Lung Cancer Screening Using Adaptive Memory-Augmented Recurrent Networks

Abstract

In this paper, we investigate the effectiveness of deep learning techniques for lung nodule classification in computed tomography scans. Using less than 10,000 training examples, our deep networks perform two times better than a standard radiology software. Visualization of the networks' neurons reveals semantically meaningful features that are consistent with the clinical knowledge and radiologists' perception. Our paper also proposes a novel framework for rapidly adapting deep networks to the radiologists' feedback, or change in the data due to the shift in sensor's resolution or patient population. The classification accuracy of our approach remains above 80% while popular deep networks' accuracy is around chance. Finally, we provide in-depth analysis of our framework by asking a radiologist to examine important networks' features and perform blind re-labeling of networks' mistakes.

References in corpus (6)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Matching Networks for One Shot Learning
• Object Detectors Emerge in Deep Scene CNNs
• Training Convolutional Networks with Noisy Labels
• One-shot Learning with Memory-Augmented Neural Networks
• Benign-Malignant Lung Nodule Classification with Geometric and Appearance Histogram Features

Cited by in corpus (4)

• Small Sample Learning in Big Data Era
• Feature Engineering and Forecasting via Derivative-free Optimization and Ensemble of Sequence-to-sequence Networks with Applications in Renewable Energy
• Text-Independent Speaker Verification Using Long Short-Term Memory Networks
• DVNet: A Memory-Efficient Three-Dimensional CNN for Large-Scale Neurovascular Reconstruction