TY - GEN
T1 - Deep convolutional activation features for large scale Brain Tumor histopathology image classification and segmentation
AU - Xu, Yan
AU - Jia, Zhipeng
AU - Ai, Yuqing
AU - Zhang, Fang
AU - Lai, Maode
AU - Chang, Eric I.Chao
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/8/4
Y1 - 2015/8/4
N2 - We propose a simple, efficient and effective method using deep convolutional activation features (CNNs) to achieve stat-of-the-art classification and segmentation for the MICCAI 2014 Brain Tumor Digital Pathology Challenge. Common traits of such medical image challenges are characterized by large image dimensions (up to the gigabyte size of an image), a limited amount of training data, and significant clinical feature representations. To tackle these challenges, we transfer the features extracted from CNNs trained with a very large general image database to the medical image challenge. In this paper, we used CNN activations trained by ImageNet to extract features (4096 neurons, 13.3% active). In addition, feature selection, feature pooling, and data augmentation are used in our work. Our system obtained 97.5% accuracy on classification and 84% accuracy on segmentation, demonstrating a significant performance gain over other participating teams.
AB - We propose a simple, efficient and effective method using deep convolutional activation features (CNNs) to achieve stat-of-the-art classification and segmentation for the MICCAI 2014 Brain Tumor Digital Pathology Challenge. Common traits of such medical image challenges are characterized by large image dimensions (up to the gigabyte size of an image), a limited amount of training data, and significant clinical feature representations. To tackle these challenges, we transfer the features extracted from CNNs trained with a very large general image database to the medical image challenge. In this paper, we used CNN activations trained by ImageNet to extract features (4096 neurons, 13.3% active). In addition, feature selection, feature pooling, and data augmentation are used in our work. Our system obtained 97.5% accuracy on classification and 84% accuracy on segmentation, demonstrating a significant performance gain over other participating teams.
KW - classification
KW - deep convolutional activation features
KW - deep learning
KW - feature learning
KW - segmentation
UR - https://www.scopus.com/pages/publications/84946045951
U2 - 10.1109/ICASSP.2015.7178109
DO - 10.1109/ICASSP.2015.7178109
M3 - 会议稿件
AN - SCOPUS:84946045951
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 947
EP - 951
BT - 2015 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 40th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015
Y2 - 19 April 2014 through 24 April 2014
ER -