Lung Cancer Screening Using Adaptive Memory-Augmented Recurrent Networks
This addresses lung cancer screening for medical diagnosis, with incremental improvements in adaptation to feedback and data shifts.
The paper tackled lung nodule classification in CT scans using deep learning, achieving twice the performance of standard radiology software with under 10,000 training examples and maintaining over 80% accuracy where other networks failed.
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.