The Effect of Image Resolution on Deep Learning in Radiography

Abstract

To examine variations of convolutional neural network (CNN) performance for multiple chest radiograph diagnoses and image resolutions. Materials and Methods: This retrospective study examined CNN performance using the publicly available National Institutes of Health chest radiograph dataset comprising 112 120 chest radiographic images from 30 805 patients. The network architectures examined included ResNet34 and DenseNet121. Image resolutions ranging from 32 3 32 to 600 3 600 pixels were investigated. Network training paradigms used 80% of samples for training and 20% for validation. CNN performance was evaluated based on area under the receiver operating characteristic curve (AUC) and label accuracy. Binary output networks were trained separately for each label or diagnosis under consideration. Results: Maximum AUCs were achieved at image resolutions between 256 3 256 and 448 3 448 pixels for binary decision networks targeting emphysema, cardiomegaly, hernias, edema, effusions, atelectasis, masses, and nodules. When comparing performance between networks that utilize lower resolution (64 3 64 pixels) versus higher (320 3 320 pixels) resolution inputs, emphysema, cardiomegaly, hernia, and pulmonary nodule detection had the highest fractional improvements in AUC at higher image resolutions. Specifically, pulmonary nodule detection had an AUC performance ratio of 80.7% 6 1.5 (standard deviation) (0.689 of 0.854) whereas thoracic mass detection had an AUC ratio of 86.7% 6 1.2 (0.767 of 0.886) for these image resolutions. Conclusion: Increasing image resolution for CNN training often has a trade-off with the maximum possible batch size, yet optimal selection of image resolution has the potential for further increasing neural network performance for various radiology-based machine learning tasks. Furthermore, identifying diagnosis-specific tasks that require relatively higher image resolution can potentially provide insight into the relative difficulty of identifying different radiology findings.

Comparing different deep learning architectures for classification of chest radiographs

“…Implementing a human-in-the loop approach for partially correcting the label noise could further improve performance of networks trained on the CheXpert dataset 21 . Our findings differ from applied techniques used in previous literature, where deeper network architectures, mainly a DenseNet-121, were used to classify the CheXpert data set 6,9,22 . The authors of the CheXpert dataset achieved an average overall AUROC of 0.889 3 , using a DenseNet-121, which was not surpassed by any of the models used in our analysis, although differences between the best performing networks and the CheXpert baseline were smaller than 0.01.…”

Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI Extension

“…Explanation: The measured performance of any AI system may be critically dependent on the nature and quality of the input data 48. A description of the input data handling, including acquisition, selection, and pre-processing prior to analysis by the AI system should be provided.…”

Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension

“…The measured performance of any AI system may be critically dependent on the nature and quality of the input data. 48 A description of the input-data handling, including acquisition, selection, and pre-processing before analysis by the AI system, should be provided. Completeness and transparency of this description are integral to the replicability of the intervention beyond the clinical trial in real-world settings.…”