Journal
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT II
Volume 11071, Issue -, Pages 693-701Publisher
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-030-00934-2_77
Keywords
Deep learning; Liver lesions; Detection; Multi-phase CT
Categories
Funding
- National Research Foundation of Korea (NRF) - Korea government (Ministry of Science and ICT) [2018R1A2B3001628]
- Interdisciplinary Research Initiatives Program from College of Engineering and College of Medicine, Seoul National University [800-20170166]
- Samsung Research Funding Center of Samsung Electronics [SRFC-IT1601-05]
- Creative Industrial Technology Development Program - Ministry of Trade, Industry & Energy (MOTIE, Korea) [10053249]
- Brain Korea 21 Plus Project in 2018
Ask authors/readers for more resources
We present a focal liver lesion detection model leveraged by custom-designed multi-phase computed tomography (CT) volumes, which reflects real-world clinical lesion detection practice using a Single Shot MultiBox Detector (SSD). We show that grouped convolutions effectively harness richer information of the multi-phase data for the object detection model, while a naive application of SSD suffers from a generalization gap. We trained and evaluated the modified SSD model and recently proposed variants with our CT dataset of 64 subjects by five-fold cross validation. Our model achieved a 53.3% average precision score and ran in under three seconds per volume, outperforming the original model and state-of-the-art variants. Results show that the one-stage object detection model is a practical solution, which runs in near real-time and can learn an unbiased feature representation from a large-volume real-world detection dataset, which requires less tedious and time consuming construction of the weak phase-level bounding box labels.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available