Tumor-aware, adversarial domain adaptation from CT to MRI for lung cancer segmentation Conference Paper
| Authors: | Jiang, J.; Hu, Y. C.; Tyagi, N.; Zhang, P.; Rimner, A.; Mageras, G. S.; Deasy, J. O.; Veeraraghavan, H. |
| Title: | Tumor-aware, adversarial domain adaptation from CT to MRI for lung cancer segmentation |
| Conference Title: | 21st International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) |
| Abstract: | We present an adversarial domain adaptation based deep learning approach for automatic tumor segmentation from T2-weighted MRI. Our approach is composed of two steps: (i) a tumor-aware unsupervised cross-domain adaptation (CT to MRI), followed by (ii) semi-supervised tumor segmentation using Unet trained with synthesized and limited number of original MRIs. We introduced a novel target specific loss, called tumor-aware loss, for unsupervised cross-domain adaptation that helps to preserve tumors on synthesized MRIs produced from CT images. In comparison, state-of-the art adversarial networks trained without our tumor-aware loss produced MRIs with ill-preserved or missing tumors. All networks were trained using labeled CT images from 377 patients with non-small cell lung cancer obtained from the Cancer Imaging Archive and unlabeled T2w MRIs from a completely unrelated cohort of 6 patients with pre-treatment and 36 on-treatment scans. Next, we combined 6 labeled pre-treatment MRI scans with the synthesized MRIs to boost tumor segmentation accuracy through semi-supervised learning. Semi-supervised training of cycle-GAN produced a segmentation accuracy of 0.66 computed using Dice Score Coefficient (DSC). Our method trained with only synthesized MRIs produced an accuracy of 0.74 while the same method trained in semi-supervised setting produced the best accuracy of 0.80 on test. Our results show that tumor-aware adversarial domain adaptation helps to achieve reasonably accurate cancer segmentation from limited MRI data by leveraging large CT datasets. © Springer Nature Switzerland AG 2018. |
| Keywords: | computerized tomography; medical imaging; tumors; patient treatment; medical computing; biological organs; diseases; non small cell lung cancer; tumor segmentation; deep learning; supervised learning; segmentation accuracy; adversarial networks; domain adaptation; learning approach; semi- supervised learning; semi-supervised trainings |
| Journal Title | Lecture Notes in Computer Science |
| Volume: | 11071 |
| Conference Dates: | 2018 Sep 16-20 |
| Conference Location: | Granada, Spain |
| ISBN: | 0302-9743 |
| Publisher: | Springer |
| Location: | Cham, Switzerland |
| Date Published: | 2018-01-01 |
| Start Page: | 777 |
| End Page: | 785 |
| Language: | English |
| DOI: | 10.1007/978-3-030-00934-2_86 |
| PROVIDER: | scopus |
| PMCID: | PMC6169798 |
| PUBMED: | 30294726 |
| DOI/URL: | |
| Notes: | Conference Paper -- Export Date: 1 November 2018 -- Source: Scopus |
Altmetric
Citation Impact
BMJ Impact Analytics
Related MSK Work