fonc.2021.785788785788Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CTXianghua Ye0Dazhou Guo1Chen-Kan Tseng2Jia Ge3Tsung-Min Hung4Ping-Ching Pai5Yanping Ren6Lu Zheng7Xinli Zhu8Ling Peng9Ying Chen10Xiaohua Chen11Chen-Yu Chou12Danni Chen13Jiaze Yu14Yuzhen Chen15Feiran Jiao16Yi Xin17Lingyun Huang18Guotong Xie19Jing Xiao20Le Lu21Senxiang Yan22Dakai Jin23Tsung-Ying Ho24Department of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaPAII Inc., Bethesda, MD, United StatesDepartment of Radiation Oncology, Chang Gung Memorial Hospital, Linkou, TaiwanDepartment of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaDepartment of Radiation Oncology, Chang Gung Memorial Hospital, Linkou, TaiwanDepartment of Radiation Oncology, Chang Gung Memorial Hospital, Linkou, TaiwanDepartment of Radiation Oncology, Huadong Hospital Affiliated to Fudan University, Shanghai, ChinaDepartment of Radiation Oncology, Lihuili Hospital, Ningbo Medical Center, Ningbo, ChinaDepartment of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaDepartment of Respiratory Disease, Zhejiang Provincial People’s Hospital, Hangzhou, ChinaDepartment of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaDepartment of Radiation Oncology, The First Hospital of Lanzhou University, Lanzhou, ChinaDepartment of Radiation Oncology, Chang Gung Memorial Hospital, Linkou, TaiwanDepartment of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaDepartment of Radiation Oncology, Haining People’s Hospital, Jiaxing, ChinaDepartment of Nuclear Medicine, Chang Gung Memorial Hospital, Linkou, Taiwan0Independent Researcher, Silver Spring, MD, United States1Ping An Technology, Shenzhen, China1Ping An Technology, Shenzhen, China1Ping An Technology, Shenzhen, China1Ping An Technology, Shenzhen, ChinaPAII Inc., Bethesda, MD, United StatesDepartment of Radiation Oncology, The First Affiliated Hospital Zhejiang University, Hangzhou, ChinaPAII Inc., Bethesda, MD, United StatesDepartment of Nuclear Medicine, Chang Gung Memorial Hospital, Linkou, TaiwanBackgroundThe current clinical workflow for esophageal gross tumor volume (GTV) contouring relies on manual delineation with high labor costs and inter-user variability.PurposeTo validate the clinical applicability of a deep learning multimodality esophageal GTV contouring model, developed at one institution whereas tested at multiple institutions.Materials and MethodsWe collected 606 patients with esophageal cancer retrospectively from four institutions. Among them, 252 patients from institution 1 contained both a treatment planning CT (pCT) and a pair of diagnostic FDG-PET/CT; 354 patients from three other institutions had only pCT scans under different staging protocols or lacking PET scanners. A two-streamed deep learning model for GTV segmentation was developed using pCT and PET/CT scans of a subset (148 patients) from institution 1. This built model had the flexibility of segmenting GTVs via only pCT or pCT+PET/CT combined when available. For independent evaluation, the remaining 104 patients from institution 1 behaved as an unseen internal testing, and 354 patients from the other three institutions were used for external testing. Degrees of manual revision were further evaluated by human experts to assess the contour-editing effort. Furthermore, the deep model’s performance was compared against four radiation oncologists in a multi-user study using 20 randomly chosen external patients. Contouring accuracy and time were recorded for the pre- and post-deep learning-assisted delineation process.https://www.frontiersin.org/articles/10.3389/fonc.2021.785788/fulldeep learningPET/CT (18)F-FDGradiotherapysegmentationdelineationesophageal cancer |
spellingShingle | Xianghua Ye Dazhou Guo Chen-Kan Tseng Jia Ge Tsung-Min Hung Ping-Ching Pai Yanping Ren Lu Zheng Xinli Zhu Ling Peng Ying Chen Xiaohua Chen Chen-Yu Chou Danni Chen Jiaze Yu Yuzhen Chen Feiran Jiao Yi Xin Lingyun Huang Guotong Xie Jing Xiao Le Lu Senxiang Yan Dakai Jin Tsung-Ying Ho Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT Frontiers in Oncology deep learning PET/CT (18)F-FDG radiotherapy segmentation delineation esophageal cancer |
title | Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT |
title_full | Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT |
title_fullStr | Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT |
title_full_unstemmed | Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT |
title_short | Multi-Institutional Validation of Two-Streamed Deep Learning Method for Automated Delineation of Esophageal Gross Tumor Volume Using Planning CT and FDG-PET/CT |
title_sort | multi institutional validation of two streamed deep learning method for automated delineation of esophageal gross tumor volume using planning ct and fdg pet ct |
topic | deep learning PET/CT (18)F-FDG radiotherapy segmentation delineation esophageal cancer |
url | https://www.frontiersin.org/articles/10.3389/fonc.2021.785788/full |
work_keys_str_mv | AT xianghuaye multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT dazhouguo multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT chenkantseng multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT jiage multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT tsungminhung multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT pingchingpai multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT yanpingren multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT luzheng multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT xinlizhu multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT lingpeng multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT yingchen multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT xiaohuachen multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT chenyuchou multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT dannichen multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT jiazeyu multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT yuzhenchen multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT feiranjiao multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT yixin multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT lingyunhuang multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT guotongxie multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT jingxiao multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT lelu multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT senxiangyan multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT dakaijin multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct AT tsungyingho multiinstitutionalvalidationoftwostreameddeeplearningmethodforautomateddelineationofesophagealgrosstumorvolumeusingplanningctandfdgpetct |