Cascaded Neural Networks with Selective Classifiers and its evaluation using Lung X-ray CT Images
This addresses the class imbalance problem in medical imaging for lung cancer diagnosis, representing an incremental improvement.
The paper tackled lung nodule detection in class-imbalanced X-ray CT images by proposing cascaded convolutional neural networks with selective classifiers to filter out non-nodules, achieving detection sensitivities of 85.3% and 90.7% at 1 and 4 false positives per scan, respectively.
Lung nodule detection is a class imbalanced problem because nodules are found with much lower frequency than non-nodules. In the class imbalanced problem, conventional classifiers tend to be overwhelmed by the majority class and ignore the minority class. We therefore propose cascaded convolutional neural networks to cope with the class imbalanced problem. In the proposed approach, cascaded convolutional neural networks that perform as selective classifiers filter out obvious non-nodules. Successively, a convolutional neural network trained with a balanced data set calculates nodule probabilities. The proposed method achieved the detection sensitivity of 85.3% and 90.7% at 1 and 4 false positives per scan in FROC curve, respectively.