TY - GEN
T1 - Prostate cancer detection and segmentation in multi-parametric mri via cnn and conditional random field
AU - Cao, Ruiming
AU - Zhong, Xinran
AU - Shakeri, Sepideh
AU - Bajgiran, Amirhossein Mohammadian
AU - Mirak, Sohrab Afshari
AU - Enzmann, Dieter
AU - Raman, Steven S.
AU - Sung, Kyunghyun
N1 - Funding Information:
This study was supported by the Integrated Diagnostics (IDX) Program, Department of Radiological Sciences & Pathology David Geffen School of Medicine at UCLA.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - Multi-parametric MRI (mp-MRI) is a powerful diagnostic tool for prostate cancer (PCa). However, interpreting prostate mp-MRI requires high-level expertise, causing significant inter-reader variations. Convolutional neural networks (CNNs) have recently shown great promise for various tasks. In this study, we propose an improved CNN to jointly detect PCa lesions and segment for accurate lesions contours. Specifically, we adapt focal loss to overcome the imbalance between cancerous and non-cancerous areas for improved lesion detection and design selective dense conditional random field (SD-CRF), a post-processing step to refine the CNN prediction into the lesion segmentation based on a specific imaging component of mp-MRI. We trained and validated the proposed CNN in 5-fold cross-validation using 397 preoperative mp-MRI exams with whole-mount histopathology-confirmed lesion annotations. In the free-response receiver operating characteristics (FROC) analysis, the proposed CNN achieved 75.1% lesion detection sensitivity at the cost of 1 false positive per patient. In the evaluation for lesion segmentation, the proposed CNN improved the Dice coefficient by 20.6% from the baseline CNN.
AB - Multi-parametric MRI (mp-MRI) is a powerful diagnostic tool for prostate cancer (PCa). However, interpreting prostate mp-MRI requires high-level expertise, causing significant inter-reader variations. Convolutional neural networks (CNNs) have recently shown great promise for various tasks. In this study, we propose an improved CNN to jointly detect PCa lesions and segment for accurate lesions contours. Specifically, we adapt focal loss to overcome the imbalance between cancerous and non-cancerous areas for improved lesion detection and design selective dense conditional random field (SD-CRF), a post-processing step to refine the CNN prediction into the lesion segmentation based on a specific imaging component of mp-MRI. We trained and validated the proposed CNN in 5-fold cross-validation using 397 preoperative mp-MRI exams with whole-mount histopathology-confirmed lesion annotations. In the free-response receiver operating characteristics (FROC) analysis, the proposed CNN achieved 75.1% lesion detection sensitivity at the cost of 1 false positive per patient. In the evaluation for lesion segmentation, the proposed CNN improved the Dice coefficient by 20.6% from the baseline CNN.
KW - Computer-aided detection and diagnosis
KW - Convolutional neural networks
KW - MRI
KW - Prostate cancer
UR - http://www.scopus.com/inward/record.url?scp=85073898500&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073898500&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2019.8759584
DO - 10.1109/ISBI.2019.8759584
M3 - Conference contribution
AN - SCOPUS:85073898500
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 1900
EP - 1904
BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
T2 - 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
Y2 - 8 April 2019 through 11 April 2019
ER -