Chest Radiograph Interpretation with Deep Learning Models: Assessment with Radiologist-adjudicated Reference Standards and Population-adjusted Evaluation
The DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index.
Abstract
Background Deep learning has the potential to augment the use of chest radiography in clinical radiology, but challenges include poor generalizability, spectrum bias, and difficulty comparing across studies. Purpose To develop and evaluate deep learning models for chest radiograph interpretation by using radiologist-adjudicated reference standards. Materials and Methods Deep learning models were developed to detect four findings (pneumothorax, opacity, nodule or mass, and fracture) on frontal chest radiographs. This retrospective study used two data sets. Data set 1 (DS1) consisted of 759 611 images from a multicity hospital network and ChestX-ray14 is a publicly available data set with 112 120 images. Natural language processing and expert review of a subset of images provided labels for 657 954 training images. Test sets consisted of 1818 and 1962 images from DS1 and ChestX-ray14, respectively. Reference standards were defined by radiologist-adjudicated image review. Performance was evaluated by area under the receiver operating characteristic curve analysis, sensitivity, specificity, and positive predictive value. Four radiologists reviewed test set images for performance comparison. Inverse probability weighting was applied to DS1 to account for positive radiograph enrichment and estimate population-level performance. Results In DS1, population-adjusted areas under the receiver operating characteristic curve for pneumothorax, nodule or mass, airspace opacity, and fracture were, respectively, 0.95 (95% confidence interval [CI]: 0.91, 0.99), 0.72 (95% CI: 0.66, 0.77), 0.91 (95% CI: 0.88, 0.93), and 0.86 (95% CI: 0.79, 0.92). With ChestX-ray14, areas under the receiver operating characteristic curve were 0.94 (95% CI: 0.93, 0.96), 0.91 (95% CI: 0.89, 0.93), 0.94 (95% CI: 0.93, 0.95), and 0.81 (95% CI: 0.75, 0.86), respectively. Conclusion Expert-level models for detecting clinically relevant chest radiograph findings were developed for this study by using adjudicated reference standards and with population-level performance estimation. Radiologist-adjudicated labels for 2412 ChestX-ray14 validation set images and 1962 test set images are provided. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Chang in this issue.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Only eight studies in ophthalmology imaging14,21,32,33,43,55,108,109, ten studies in respiratory imaging64,66,70,72,75,79,82,87,89,110 and six studies in breast imaging62,91,97,104,106,111 mentioned adherence to the STARD-2015 guidelines or had a STARD flow diagram in the manuscript.”
From Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis (2021) · cites this paper as [82]
Citations
Cites
Fleischner Society: Glossary of Terms for Thoracic Imaging
Hansell, D. M., Bankier, A. A., MacMahon, H., McLoud, T. C., Müller, N. L., & Remy, J. (2008) · Radiology · doi:10.1148/radiol.2462070712 · Cited by 4,198 · cited 1×
“Clinical definitions for these categories were on the basis of the Fleischner Society Glos-sary of Terms for Thoracic Imaging, except for osseous fracture, which was defined as visible rib, clavicle, humeral, or vertebral body fractures (20).”
MethodArmato, S. G., McLennan, G., Bidaut, L., McNitt‐Gray, M. F., Meyer, C. R., Reeves, A. P., Zhao, B., Aberle, D. R., Henschke, C. I., Hoffman, E. A., et al. (2011) · Medical Physics · doi:10.1118/1.3528204 · Cited by 2,887
Revisiting Unreasonable Effectiveness of Data in Deep Learning Era
Sun, C., Shrivastava, A., Singh, S., & Gupta, A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.97 · Cited by 2,484 · cited 1×
“We used Xception (22) as the convolutional neural network architecture that was pretrained on 300 000 000 natural images (23).”
MethodRajpurkar, P., Irvin, J., Ball, R. L., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., Langlotz, C. P., et al. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002686 · Cited by 1,449 · cited 2×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundVery Deep Convolutional Networks for Text Classification
Conneau, A., Schwenk, H., Barrault, L., & Lecun, Y. (2017) · Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers · doi:10.18653/v1/e17-1104 · Cited by 1,046
Park, S. H., & Han, K. (2018) · Radiology · doi:10.1148/radiol.2017171920 · Cited by 894 · cited 1×
“However, because such data do not necessarily reflect real-world prevalence or diversity (33,34), such enrichment can also prevent meaningful clinical interpretation of diagnostic performance.”
BackgroundKrause, J., Gulshan, V., Rahimy, E., Karth, P., Widner, K., Corrado, G. S., Peng, L., & Webster, D. R. (2018) · Ophthalmology · doi:10.1016/j.ophtha.2018.01.034 · Cited by 564 · cited 1×
“…adjudicated labels, extensive work has been performed on the value of multiple interpretations in mammography (30) and other clinical settings (31,32). the most obviously positive and negative for pneumothorax, opacity, nodule or mass, or fracture for training, which can fail to address the…”
BackgroundSelf-citationMazurowski, M. A., Buda, M., Saha, A., & Bashir, M. R. (2019) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26534 · Cited by 554 · cited 1×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundNam, J. G., Park, S., Hwang, E. J., Lee, J. H., Jin, K.-N., Lim, K. Y., Vu, T. H., Sohn, J. H., Hwang, S., Goo, J. M., et al. (2019) · Radiology · doi:10.1148/radiol.2018180237 · Cited by 539 · cited 1×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundMcBee, M. P., Awan, O. A., Colucci, A. T., Ghobadi, C. W., Kadom, N., Kansagra, A. P., Tridandapani, S., & Auffermann, W. F. (2018) · Academic Radiology · doi:10.1016/j.acra.2018.02.018 · Cited by 496
Research works printed but not linked: 3
- Auffermann WF, 2019, AJR Am J Roentgenol, 1
- Summers RM, NIH Chest X-ray Dataset of 14 Common Thorax Disease Categories
- Chollet F. Xception: Deep Learning with Depthwise Separable Convolutions. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, July 21–26, 2017. Piscataway, NJ: IEEE, 2017; 1800–1807.
Cited by
Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis
Aggarwal, R., Sounderajah, V., Martin, G., Ting, D. S. W., Karthikesalingam, A., King, D., Ashrafian, H., & Darzi, A. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00438-z · Cited by 946
Huang, S.-C., Pareek, A., Seyyedi, S., Banerjee, I., & Lungren, M. P. (2020) · npj Digital Medicine · doi:10.1038/s41746-020-00341-z · Cited by 934
“As an example consider the “ simple ” task in radiology of identifying pneumonia on a chest radiograph, something that has been achieved by many investigators training deep learning models for automated detection and classi fi cation of pathologies on chest X-rays 19,20 .”
COVID-19 Image Data Collection: Prospective Predictions are the Future
Cohen, J. P., Morrison, P., Dao, L., Roth, K., Duong, T., & Ghassem, M. (2020) · Machine Learning for Biomedical Imaging · doi:10.59275/j.melba.2020-48g7 · Cited by 927
“, 2019]), neural networks have even reported being able to achieve performance near radiologist levels [Rajpurkar et al., 2018, 2017; Irvin et al., 2019; Putha et al., 2018a; Majkowska et al., 2019; Putha et al., 2018b].”
BackgroundDeep learning for chest X-ray analysis: A survey
Çallı, E., Sogancioglu, E., van Ginneken, B., van Leeuwen, K. G., & Murphy, K. (2021) · Medical Image Analysis · doi:10.1016/j.media.2021.102125 · Cited by 506
VinDr-CXR: An open dataset of chest X-rays with radiologist’s annotations
Nguyen, H. Q., Lam, K., Le, L. T., Pham, H. H., Tran, D. Q., Nguyen, D. B., Le, D. D., Pham, C. M., Tong, H. T. T., Dinh, D. H., et al. (2022) · Scientific Data · doi:10.1038/s41597-022-01498-w · Cited by 435
Rethinking CNN Models for Audio Classification
Palanisamy, K., Singhania, D., & Yao, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.11154 · Cited by 107
“In recent years deep models trained on a large corpus like ImageNet for classification have been widely used for transfer learning for tasks such as Image Segmentation [39], [40], Medical Image Analysis [41], [42].”
MethodPredicting COVID-19 Pneumonia Severity on Chest X-ray With Deep Learning
Cohen, J. P., Dao, L., Roth, K., Morrison, P., Bengio, Y., Abbasi, A. F., Shen, B., Mahsa, H. K., Ghassemi, M., Li, H., et al. (2020) · Cureus · doi:10.7759/cureus.9448 · Cited by 259
Introduction to artificial intelligence in ultrasound imaging in obstetrics and gynecology
Drukker, L., Noble, J. A., & Papageorghiou, A. T. (2020) · Ultrasound in Obstetrics & Gynecology · doi:10.1002/uog.22122 · Cited by 240
“Promising medical AI applications are emerging in the areas of screening 5,6 , prediction 7–9 , triage 10,11 , diagnosis 12,13 , drug development 14,15 , treatment 16,17 , monitoring 18 and imaging interpretation 19,20 .”
BackgroundHolzinger, A., Dehmer, M., Emmert-Streib, F., Cucchiara, R., Augenstein, I., Ser, J. D., Samek, W., Jurisica, I., & Díaz-Rodríguez, N. (2022) · Information Fusion · doi:10.1016/j.inffus.2021.10.007 · Cited by 213
Machine Learning in oncology: A clinical appraisal
Cuocolo, R., Caruso, M., Perillo, T., Ugga, L., & Petretta, M. (2020) · Cancer Letters · doi:10.1016/j.canlet.2020.03.032 · Cited by 210
Heterogeneity and predictors of the effects of AI assistance on radiologists
Yu, F., Moehring, A., Banerjee, O., Salz, T., Agarwal, N., & Rajpurkar, P. (2024) · Nature Medicine · doi:10.1038/s41591-024-02850-w · Cited by 200
Jin, L., Yang, J., Kuang, K., Ni, B., Gao, Y., Sun, Y., Gao, P., Ma, W., Tan, M., Kang, H., et al. (2020) · eBioMedicine · doi:10.1016/j.ebiom.2020.103106 · Cited by 193
“The inspiring results achieved by human-computer collaboration were consistent with a pervious study [4] in chest radiograph interpretation.”
MethodResultInfluentialBone Fracture Detection Using Deep Supervised Learning from Radiological Images: A Paradigm Shift
Meena, T., & Roy, S. (2022) · Diagnostics · doi:10.3390/diagnostics12102420 · Cited by 184
“[44] 2020 Radiographic images Xception Chest The authors developed a model to detect opacity, pneumothorax, mass or nodule, and fracture.”
BackgroundBaclic, O., Tunis, M., Young, K., Doan, C., & Swerdfeger, H. (2020) · Canada Communicable Disease Report · doi:10.14745/ccdr.v46i06a02 · Cited by 134
“AI research has produced models that can interpret a radiograph (1,2), detect irregular heartbeats using a smartwatch (3), automatically identify reports of infectious disease in the media (4), ascertain cardiovascular risk factors from retinal images (5) and find new targets for existing medications (6,7).”
BackgroundWu, J. T., Wong, K. C. L., Gur, Y., Ansari, N., Karargyris, A., Sharma, A., Morris, M., Saboury, B., Ahmad, H., Boyko, O., et al. (2020) · JAMA Network Open · doi:10.1001/jamanetworkopen.2020.22779 · Cited by 149
Active label cleaning for improved dataset quality under resource constraints
Roschewitz, M., Castro, D. C., Tanno, R., Schwaighofer, A., Tezcan, K. C., Monteiro, M., Bannur, S., Lungren, M. P., Nori, A., Glocker, B., et al. (2022) · Nature Communications · doi:10.1038/s41467-022-28818-3 · Cited by 168
“(2), as the framework makes no assumptions about the family of functions that can be used for label cleaning.”
BackgroundInfluentialAzemin, M. Z. C., Hassan, R., Tamrin, M. I. M., & Ali, M. A. B. M. (2020) · International Journal of Biomedical Imaging · doi:10.1155/2020/8828855 · Cited by 162
“The labels were provided by a recent research from Google Health [11]; the research was motivated by the need of more accurate ground truth for chest X-ray diagnosis.”
BackgroundMethodResultInfluentialExplainable Deep Learning Methods in Medical Image Classification: A Survey
Patrício, C., Neves, J. C., & Teixeira, L. F. (2024) · ACM Computing Surveys · doi:10.1145/3625287 · Cited by 161
“Medical image analysis is one of the applications where the progress made ensured human-level accuracy on the classification of different types of medical data (e.g., chest X-rays [90] and corneal images [166]).”
BackgroundTriple attention learning for classification of 14 thoracic diseases using chest radiography
Wang, H., Wang, S., Qin, Z., Zhang, Y., Li, R., & Xia, Y. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101846 · Cited by 157
CTransCNN: Combining transformer and CNN in multilabel medical image classification
Wu, X., Feng, Y., Xu, H. L., Lin, Z., Chen, T., Li, S., Qiu, S., Liu, Q., Ma, Y., & Zhang, S. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.111030 · Cited by 155
“Specifically, CNNs have demonstrated remarkable performance [2].”
Machine and Deep Learning for Tuberculosis Detection on Chest X-Rays: Systematic Literature Review
Hansun, S., Argha, A., Liaw, S., Celler, B. G., & Marks, G. B. (2023) · Journal of Medical Internet Research · doi:10.2196/43154 · Cited by 155
“By contrast, ChestX-ray14 contains around 112,120 CXR images of 30,805 patients [79], which is considered as a large data set [80-82].”
On the limits of cross-domain generalization in automated X-ray prediction
Cohen, J., Hashir, M., Brooks, R., & Bertrand, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.02497 · Cited by 59
“, 2016), Google (Majkowska et al., 2019), Kaggle aka the RSNA Pneumonia Detection Challenge3.”
BackgroundMethodInfluentialElaziz, M. A., Ewees, A. A., Yousri, D., Alwerfali, H. S. N., Awad, Q. A., Lu, S., & Al‐qaness, M. A. A. (2020) · IEEE Access · doi:10.1109/access.2020.3007928 · Cited by 144
“To assess the quality of the segmentation method for COVID-19 CT images, a set of thirteen images is used VOLUME 8, 2020 from [53] as in Figure 7. images are collected from different datasets such as CheX aka CheXpert [94], OpenI [95], Google [96], PC aka PadChest [97], NIH aka Chest X-ray14 [98], and MIMIC-CXR [99].”
Yoo, H., Kim, K. H., Singh, R., Digumarthy, S. R., & Kalra, M. K. (2020) · JAMA Network Open · doi:10.1001/jamanetworkopen.2020.17135 · Cited by 122
Zhu, J., Chen, Z., Zhao, J., Yu, Y., Li, X., Shi, K., Zhang, F., Yu, F., Shi, K., Sun, Z., et al. (2023) · BMC Oral Health · doi:10.1186/s12903-023-03027-6 · Cited by 133
“The annotated images were reviewed and revised by another 2 oral and maxillofacial imaging experts and achieved final confirmation [26].”
MethodKutbi, M. (2024) · Diagnostics · doi:10.3390/diagnostics14171879 · Cited by 88
“High-quality, annotated datasets are crucial for training effective AI models, but their availability is often limited, and obtaining such datasets can be costly and time-consuming [37].”
BackgroundMethodInfluentialEslami, M., Tabarestani, S., Albarqouni, S., Adeli, E., Navab, N., & Adjouadi, M. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.2974159 · Cited by 120
“Four different deep learning based methods are investigated in [12] and”
VILA-M3: Enhancing Vision-Language Models with Medical Expert Knowledge
Nath, V., Li, W., Yang, D., Myronenko, A., Zheng, M., Lu, Y., Liu, Z., Yin, H., Law, Y. M., Tang, Y., et al. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01378 · Cited by 120
“One is the subset of ChestX-ray14 [50] curated by Majkowska and Mittal et al. [35], consisted of 1,962 images after an adjudi-cation process of four radiologists.”
Artificial intelligence in healthcare: a comprehensive review of its ethical concerns
Kooli, C., & Al Muftah, H. (2022) · Technological Sustainability · doi:10.1108/techs-12-2021-0029 · Cited by 83
Exploring scalable medical image encoders beyond text supervision
Pérez-García, F., Sharma, H., Bond-Taylor, S., Bouzid, K., Salvatelli, V., Ilse, M., Bannur, S., Castro, D. C., Schwaighofer, A., Lungren, M. P., et al. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-024-00965-w · Cited by 115
“In these benchmarks, diagnostic labels were extracted from radiology reports using a parser, resulting in significant label noise [112, 113].”
A scalable physician-level deep learning algorithm detects universal trauma on pelvic radiographs
Cheng, C., Wang, Y., Chen, H., Hsiao, P.-M., Yeh, C., Hsieh, C., Miao, S., Xiao, J. Z., Liao, C., & Lü, L. (2021) · Nature Communications · doi:10.1038/s41467-021-21311-3 · Cited by 113
AI-based radiodiagnosis using chest X-rays: A review
Akhter, Y., Singh, R., & Vatsa, M. (2023) · Frontiers in Big Data · doi:10.3389/fdata.2023.1120989 · Cited by 113
Kazemzadeh, S., Yu, J., Jamshy, S., Pilgrim, R., Nabulsi, Z., Chen, C., Beladia, N., Lau, C., McKinney, S. M., Hughes, T., et al. (2023) · Radiology · doi:10.1148/radiol.212213 · Cited by 86
Self-citationDeep Mining External Imperfect Data for Chest X-Ray Disease Screening
Luo, L., Yu, L., Chen, H., Liu, Q., Wang, X., Xu, J., & Heng, P. (2020) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2020.3000949 · Cited by 104
“With TenCrop, the proposed method further achieves competitive AUC with the ensemble model in [54] on ”Pneumothorax” ( 0 .”
BackgroundMethodInfluentialNam, J. G., Kim, M., Park, J., Hwang, E. J., Lee, J. H., Hong, J. H., Goo, J. M., & Park, C. M. (2021) · European Respiratory Journal · doi:10.1183/13993003.03061-2020 · Cited by 103
“The training data of DLAD-10 were curated by radiologists mostly without CT reference, intended to resemble radiologists ’ performance, resulting in reasonable output for the readers [28, 29].”
Method“Who goes first? Influences of human-AI workflow on decision making in clinical imaging,”
Fogliato, R., Chappidi, S., Lungren, M., Fisher, P., Wilson, D., Fitzke, M., Parkinson, M., Horvitz, E., Inkpen, K., & Nushi, B. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533193 · Cited by 103
Objective evaluation of deep uncertainty predictions for COVID-19 detection
Asgharnezhad, H., Shamsi, A., Alizadehsani, R., Khosravi, A., Nahavandi, S., Sani, Z. A., Srinivasan, D., & Islam, S. M. S. (2022) · Scientific Reports · doi:10.1038/s41598-022-05052-x · Cited by 95
Öztürk, Ş., Turali, M. Y., & Çukur, T. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106959 · Cited by 96
Xue, C., Yu, L., Chen, P., Dou, Q., & Heng, P.-A. (2022) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3140140 · Cited by 94
“We used 1,962 manual labeled images [49] as clean testing data.”
BackgroundMethodSimplified Transfer Learning for Chest Radiography Models Using Less Data
Sellergren, A. B., Chen, C., Nabulsi, Z., Li, Y., Maschinot, A., Sarna, A., Huang, J., Lau, C., Kalidindi, S. R., Etemadi, M., et al. (2022) · Radiology · doi:10.1148/radiol.212482 · Cited by 58
Self-citationGarcia Santa Cruz, B., Bossa, M. N., Sölter, J., & Husch, A. D. (2021) · Medical Image Analysis · doi:10.1016/j.media.2021.102225 · Cited by 89
Nowroozi, A. A., Salehi, M. A., Shobeiri, P., Agahi, S., Momtazmanesh, S., Kaviani, P., & Kalra, M. K. (2024) · Clinical Radiology · doi:10.1016/j.crad.2024.04.009 · Cited by 87
TorchXRayVision: A library of chest X-ray datasets and models
Cohen, J. P., Viviano, J. D., Bertin, P., Morrison, P., Torabian, P., Guarrera, M., Lungren, M. P., Chaudhari, A., Brooks, R., Hashir, M., et al. (2021) · arXiv · doi:10.48550/arxiv.2111.00595 · Cited by 82
Houssein, E. H., Helmy, B. E.-D., Oliva, D. A., Jangir, P., Premkumar, M., Elngar, A. A., & Shaban, H. (2021) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2021.103401 · Cited by 81
Using generative AI to investigate medical imagery models and datasets
Lang, O., Stupp, D., Traynis, I., Cole-Lewis, H. J., Bennett, C. R., Lyles, C. R., Lau, C., Irani, M., Semturs, C., Webster, D. R., et al. (2024) · EBioMedicine · doi:10.1016/j.ebiom.2024.105075 · Cited by 79
Self-citationJin, H., Geng, J., Yin, Y., Hu, M., Yang, G., Xiang, S., Zhai, X., Ji, Z., Fan, X., Hu, P., et al. (2020) · Journal of NeuroInterventional Surgery · doi:10.1136/neurintsurg-2020-015824 · Cited by 79
“15 16 More and more aneurysm detection and segmentation procedures based on a deep learning technique have been reported recently, including abdominal aortic aneurysm detection and segmentation on CTA images using a conventional deep convolutional neural network approach, 17 and cerebral aneurysm…”
MethodContinual Learning for Domain Adaptation in Chest X-ray ClassificationPreprint
Lenga, M., Schulz, H., & Saalbach, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.05922 · Cited by 78
“More recently, for all findings in the ChestX-ray14 dataset (Wang et al., 2017), a performance similar to radiologists was reported (Majkowska et al., 2019).”
MethodZhou, Q., Zhu, W., Li, F., Yuan, M., Zheng, L., & Liu, X. (2021) · Current Pharmaceutical Design · doi:10.2174/1381612827666211213143357 · Cited by 76
Nabulsi, Z., Sellergren, A., Jamshy, S., Lau, C. T., Santos, E. R. G., Kiraly, A. P., Ye, W., Yang, J., Pilgrim, R., Kazemzadeh, S., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-93967-2 · Cited by 74
Self-citationCohen, J., Brooks, R., En, S., Zucker, E. J., Pareek, A., Lungren, M. P., & Chaudhari, A. S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.09475 · Cited by 79
Diagnostic accuracy of deep learning in orthopaedic fractures: a systematic review and meta-analysis
Yang, S., Yin, B., Cao, W., Feng, C., Fan, G., & He, S. (2020) · Clinical Radiology · doi:10.1016/j.crad.2020.05.021 · Cited by 72
Güngör, İ., Günaydın, B., Oktar, S. Ö., Yeşil, B. M. B., Bağcaz, S., Özdemir, M. G., & İnan, G. (2021) · Journal of Anesthesia · doi:10.1007/s00540-021-02947-3 · Cited by 72
Attention-based Saliency Maps Improve Interpretability of Pneumothorax Classification
Wollek, A., Graf, R., Čečatka, S., Fink, N., Willem, T., Sabel, B. O., & Lasser, T. (2023) · Radiology: Artificial Intelligence · doi:10.1148/ryai.220187 · Cited by 68
Machine Learning Augmented Interpretation of Chest X-rays: A Systematic Review
Ahmad, H. K., Milne, M., Buchlak, Q. D., Ektas, N., Sanderson, G., Chamtie, H., Karunasena, S., Chiang, J., Holt, X., Tang, C. H. M., et al. (2023) · Diagnostics · doi:10.3390/diagnostics13040743 · Cited by 67
“Study Study Aim Study Design Datasets Used Dataset Size Number of PathologiesInvestigated Ahn et al., 2022 [43] Evaluate whether a deep-learning–based AI engine used concurrently can improve reader performance and efficiency in interpreting CXR abnormalities Retrospective reader study Two sources: MIMIC-CXR(public) an…”
InfluentialHow Transferable Are Self-supervised Features in Medical Image Classification Tasks?Preprint
Truong, T., Mohammadi, S., & Lenga, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.10048 · Cited by 66
“We use two subsets of the original NIH Chest X-ray dataset, which are referred to as validation and test set in the study at (Majkowska et al., 2020).”
MethodDawood, Y., Buijtendijk, M. F. J., Shah, H., Smit, J. A., Jacobs, K., Hagoort, J., Oostra, R., Bourne, T. H., van den Hoff, M. J. B., & de Bakker, B. S. (2022) · Seminars in Cell and Developmental Biology · doi:10.1016/j.semcdb.2022.02.023 · Cited by 65
Human–large language model collaboration in clinical medicine: a systematic review and meta-analysis
Wang, G., Zhang, K., Jiang, J., Wang, C., Bi, H., Liang, H., Qi, Z., Huang, Y., Li, Y., & Yang, X. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02382-2 · Cited by 67
Andersson, P., Johnsson, J., Björnsson, O., Cronberg, T., Hassager, C., Zetterberg, H., Stammet, P., Undén, J., Kjærgaard, J., Friberg, H., et al. (2021) · Critical Care · doi:10.1186/s13054-021-03505-9 · Cited by 64
“Improvements in machine learning algorithms and increased computational power have led to an enhanced diagnostic and prognostic capability in a variety of medical fields, ranging from radiology to intensive care medicine [17, 18].”
BackgroundEvaluating the Clinical Realism of Synthetic Chest X-Rays Generated Using Progressively Growing GANs
Segal, B., Rubin, D. M., Rubin, G., & Pantanowitz, A. (2021) · SN Computer Science · doi:10.1007/s42979-021-00720-7 · Cited by 64
Deep Learning to Quantify Pulmonary Edema in Chest Radiographs
Horng, S., Liao, R., Wang, X., Dalal, S., Golland, P., & Berkowitz, S. J. (2021) · Radiology Artificial Intelligence · doi:10.1148/ryai.2021190228 · Cited by 62
Pham, H., Nguyen, N. H., Tran, T. T., Nguyen, T. N.-M., & Nguyen, H. Q. (2023) · Scientific Data · doi:10.1038/s41597-023-02102-5 · Cited by 62
Correcting data imbalance for semi-supervised COVID-19 detection using X-ray chest images
Calderon-Ramirez, S., Shengxiang-Yang, Moemeni, A., Elizondo, D. A., Colreavy‐Donnelly, S., Chavarría-Estrada, L. F., & Molina‐Cabello, M. A. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107692 · Cited by 63
“A wider range of datasets were used in such work for training the feature extractor [19], [11], [26], [30],”
MethodJang, H., McCormack, D. R., & Tong, F. (2021) · PLoS Biology · doi:10.1371/journal.pbio.3001418 · Cited by 61
Dynamic Chest X-Ray Using a Flat-Panel Detector System: Technique and Applications
Hata, A., Yamada, Y., Tanaka, R., Nishino, M., Hida, T., Hino, T., Ueyama, M., Yanagawa, M., Kamitani, T., Kurosaki, A., et al. (2020) · Korean Journal of Radiology · doi:10.3348/kjr.2020.1136 · Cited by 60
“Many studies have focused on using chest radiography with deep-learning-based algorithms to achieve high accuracy comparable to that obtained with human experts (75-79).”
ResultCoherent Concept-based Explanations in Medical Image and Its Application to Skin Lesion Diagnosis
Patrício, C., Neves, J. C., & Teixeira, L. F. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00394 · Cited by 56
“The research on automated medical image analysis is nowadays dominated by the use of deep learning models, which supported a remarkable increase in the accuracy of several medical image problems, assuring in some cases results that surpassed human performance [27, 36].”
BackgroundDeep learning for pneumothorax diagnosis: a systematic review and meta-analysis
Sugibayashi, T., Walston, S. L., Matsumoto, T., Mitsuyama, Y., Miki, Y., & Ueda, D. (2023) · European Respiratory Review · doi:10.1183/16000617.0259-2022 · Cited by 57
“…AUC: AI-one:0.89 AI-two:0.91 AUC(prediction): AI-one:0.50(0.45-0.57) AI-two:0.62(0.56-0.69) Mean of the change in probability:19.7%(9.729.7) [50] Majkowska 2020 Xception CXR14:72.8(64.1-81.0),90.8(88.9-93.1) DS1:64.8(47.7-78.4),99.7(99.3-100) Positive predictive value: CXR14:48.7%(43.8-55.8)…”
InfluentialDilated SE-DenseNet for brain tumor MRI classification
Mao, Y., Kim, J., Podina, L., & Kohandel, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-86752-y · Cited by 55
Zhang, X., Han, L., Sobeih, T., Han, L., Dempsey‐Hibbert, N., Lechareas, S., Tridente, A., Chen, H., White, S. J., & Zhang, D. (2022) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3220813 · Cited by 54
“divided into three categories: Lung disease classification [9], [10], [11], deep learning-based explanation[12], [13], [14], [15] and Organ area segmentation[16], [17], [18], [19].”
BackgroundUllah, N., Khan, J. A., Almakdi, S., Khan, M. S., Alshehri, M., Alboaneen, D., & Raza, A. (2022) · Applied Sciences · doi:10.3390/app12126269 · Cited by 53
Thian, Y. L., Ng, D., Hallinan, J. T. P. D., Jagmohan, P., Sia, S. Y., Tan, C. H., Ting, Y. H., Kei, P. L., Pulickal, G. G., Tiong, V. T. Y., et al. (2021) · Radiology Artificial Intelligence · doi:10.1148/ryai.2021200190 · Cited by 53
“It is noteworthy that Majkowska et al (7) and Hwang et al (23) tested commercial deep learning models on combined multicenter data from up to three institutions, but their results were limited by the fact that data from the same institution were used in both the training and the test sets.”
BackgroundTam, M. D. B. S., Dyer, T., Dissez, G., Morgan, T. N., Hughes, M., Illes, J., Rasalingham, R., & Rasalingham, S. (2021) · Clinical Radiology · doi:10.1016/j.crad.2021.03.021 · Cited by 43
Une brève introduction à l’intelligence artificielle
Jean, A. (2020) · médecine/sciences · doi:10.1051/medsci/2020189 · Cited by 52
Hwang, E. J., Kim, K. B., Kim, J. Y., Lim, J., Nam, J. G., Choi, H., Kim, H., Yoon, S. H., Goo, J. M., & Park, C. M. (2021) · PLoS ONE · doi:10.1371/journal.pone.0252440 · Cited by 51
CheXGAT: A disease correlation-aware network for thorax disease diagnosis from chest X-ray images
Lee, Y., Huang, S.-K., & Chang, R. (2022) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2022.102382 · Cited by 50
Validation study of machine-learning chest radiograph software in primary and emergency medicine
van Beek, E. J. R., Ahn, J. S., Kim, M. J., & Murchison, J. T. (2022) · Clinical Radiology · doi:10.1016/j.crad.2022.08.129 · Cited by 51
Large-scale long-tailed disease diagnosis on radiology images
Zheng, Q., Zhao, W., Wu, C., Zhang, X., Dai, L., Guan, H., Li, Y., Zhang, Y., Wang, Y., & Xie, W. (2024) · Nature Communications · doi:10.1038/s41467-024-54424-6 · Cited by 52
“Existing benchmarks typically focus on a very limited number of diseases, with single modality inputs 5,6 .”
Sellergren, A., Kazemzadeh, S., Jaroensri, T., Kiraly, A., Traverse, M., Kohlberger, T., Xu, S., Jamil, F., Hughes, C., Lau, C., et al. (2025) · arXiv · doi:10.48550/arxiv.2507.05201 · Cited by 49
Self-citationRudolph, J., Huemmer, C., Ghesu, F.-C., Mansoor, A., Preuhs, A., Fieselmann, A., Fink, N., Dinkel, J., Koliogiannis, V., Schwarze, V., et al. (2022) · Investigative Radiology · doi:10.1097/rli.0000000000000813 · Cited by 46
Ramli, P. N. M., Aizuddin, A. N., Ahmad, N., Hamid, Z. A., & Ismail, K. I. (2025) · Diagnostics · doi:10.3390/diagnostics15030246 · Cited by 47
In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defer
Cao, Y., Mozannar, H., Feng, L., Wei, H., & An, B. (2023) · neural information processing systems · doi:10.52202/075280-1671 · Cited by 47
Wang, X., Yang, S., Lan, J., Fang, Y., He, J., Wang, M., Zhang, J., & Han, X. (2020) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2020.3035572 · Cited by 44
“Different variants of DCNNs have been used, such as fully convolutional network (FCN) [11], DenseNets [20], [23]–[28], ResNet [20], [29], Xception [30], long short-term memory network (LSTM) [13], and squeeze-and-excitation (SE) block network [28].”
MethodInfluentialMir, A. N., & Rizvi, D. R. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.111413 · Cited by 45
CheXRelNet: An Anatomy-Aware Model for Tracking Longitudinal Relationships Between Chest X-Rays
Karwande, G., Mbakwe, A. B., Wu, J. T.-Y., Celi, L. A., Moradi, M., & Lourentzou, I. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16431-6_55 · Cited by 43
CELM: An Ensemble Deep Learning Model for Early Cardiomegaly Diagnosis in Chest Radiography
Yanar, E., Hardalaç, F., & Ayturan, K. (2025) · Diagnostics · doi:10.3390/diagnostics15131602 · Cited by 41
“Majkowska et al. [20] benchmarked deep CNN ensembles on radiographs, showing competitive AUCs across multiple thoracic pathologies.”
Automatic clustering method to segment COVID-19 CT images
Elaziz, M. A., Al‐qaness, M. A. A., Zaid, E. O. A., Lu, S., Ibrahim, R. A., & Ewees, A. A. (2021) · PLoS ONE · doi:10.1371/journal.pone.0244416 · Cited by 41
“These images are collected from different datasets such as CheX aka CheXpert [33], OpenI [34], Google [35], PC aka PadChest [36], NIH aka Chest X-ray14 [37], and MIMIC-CXR [38].”
MethodKulkarni, V., Gawali, M., & Kharat, A. T. (2021) · JMIR Medical Informatics · doi:10.2196/28776 · Cited by 41
“This risk can be mitigated by employing multiphase reviews [31] or expert adjudication [32] to create high-quality labels.”
BackgroundThian, Y. L., Ng, D. W., Hallinan, J. T. P. D., Jagmohan, P., Sia, S. Y., Adnan, J. S., Quek, S. T., & Feng, M. (2022) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-022-00594-y · Cited by 41
“…to computer vision tasks in radiology, high accuracies for specific pathologies have been reported in literature for deep learning systems using training sets as small as 1389 images for radius fracture detection [9] to as many as almost a million images for chest radiograph interpretation [10].”
BackgroundPham, H., Nguyen, H. Q., Lam, K., Le, L. T., Nguyen, D. B., Nguyen, H. T., Le, T. T., Nguyen, T. V., Dao, M., & Vu, V. (2022) · IEEE Access · doi:10.1109/access.2022.3210468 · Cited by 40
“In particular, previous studies [31], [37], [38], [39], [40], [41], [42], [43], [44], [45] have indicated that a deep learning (DL) system trained on a large-scale, annotatedmedical imaging dataset can reach a level of performance comparable to practicing radiologists in detecting common thoracic diseases [31], [37],…”
BackgroundResultAttack-agnostic Adversarial Detection on Medical Data Using Explainable Machine Learning
Watson, M., & Moubayed, N. A. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412560 · Cited by 38
Gipson, J., Tang, V. D., Seah, J. C. Y., Kavnoudias, H., Zia, A., Lee, R., Mitra, B., & Clements, W. (2022) · British Journal of Radiology · doi:10.1259/bjr.20210979 · Cited by 38
Reducing Errors Resulting From Commonly Missed Chest Radiography Findings
Gefter, W. B., & Hatabu, H. (2022) · CHEST Journal · doi:10.1016/j.chest.2022.12.003 · Cited by 38
Deep Learning to Quantify Pulmonary Edema in Chest RadiographsPreprint
Horng, S., Liao, R., Wang, X., Dalal, S., Golland, P., & Berkowitz, S. J. (2021) · DSpace@MIT (Massachusetts Institute of Technology) · Cited by 38
Cid, Y. D., Macpherson, M., Gervais-Andre, L., Zhu, Y., Franco, G., Santeramo, R., Lim, C., Selby, I., Muthuswamy, K., Amlani, A., et al. (2024) · The Lancet Digital Health · doi:10.1016/s2589-7500(23)00218-2 · Cited by 37
Jin, K. N., Kim, E. Y., Kim, Y. J., Lee, G. P., Kim, H., Oh, S., Kim, Y., Han, J., & Cho, Y. J. (2022) · European Radiology · doi:10.1007/s00330-021-08397-5 · Cited by 37
“Consensus reading can be used for standards reference [27].”
MethodIkechukwu, A. V., & Murali, S. (2023) · Machine Learning Science and Technology · doi:10.1088/2632-2153/acd2a5 · Cited by 38
Automated Rib Fracture Detection on Chest X-Ray Using Contrastive Learning
Sun, H., Wang, X., Li, Z., Liu, A., Xu, S., Jiang, Q., Li, Q., Xue, Z., Gong, J., Chen, L., et al. (2023) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-023-00868-z · Cited by 37
“In one study [16], a DL-based model of chest radiography was established to identify four types of features (fracture, pneumothorax, nodule, and opacity), and the sensitivity and specificity for fracture detection based on the public dataset ChestX-Ray14 were 43.”
BackgroundMethodExploring Vision Transformers for 3D Human Motion-Language Models with Motion Patches
Yu, Q., Tanaka, M., & Fujiwara, K. (2024) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52733.2024.00095 · Cited by 38
Dreizin, D., Staziaki, P. V., Khatri, G. D., Beckmann, N. M., Feng, Z., Liang, Y., DelProposto, Z., Klug, M., Spann, J. S., Sarkar, N., et al. (2023) · Emergency Radiology · doi:10.1007/s10140-023-02120-1 · Cited by 36
“Dataset size was related to the imaging modality and anatomical region/system, with the largest datasets of more than 500,000 patients for chest radiography [35, 36], followed by musculoskeletal radiography with 314,866 patients [37], and a head CT dataset with 313,318 exams [38].”
MethodShetty, S., Ananthanarayana, V. S., & Mahale, A. R. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-14940-x · Cited by 37
“We can illustrate with an elementary exemplar like detection of pneumonia from a chest radiograph, where researchers have previously worked on building DL models to automate the process of identifying and classifying pathologies from chest radiographs [48, 60].”
MethodSelf-citationNodule Detection and Generation on Chest X-Rays: NODE21 Challenge
Sogancioglu, E., van Ginneken, B., Behrendt, F., Bengs, M., Schlaefer, A., Radu, M., Xu, D., Sheng, K., Scalzo, F., Marcus, E., et al. (2024) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2024.3382042 · Cited by 36
“Many of the works using these public datasets used the NLP-generated labels [14], [15], [16], [17], [18] for evaluation or, at best, radiologist assessment of CXR images [19], [20], [21], [22].”
Metadata
- Authors
- Majkowska, A., Mittal, S., Steiner, D. F., Reicher, J. J., McKinney, S. M., Duggan, G. E., Eswaran, K., Cameron Chen, P.-H., Liu, Y., Kalidindi, S. R., Ding, A., Corrado, G. S., Tse, D., & Shetty, S.
- Affiliations
- Stanford Health Care (US)
- From Google Health, 1600 Amphitheatre Pkwy, Mountain View, CA 94043 (A.M., S.M., D.F.S., S.M.M., G.E.D., K.E., P.H.C.C., Y.L., G.S.C., D.T., S.S.); Stanford Healthcare and Palo Alto Veterans Affairs, Palo Alto, Calif (J.J.R.); Apollo Radiology International, Hyderabad, India (S.R.K.); and California Advanced Imaging, Novato, Calif (A.D.).
- ORCID
- Venue
- Radiology
- Published
- 2020-02
- Volume
- 294
- Issue
- 2
- Pages
- 421-431
- ISSN
- 0033-8419 (ISSN-L)
- 1527-1315
- OpenAlex
- W2991676991
- Cited by
- 277
- References
- 30
- Citations by year
- 2026: 21 · 2025: 37 · 2024: 36 · 2023: 48 · 2022: 48 · 2021: 52 · 2020: 34
- FWCI
- 22.28
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Radiology (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- chest X-ray interpretation, deep learning, pneumothorax detection, lung nodule detection, lung opacity, rib fracture detection, inverse probability weighting, spectrum bias, generalizability, ChestX-ray14, positive predictive value, sensitivity and specificity
- MeSH terms
- Radiologists, Deep Learning, Adolescent, Adult, Aged, Aged, 80 and over, Child, Child, Preschool, Female, Humans, Infant, Male, Middle Aged, Pneumothorax, Radiographic Image Interpretation, Computer-Assisted, Reference Standards, Respiratory Tract Diseases, Retrospective Studies, Sensitivity and Specificity, Thoracic Injuries, Radiography, Thoracic, Reproducibility of Results, Young Adult
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.