Loading...
Please wait, while we are loading the content...
Similar Documents
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
| Content Provider | IEEE Xplore Digital Library |
|---|---|
| Author | Z. Xu L. Lu J. Yao H. R. Roth M. Gao R. M. Summers H. Shin D. Mollura I. Nogues |
| Copyright Year | 2016 |
| Abstract | Remarkable progress has been made in image recognition, primarily due to the availability of large-scale annotated datasets and deep convolutional neural networks (CNNs). CNNs enable learning data-driven, highly representative, hierarchical image features from sufficient training data. However, obtaining datasets as comprehensively annotated as ImageNet in the medical imaging domain remains a challenge. There are currently three major techniques that successfully employ CNNs to medical image classification: training the CNN from scratch, using off-the-shelf pre-trained CNN features, and conducting unsupervised CNN pre-training with supervised fine-tuning. Another effective method is transfer learning, i.e., fine-tuning CNN models pre-trained from natural image dataset to medical image tasks. In this paper, we exploit three important, but previously understudied factors of employing deep convolutional neural networks to computer-aided detection problems. We first explore and evaluate different CNN architectures. The studied models contain 5 thousand to 160 million parameters, and vary in numbers of layers. We then evaluate the influence of dataset scale and spatial image context on performance. Finally, we examine when and why transfer learning from pre-trained ImageNet (via fine-tuning) can be useful. We study two specific computer-aided detection (CADe) problems, namely thoraco-abdominal lymph node (LN) detection and interstitial lung disease (ILD) classification. We achieve the state-of-the-art performance on the mediastinal LN detection, and report the first five-fold cross-validation classification results on predicting axial CT slices with ILD categories. Our extensive empirical evaluation, CNN model analysis and valuable insights can be extended to the design of high performance CAD systems for other medical imaging tasks. |
| Starting Page | 1285 |
| Ending Page | 1298 |
| Page Count | 14 |
| File Format | HTM / HTML |
| ISSN | 1558254X |
| Journal | IEEE Transactions on Medical Imaging |
| DOI | 10.1109/TMI.2016.2528162 |
| Volume Number | 35 |
| Issue Number | 5 |
| Language | English |
| Access Restriction | Open |
| Subject Keyword | Medical image processing Solid modeling Medical imaging domain Image recognition Five-fold cross-validation classification Transfer learning Lung Highly representative hierarchical image features Fine-tuning CNN models Off-the-shelf pretrained CNN features Lymph nodes Dataset characteristics Computed tomography Spatial image context CNN model analysis Medical image classification Image representation CNN architectures Computer aided diagnosis Mediastinal LN detection Unsupervised CNN pretraining High performance CAD systems Thoraco-abdominal lymph node detection Pretrained imagenet Natural image dataset Biomedical imaging Deep convolutional neural networks Learning data-driven Computational modeling Interstitial lung disease classification Computerised tomography Computer-aided detection problems State-of-the-art performance Supervised fine-tuning Diseases Neurophysiology Medical image tasks Image analysis Reviews Lungs Neural networks Machine learning Learning (artificial intelligence) Axial CT slices Computer-aided detection Image classification |
| Content Type | Text |
| Resource Type | Preprint Article |
| Subject | Electrical and Electronic Engineering Computer Science Applications Radiological and Ultrasound Technology Software |
