Recalibration of deep learning models for abnormality detection in smartphone-captured chest radiograph
Open Access
- 15 February 2021
- journal article
- research article
- Published by Springer Science and Business Media LLC in npj Digital Medicine
- Vol. 4 (1), 1-10
- https://doi.org/10.1038/s41746-021-00393-9
Abstract
Image-based teleconsultation using smartphones has become increasingly popular. In parallel, deep learning algorithms have been developed to detect radiological findings in chest X-rays (CXRs). However, the feasibility of using smartphones to automate this process has yet to be evaluated. This study developed a recalibration method to build deep learning models to detect radiological findings on CXR photographs. Two publicly available databases (MIMIC-CXR and CheXpert) were used to build the models, and four derivative datasets containing 6453 CXR photographs were collected to evaluate model performance. After recalibration, the model achieved areas under the receiver operating characteristic curve of 0.80 (95% confidence interval: 0.78–0.82), 0.88 (0.86–0.90), 0.81 (0.79–0.84), 0.79 (0.77–0.81), 0.84 (0.80–0.88), and 0.90 (0.88–0.92), respectively, for detecting cardiomegaly, edema, consolidation, atelectasis, pneumothorax, and pleural effusion. The recalibration strategy, respectively, recovered 84.9%, 83.5%, 53.2%, 57.8%, 69.9%, and 83.0% of performance losses of the uncalibrated model. We conclude that the recalibration method can transfer models from digital CXRs to CXR photographs, which is expected to help physicians’ clinical works.Keywords
Funding Information
- U.S. Department of Health & Human Services | National Institutes of Health (R01 grant EB017205, R01 grant EB017205)
- Ministry of Science and Technology, Taiwan
- Fulbright Association
- U.S. Department of Health & Human Services | National Institutes of Health
This publication has 32 references indexed in Scilit:
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based LocalizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural NetworksRadiology, 2017
- Densely Connected Convolutional NetworksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Spatiotemporal Pyramid Network for Video Action RecognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2017
- Image-based teleconsultation using smartphones or tablets: qualitative assessment of medical expertsEmergency Medicine Journal, 2016
- The U.S. Radiologist Workforce: An Analysis of Temporal and Geographic Variation by Using Large National DatasetsRadiology, 2016
- scikit-image: image processing in PythonPeerJ, 2014
- Simulation and measurement of moiré patterns at finite distance.Optics Express, 2012
- Complex Wavelet Structural Similarity: A New Image Similarity IndexIEEE Transactions on Image Processing, 2009
- Image Noise ModelsPublished by Elsevier BV ,2009