The effect of Gaussian noise on pneumonia detection on chest radiographs,using convolutional neural networks |
| |
Institution: | 1. Department of Radiology and Nuclear Medicine, Hospital of South West Jutland, University Hospital of Southern Denmark, Esbjerg, Denmark;2. Department of Regional Health Research, Faculty of Health Sciences, University of Southern Denmark, Odense, Denmark;3. Imaging Research Initiative Southwest (IRIS), Hospital of South West Jutland, University Hospital of Southern Denmark, Esbjerg, Denmark |
| |
Abstract: | IntroductionChest X-rays (CXR) with under-exposure increase image noise and this may affect convolutional neural network (CNN) performance. This study aimed to train and validate CNNs for classifying pneumonia on CXR as normal or pneumonia acquired at different image noise levels.MethodsThe study used the curated and publicly available “Chest X-Ray Pneumonia” dataset of 5856 AP CXR classified into 1583 normal, 4273 viral and bacterial pneumonia cases. Gaussian noise with zero mean was added to the images, at 5 image noise variance levels, corresponding to decreasing exposure. Each noise-level dataset was split into 80% for training, 10% for validation, and 10% for test data and then classified using custom trained sequential CNN architecture. Six classification tasks were developed for five Gaussian noise levels and the original dataset. Sensitivity, specificity, predictive values and accuracy were used as evaluation performance metrics.ResultsCNN evaluation on the different datasets revealed no performance drop from the original dataset to the five datasets with different noise levels. Sensitivity, specificity and accuracy for the normal datasets were 98.7%, 76.1% and 90.2%. For the five Gaussian noise levels the sensitivity, specificity and accuracy ranged from 96.9% to 98.2%, 94.4%–98.7% and 96.8%–97.6%, respectively. A heat map was used for visual explanation of the CNNs.ConclusionThe CNNs sensitivity maintained, and the specificity increased in distinguishing between normal and pneumonia CXR with the introduction of image noise.Implications for practiceNo performance drops of CNNs in distinguishing cases with and without pneumonia CXR with different Gaussian noise levels was observed. This has potential for decreasing radiation dose to patients or maintaining exposure parameters for patients that require additional radiographs. |
| |
Keywords: | Digital radiography Gaussian noise Convolutional neural networks Pneumonia Deep learning AI"} {"#name":"keyword" "$":{"id":"kwrd0040"} "$$":[{"#name":"text" "_":"Artificial Intelligence CNN"} {"#name":"keyword" "$":{"id":"kwrd0050"} "$$":[{"#name":"text" "_":"Convolutional neural network CXR"} {"#name":"keyword" "$":{"id":"kwrd0060"} "$$":[{"#name":"text" "_":"Chest Radiographs DL"} {"#name":"keyword" "$":{"id":"kwrd0070"} "$$":[{"#name":"text" "_":"Deep learning DR"} {"#name":"keyword" "$":{"id":"kwrd0080"} "$$":[{"#name":"text" "_":"Digital radiography GAN"} {"#name":"keyword" "$":{"id":"kwrd0090"} "$$":[{"#name":"text" "_":"Generative Adversarial Network Grad-CAM++"} {"#name":"keyword" "$":{"id":"kwrd0100"} "$$":[{"#name":"text" "_":"Generalized Gradient-Based Visual Explanation for Deep Convolutional Networks ML"} {"#name":"keyword" "$":{"id":"kwrd0110"} "$$":[{"#name":"text" "_":"Machine learning PACS"} {"#name":"keyword" "$":{"id":"kwrd0120"} "$$":[{"#name":"text" "_":"Picture and archiving system |
本文献已被 ScienceDirect 等数据库收录! |
|