TY - GEN
T1 - AdaBoosted Deep Ensembles
T2 - 11th International Workshop on Machine Learning in Medical Imaging, MLMI 2020, held in conjunction with the 23rd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2020
AU - Reza, Syed M.S.
AU - Butman, John A.
AU - Park, Deric M.
AU - Pham, Dzung L.
AU - Roy, Snehashis
N1 - Publisher Copyright: © 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Even though state-of-the-art convolutional neural networks (CNNs) have shown outstanding performance in a wide range of imaging applications, they typically require large amounts of high-quality training data to prevent over fitting. In the case of medical image segmentation, it is often difficult to gain access to large data sets, particularly those involving rare diseases, such as skull-based chordoma tumors. This challenge is exacerbated by the difficulty in performing manual delineations, which are time-consuming and can have inconsistent quality. In this work, we propose a deep ensemble method that learns multiple models, trained using a leave-one-out strategy, and then aggregates the outputs for test data through a boosting strategy. The proposed method was evaluated for chordoma tumor segmentation in head magnetic resonance images using three well-known CNN architectures; VNET, UNET, and Feature pyramid network (FPN). Significantly improved Dice scores (up to 27%) were obtained using the proposed ensemble method when compared to a single model trained with all available training subjects. The proposed ensemble method can be applied to any neural network based segmentation method to potentially improve generalizability when learning from a small sized dataset.
AB - Even though state-of-the-art convolutional neural networks (CNNs) have shown outstanding performance in a wide range of imaging applications, they typically require large amounts of high-quality training data to prevent over fitting. In the case of medical image segmentation, it is often difficult to gain access to large data sets, particularly those involving rare diseases, such as skull-based chordoma tumors. This challenge is exacerbated by the difficulty in performing manual delineations, which are time-consuming and can have inconsistent quality. In this work, we propose a deep ensemble method that learns multiple models, trained using a leave-one-out strategy, and then aggregates the outputs for test data through a boosting strategy. The proposed method was evaluated for chordoma tumor segmentation in head magnetic resonance images using three well-known CNN architectures; VNET, UNET, and Feature pyramid network (FPN). Significantly improved Dice scores (up to 27%) were obtained using the proposed ensemble method when compared to a single model trained with all available training subjects. The proposed ensemble method can be applied to any neural network based segmentation method to potentially improve generalizability when learning from a small sized dataset.
UR - http://www.scopus.com/inward/record.url?scp=85092712667&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-59861-7_58
DO - 10.1007/978-3-030-59861-7_58
M3 - Conference contribution
SN - 9783030598600
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 572
EP - 582
BT - Machine Learning in Medical Imaging - 11th International Workshop, MLMI 2020, Held in Conjunction with MICCAI 2020, Proceedings
A2 - Liu, Mingxia
A2 - Lian, Chunfeng
A2 - Yan, Pingkun
A2 - Cao, Xiaohuan
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 4 October 2020 through 4 October 2020
ER -