WindowNet: Learnable Windows for Chest X-ray Classification

Chest X-ray (CXR) images are commonly compressed to a lower resolution and bit depth to reduce their size, potentially altering subtle diagnostic features. Radiologists use windowing operations to enhance image contrast, but the impact of such operations on CXR classification performance is unclear. In this study, we show that windowing can improve CXR classification performance, and propose WindowNet, a model that learns optimal window settings. We first investigate the impact of bit-depth on classification performance and find that a higher bit-depth (12-bit) leads to improved performance. We then evaluate different windowing settings and show that training with a distinct window generally improves pathology-wise classification performance. Finally, we propose and evaluate WindowNet, a model that learns optimal window settings, and show that it significantly improves performance compared to the baseline model without windowing.

[1]  B. Sabel,et al.  Attention-based Saliency Maps Improve Interpretability of Pneumothorax Classification , 2023, Radiology. Artificial intelligence.

[2]  Manohar Karki,et al.  CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings , 2020, Artif. Intell. Medicine.

[3]  Kihwan Choi,et al.  Trainable Multi-contrast Windowing for Liver CT Segmentation , 2020, 2020 IEEE International Conference on Big Data and Smart Computing (BigComp).

[4]  Shunxing Bao,et al.  Stochastic tissue window normalization of deep learning on computed tomography , 2019, Journal of medical imaging.

[5]  Hyunkwang Lee,et al.  Practical Window Setting Optimization for Medical Image Deep Learning , 2018, ArXiv.

[6]  Frank Hutter,et al.  Decoupled Weight Decay Regularization , 2017, ICLR.

[7]  Ronald M. Summers,et al.  ChestX-ray: Hospital-Scale Chest X-ray Database and Benchmarks on Weakly Supervised Classification and Localization of Common Thorax Diseases , 2019, Deep Learning and Convolutional Neural Networks for Medical Imaging and Clinical Informatics.

[8]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Tom Kimpe,et al.  Increasing the Number of Gray Shades in Medical Display Systems—How Much is Enough? , 2007, Journal of Digital Imaging.