Tips and Tricks to Improve CNN-based Chest X-ray Diagnosis: A Survey
arXiv:2106.00997
Abstract
Convolutional Neural Networks (CNNs) intrinsically requires large-scale data whereas Chest X-Ray (CXR) images tend to be data/annotation-scarce, leading to over-fitting. Therefore, based on our development experience and related work, this paper thoroughly introduces tricks to improve generalization in the CXR diagnosis: how to (i) leverage additional data, (ii) augment/distillate data, (iii) regularize training, and (iv) conduct efficient segmentation. As a development example based on such optimization techniques, we also feature LPIXEL's CNN-based CXR solution, EIRL Chest Nodule, which improved radiologists/non-radiologists' nodule detection sensitivity by 0.100/0.131, respectively, while maintaining specificity.
arXiv comment7 pages, 2 figures, to be published in Japanese Journal of Medical Imaging and Information Sciences
References in corpus (6)
- SMOTE: Synthetic Minority Over-sampling Technique
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Deep Learning for Chest X-ray Analysis: A Survey
- Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation
- Interactive segmentation of medical images through fully convolutional neural networks
- MoCo-CXR: MoCo Pretraining Improves Representation and Transferability of Chest X-ray Models