Deep convolutional neural network with transfer learning for rectum toxicity prediction in cervical cancer radiotherapy

A feasibility study

Xin Zhen, Jiawei Chen, Zichun Zhong, Brian Hrycushko, Linghong Zhou, Steve Jiang, Kevin Albuquerque, Xuejun Gu

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Better understanding of the dose-toxicity relationship is critical for safe dose escalation to improve local control in late-stage cervical cancer radiotherapy. In this study, we introduced a convolutional neural network (CNN) model to analyze rectum dose distribution and predict rectum toxicity. Forty-two cervical cancer patients treated with combined external beam radiotherapy (EBRT) and brachytherapy (BT) were retrospectively collected, including twelve toxicity patients and thirty non-toxicity patients. We adopted a transfer learning strategy to overcome the limited patient data issue. A 16-layers CNN developed by the visual geometry group (VGG-16) of the University of Oxford was pre-trained on a large-scale natural image database, ImageNet, and fine-tuned with patient rectum surface dose maps (RSDMs), which were accumulated EBRT + BT doses on the unfolded rectum surface. We used the adaptive synthetic sampling approach and the data augmentation method to address the two challenges, data imbalance and data scarcity. The gradient-weighted class activation maps (Grad-CAM) were also generated to highlight the discriminative regions on the RSDM along with the prediction model. We compare different CNN coefficients fine-tuning strategies, and compare the predictive performance using the traditional dose volume parameters, e.g. D 0.1/1/2cc, and the texture features extracted from the RSDM. Satisfactory prediction performance was achieved with the proposed scheme, and we found that the mean Grad-CAM over the toxicity patient group has geometric consistence of distribution with the statistical analysis result, which indicates possible rectum toxicity location. The evaluation results have demonstrated the feasibility of building a CNN-based rectum dose-toxicity prediction model with transfer learning for cervical cancer radiotherapy.

Original languageEnglish (US)
Pages (from-to)8246-8263
Number of pages18
JournalPhysics in Medicine and Biology
Volume62
Issue number21
DOIs
StatePublished - Oct 11 2017

Fingerprint

Feasibility Studies
Rectum
Uterine Cervical Neoplasms
Radiotherapy
Brachytherapy
Statistical Distributions
Neural Networks (Computer)
Transfer (Psychology)
Databases

Keywords

  • convolutional neural networks
  • deformable image registration
  • rectum surface dose maps
  • rectum toxicity prediction
  • transfer learning

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging

Cite this

Deep convolutional neural network with transfer learning for rectum toxicity prediction in cervical cancer radiotherapy : A feasibility study. / Zhen, Xin; Chen, Jiawei; Zhong, Zichun; Hrycushko, Brian; Zhou, Linghong; Jiang, Steve; Albuquerque, Kevin; Gu, Xuejun.

In: Physics in Medicine and Biology, Vol. 62, No. 21, 11.10.2017, p. 8246-8263.

Research output: Contribution to journalArticle

@article{af80474fa85749b4a09f5fcd44a8b73b,
title = "Deep convolutional neural network with transfer learning for rectum toxicity prediction in cervical cancer radiotherapy: A feasibility study",
abstract = "Better understanding of the dose-toxicity relationship is critical for safe dose escalation to improve local control in late-stage cervical cancer radiotherapy. In this study, we introduced a convolutional neural network (CNN) model to analyze rectum dose distribution and predict rectum toxicity. Forty-two cervical cancer patients treated with combined external beam radiotherapy (EBRT) and brachytherapy (BT) were retrospectively collected, including twelve toxicity patients and thirty non-toxicity patients. We adopted a transfer learning strategy to overcome the limited patient data issue. A 16-layers CNN developed by the visual geometry group (VGG-16) of the University of Oxford was pre-trained on a large-scale natural image database, ImageNet, and fine-tuned with patient rectum surface dose maps (RSDMs), which were accumulated EBRT + BT doses on the unfolded rectum surface. We used the adaptive synthetic sampling approach and the data augmentation method to address the two challenges, data imbalance and data scarcity. The gradient-weighted class activation maps (Grad-CAM) were also generated to highlight the discriminative regions on the RSDM along with the prediction model. We compare different CNN coefficients fine-tuning strategies, and compare the predictive performance using the traditional dose volume parameters, e.g. D 0.1/1/2cc, and the texture features extracted from the RSDM. Satisfactory prediction performance was achieved with the proposed scheme, and we found that the mean Grad-CAM over the toxicity patient group has geometric consistence of distribution with the statistical analysis result, which indicates possible rectum toxicity location. The evaluation results have demonstrated the feasibility of building a CNN-based rectum dose-toxicity prediction model with transfer learning for cervical cancer radiotherapy.",
keywords = "convolutional neural networks, deformable image registration, rectum surface dose maps, rectum toxicity prediction, transfer learning",
author = "Xin Zhen and Jiawei Chen and Zichun Zhong and Brian Hrycushko and Linghong Zhou and Steve Jiang and Kevin Albuquerque and Xuejun Gu",
year = "2017",
month = "10",
day = "11",
doi = "10.1088/1361-6560/aa8d09",
language = "English (US)",
volume = "62",
pages = "8246--8263",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "21",

}

TY - JOUR

T1 - Deep convolutional neural network with transfer learning for rectum toxicity prediction in cervical cancer radiotherapy

T2 - A feasibility study

AU - Zhen, Xin

AU - Chen, Jiawei

AU - Zhong, Zichun

AU - Hrycushko, Brian

AU - Zhou, Linghong

AU - Jiang, Steve

AU - Albuquerque, Kevin

AU - Gu, Xuejun

PY - 2017/10/11

Y1 - 2017/10/11

N2 - Better understanding of the dose-toxicity relationship is critical for safe dose escalation to improve local control in late-stage cervical cancer radiotherapy. In this study, we introduced a convolutional neural network (CNN) model to analyze rectum dose distribution and predict rectum toxicity. Forty-two cervical cancer patients treated with combined external beam radiotherapy (EBRT) and brachytherapy (BT) were retrospectively collected, including twelve toxicity patients and thirty non-toxicity patients. We adopted a transfer learning strategy to overcome the limited patient data issue. A 16-layers CNN developed by the visual geometry group (VGG-16) of the University of Oxford was pre-trained on a large-scale natural image database, ImageNet, and fine-tuned with patient rectum surface dose maps (RSDMs), which were accumulated EBRT + BT doses on the unfolded rectum surface. We used the adaptive synthetic sampling approach and the data augmentation method to address the two challenges, data imbalance and data scarcity. The gradient-weighted class activation maps (Grad-CAM) were also generated to highlight the discriminative regions on the RSDM along with the prediction model. We compare different CNN coefficients fine-tuning strategies, and compare the predictive performance using the traditional dose volume parameters, e.g. D 0.1/1/2cc, and the texture features extracted from the RSDM. Satisfactory prediction performance was achieved with the proposed scheme, and we found that the mean Grad-CAM over the toxicity patient group has geometric consistence of distribution with the statistical analysis result, which indicates possible rectum toxicity location. The evaluation results have demonstrated the feasibility of building a CNN-based rectum dose-toxicity prediction model with transfer learning for cervical cancer radiotherapy.

AB - Better understanding of the dose-toxicity relationship is critical for safe dose escalation to improve local control in late-stage cervical cancer radiotherapy. In this study, we introduced a convolutional neural network (CNN) model to analyze rectum dose distribution and predict rectum toxicity. Forty-two cervical cancer patients treated with combined external beam radiotherapy (EBRT) and brachytherapy (BT) were retrospectively collected, including twelve toxicity patients and thirty non-toxicity patients. We adopted a transfer learning strategy to overcome the limited patient data issue. A 16-layers CNN developed by the visual geometry group (VGG-16) of the University of Oxford was pre-trained on a large-scale natural image database, ImageNet, and fine-tuned with patient rectum surface dose maps (RSDMs), which were accumulated EBRT + BT doses on the unfolded rectum surface. We used the adaptive synthetic sampling approach and the data augmentation method to address the two challenges, data imbalance and data scarcity. The gradient-weighted class activation maps (Grad-CAM) were also generated to highlight the discriminative regions on the RSDM along with the prediction model. We compare different CNN coefficients fine-tuning strategies, and compare the predictive performance using the traditional dose volume parameters, e.g. D 0.1/1/2cc, and the texture features extracted from the RSDM. Satisfactory prediction performance was achieved with the proposed scheme, and we found that the mean Grad-CAM over the toxicity patient group has geometric consistence of distribution with the statistical analysis result, which indicates possible rectum toxicity location. The evaluation results have demonstrated the feasibility of building a CNN-based rectum dose-toxicity prediction model with transfer learning for cervical cancer radiotherapy.

KW - convolutional neural networks

KW - deformable image registration

KW - rectum surface dose maps

KW - rectum toxicity prediction

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85032684258&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032684258&partnerID=8YFLogxK

U2 - 10.1088/1361-6560/aa8d09

DO - 10.1088/1361-6560/aa8d09

M3 - Article

VL - 62

SP - 8246

EP - 8263

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 21

ER -