Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation
Multiparametric magnetic resonance imaging (mpMRI) has emerged as a valuable diagnostic tool in prostate lesion assessment. However, training convolutional neural networks (CNNs) inevitably involves magnetic resonance (MR) images from multiple cohorts. There always exists variation in scanning proto...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-09-01
|
| Series: | Advanced Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/aisy.202200246 |
| _version_ | 1797676497844043776 |
|---|---|
| author | Jing Dai Xiaomei Wang Yingqi Li Zhiyu Liu Yui-Lun Ng Jiaren Xiao Joe King Man Fan James Lam Qi Dou Varut Vardhanabhuti Ka-Wai Kwok |
| author_facet | Jing Dai Xiaomei Wang Yingqi Li Zhiyu Liu Yui-Lun Ng Jiaren Xiao Joe King Man Fan James Lam Qi Dou Varut Vardhanabhuti Ka-Wai Kwok |
| author_sort | Jing Dai |
| collection | DOAJ |
| description | Multiparametric magnetic resonance imaging (mpMRI) has emerged as a valuable diagnostic tool in prostate lesion assessment. However, training convolutional neural networks (CNNs) inevitably involves magnetic resonance (MR) images from multiple cohorts. There always exists variation in scanning protocol among cohorts, inducing significant changes in data distribution between source and target domains. This challenge has greatly limited clinical adoption on a large scale. Herein, a coarse mask‐guided deep domain adaptation network (CMD2A‐Net) is proposed to develop a fully automated framework for prostate lesion detection and classification (PLDC). No category or mask label is required from the target domain. A coarse segmentation module is trained to cover the possible lesion‐related regions, so that attention maps can be generated to dedicate the local feature extraction of lesions within those regions. Experiments are performed on 512 mpMRI sets from datasets of PROSTATEx (330 sets) and two cohorts, A (74 sets) and B (108 sets). Using ensemble learning, CMD2A‐Net accomplishes an AUC of 0.921 in cohort A and 0.913 in cohort B, demonstrating its transferability from a large‐scale public dataset PROSTATEx to small‐scale target domains. Results from an ablation study also support its effectiveness in classification between benign and malignant lesions, compared to the state‐of‐the‐art models. An interactive preprint version of the article can be found here: https://doi.org/10.22541/au.166081031.11420810/v1. |
| first_indexed | 2024-03-11T22:30:04Z |
| format | Article |
| id | doaj.art-983e13490619470db7d7ca6e308e259f |
| institution | Directory Open Access Journal |
| issn | 2640-4567 |
| language | English |
| last_indexed | 2024-03-11T22:30:04Z |
| publishDate | 2023-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advanced Intelligent Systems |
| spelling | doaj.art-983e13490619470db7d7ca6e308e259f2023-09-23T07:09:23ZengWileyAdvanced Intelligent Systems2640-45672023-09-0159n/an/a10.1002/aisy.202200246Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain AdaptationJing Dai0Xiaomei Wang1Yingqi Li2Zhiyu Liu3Yui-Lun Ng4Jiaren Xiao5Joe King Man Fan6James Lam7Qi Dou8Varut Vardhanabhuti9Ka-Wai Kwok10Department of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Surgery The University of Hong Kong-Shenzhen Hospital Shenzhen Guangdong 518000 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaDepartment of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong 999077 ChinaDepartment of Diagnostic Radiology The University of Hong Kong Hong Kong 999077 ChinaDepartment of Mechanical Engineering The University of Hong Kong Hong Kong 999077 ChinaMultiparametric magnetic resonance imaging (mpMRI) has emerged as a valuable diagnostic tool in prostate lesion assessment. However, training convolutional neural networks (CNNs) inevitably involves magnetic resonance (MR) images from multiple cohorts. There always exists variation in scanning protocol among cohorts, inducing significant changes in data distribution between source and target domains. This challenge has greatly limited clinical adoption on a large scale. Herein, a coarse mask‐guided deep domain adaptation network (CMD2A‐Net) is proposed to develop a fully automated framework for prostate lesion detection and classification (PLDC). No category or mask label is required from the target domain. A coarse segmentation module is trained to cover the possible lesion‐related regions, so that attention maps can be generated to dedicate the local feature extraction of lesions within those regions. Experiments are performed on 512 mpMRI sets from datasets of PROSTATEx (330 sets) and two cohorts, A (74 sets) and B (108 sets). Using ensemble learning, CMD2A‐Net accomplishes an AUC of 0.921 in cohort A and 0.913 in cohort B, demonstrating its transferability from a large‐scale public dataset PROSTATEx to small‐scale target domains. Results from an ablation study also support its effectiveness in classification between benign and malignant lesions, compared to the state‐of‐the‐art models. An interactive preprint version of the article can be found here: https://doi.org/10.22541/au.166081031.11420810/v1.https://doi.org/10.1002/aisy.202200246convolutional neural networksdomain adaptationmultiparametric magnetic resonance imaging (mpMRI)prostate lesion detection and classification |
| spellingShingle | Jing Dai Xiaomei Wang Yingqi Li Zhiyu Liu Yui-Lun Ng Jiaren Xiao Joe King Man Fan James Lam Qi Dou Varut Vardhanabhuti Ka-Wai Kwok Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation Advanced Intelligent Systems convolutional neural networks domain adaptation multiparametric magnetic resonance imaging (mpMRI) prostate lesion detection and classification |
| title | Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation |
| title_full | Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation |
| title_fullStr | Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation |
| title_full_unstemmed | Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation |
| title_short | Automatic Multiparametric Magnetic Resonance Imaging‐Based Prostate Lesions Assessment with Unsupervised Domain Adaptation |
| title_sort | automatic multiparametric magnetic resonance imaging based prostate lesions assessment with unsupervised domain adaptation |
| topic | convolutional neural networks domain adaptation multiparametric magnetic resonance imaging (mpMRI) prostate lesion detection and classification |
| url | https://doi.org/10.1002/aisy.202200246 |
| work_keys_str_mv | AT jingdai automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT xiaomeiwang automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT yingqili automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT zhiyuliu automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT yuilunng automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT jiarenxiao automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT joekingmanfan automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT jameslam automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT qidou automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT varutvardhanabhuti automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation AT kawaikwok automaticmultiparametricmagneticresonanceimagingbasedprostatelesionsassessmentwithunsuperviseddomainadaptation |