Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization

Breast-conserving surgery requires supportive radiotherapy to prevent cancer recurrence. However, the task of localizing the tumor bed to be irradiated is not trivial. The automatic image registration could significantly aid the tumor bed localization and lower the radiation dose delivered to the su...

Full description

Bibliographic Details
Main Authors: Marek Wodzinski, Izabela Ciepiela, Tomasz Kuszewski, Piotr Kedzierawski, Andrzej Skalski
Format: Article
Language:English
Published: MDPI AG 2021-06-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/21/12/4085
id doaj-7060f9e977ad44fc89158518499e689a
record_format Article
spelling doaj-7060f9e977ad44fc89158518499e689a2021-07-01T00:07:21ZengMDPI AGSensors1424-82202021-06-01214085408510.3390/s21124085Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed LocalizationMarek Wodzinski0Izabela Ciepiela1Tomasz Kuszewski2Piotr Kedzierawski3Andrzej Skalski4Department of Measurement and Electronics, AGH University of Science and Technology, PL30059 Kraków, PolandDepartment of Radiotherapy, The Holycross Cancer Center, PL25734 Kielce, PolandDepartment of Medical Physics, The Holycross Cancer Center, PL25734 Kielce, PolandDepartment of Radiotherapy, The Holycross Cancer Center, PL25734 Kielce, PolandDepartment of Measurement and Electronics, AGH University of Science and Technology, PL30059 Kraków, PolandBreast-conserving surgery requires supportive radiotherapy to prevent cancer recurrence. However, the task of localizing the tumor bed to be irradiated is not trivial. The automatic image registration could significantly aid the tumor bed localization and lower the radiation dose delivered to the surrounding healthy tissues. This study proposes a novel image registration method dedicated to breast tumor bed localization addressing the problem of missing data due to tumor resection that may be applied to real-time radiotherapy planning. We propose a deep learning-based nonrigid image registration method based on a modified U-Net architecture. The algorithm works simultaneously on several image resolutions to handle large deformations. Moreover, we propose a dedicated volume penalty that introduces the medical knowledge about tumor resection into the registration process. The proposed method may be useful for improving real-time radiation therapy planning after the tumor resection and, thus, lower the surrounding healthy tissues’ irradiation. The data used in this study consist of 30 computed tomography scans acquired in patients with diagnosed breast cancer, before and after tumor surgery. The method is evaluated using the target registration error between manually annotated landmarks, the ratio of tumor volume, and the subjective visual assessment. We compare the proposed method to several other approaches and show that both the multilevel approach and the volume regularization improve the registration results. The mean target registration error is below 6.5 mm, and the relative volume ratio is close to zero. The registration time below 1 s enables the real-time processing. These results show improvements compared to the classical, iterative methods or other learning-based approaches that do not introduce the knowledge about tumor resection into the registration process. In future research, we plan to propose a method dedicated to automatic localization of missing regions that may be used to automatically segment tumors in the source image and scars in the target image.https://www.mdpi.com/1424-8220/21/12/4085deep learningimage registrationmissing dataradiotherapybreast-conserving surgery
collection DOAJ
language English
format Article
sources DOAJ
author Marek Wodzinski
Izabela Ciepiela
Tomasz Kuszewski
Piotr Kedzierawski
Andrzej Skalski
spellingShingle Marek Wodzinski
Izabela Ciepiela
Tomasz Kuszewski
Piotr Kedzierawski
Andrzej Skalski
Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
Sensors
deep learning
image registration
missing data
radiotherapy
breast-conserving surgery
author_facet Marek Wodzinski
Izabela Ciepiela
Tomasz Kuszewski
Piotr Kedzierawski
Andrzej Skalski
author_sort Marek Wodzinski
title Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
title_short Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
title_full Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
title_fullStr Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
title_full_unstemmed Semi-Supervised Deep Learning-Based Image Registration Method with Volume Penalty for Real-Time Breast Tumor Bed Localization
title_sort semi-supervised deep learning-based image registration method with volume penalty for real-time breast tumor bed localization
publisher MDPI AG
series Sensors
issn 1424-8220
publishDate 2021-06-01
description Breast-conserving surgery requires supportive radiotherapy to prevent cancer recurrence. However, the task of localizing the tumor bed to be irradiated is not trivial. The automatic image registration could significantly aid the tumor bed localization and lower the radiation dose delivered to the surrounding healthy tissues. This study proposes a novel image registration method dedicated to breast tumor bed localization addressing the problem of missing data due to tumor resection that may be applied to real-time radiotherapy planning. We propose a deep learning-based nonrigid image registration method based on a modified U-Net architecture. The algorithm works simultaneously on several image resolutions to handle large deformations. Moreover, we propose a dedicated volume penalty that introduces the medical knowledge about tumor resection into the registration process. The proposed method may be useful for improving real-time radiation therapy planning after the tumor resection and, thus, lower the surrounding healthy tissues’ irradiation. The data used in this study consist of 30 computed tomography scans acquired in patients with diagnosed breast cancer, before and after tumor surgery. The method is evaluated using the target registration error between manually annotated landmarks, the ratio of tumor volume, and the subjective visual assessment. We compare the proposed method to several other approaches and show that both the multilevel approach and the volume regularization improve the registration results. The mean target registration error is below 6.5 mm, and the relative volume ratio is close to zero. The registration time below 1 s enables the real-time processing. These results show improvements compared to the classical, iterative methods or other learning-based approaches that do not introduce the knowledge about tumor resection into the registration process. In future research, we plan to propose a method dedicated to automatic localization of missing regions that may be used to automatically segment tumors in the source image and scars in the target image.
topic deep learning
image registration
missing data
radiotherapy
breast-conserving surgery
url https://www.mdpi.com/1424-8220/21/12/4085
work_keys_str_mv AT marekwodzinski semisuperviseddeeplearningbasedimageregistrationmethodwithvolumepenaltyforrealtimebreasttumorbedlocalization
AT izabelaciepiela semisuperviseddeeplearningbasedimageregistrationmethodwithvolumepenaltyforrealtimebreasttumorbedlocalization
AT tomaszkuszewski semisuperviseddeeplearningbasedimageregistrationmethodwithvolumepenaltyforrealtimebreasttumorbedlocalization
AT piotrkedzierawski semisuperviseddeeplearningbasedimageregistrationmethodwithvolumepenaltyforrealtimebreasttumorbedlocalization
AT andrzejskalski semisuperviseddeeplearningbasedimageregistrationmethodwithvolumepenaltyforrealtimebreasttumorbedlocalization
_version_ 1721349459363758080