Assessment of Convolutional Neural Networks for Automated Classification of Chest Radiographs
Open accessThe DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.
Abstract
Purpose To assess the ability of convolutional neural networks (CNNs) to enable high-performance automated binary classification of chest radiographs. Materials and Methods In a retrospective study, 216 431 frontal chest radiographs obtained between 1998 and 2012 were procured, along with associated text reports and a prospective label from the attending radiologist. This data set was used to train CNNs to classify chest radiographs as normal or abnormal before evaluation on a held-out set of 533 images hand-labeled by expert radiologists. The effects of development set size, training set size, initialization strategy, and network architecture on end performance were assessed by using standard binary classification metrics; detailed error analysis, including visualization of CNN activations, was also performed. Results Average area under the receiver operating characteristic curve (AUC) was 0.96 for a CNN trained with 200 000 images. This AUC value was greater than that observed when the same model was trained with 2000 images (AUC = 0.84, P < .005) but was not significantly different from that observed when the model was trained with 20 000 images (AUC = 0.95, P > .05). Averaging the CNN output score with the binary prospective label yielded the best-performing classifier, with an AUC of 0.98 (P < .005). Analysis of specific radiographs revealed that the model was heavily influenced by clinically relevant spatial regions but did not reliably generalize beyond thoracic disease. Conclusion CNNs trained with a modestly sized collection of prospectively labeled chest radiographs achieved high diagnostic performance in the classification of chest radiographs as normal or abnormal; this function may be useful for automated prioritization of abnormal chest radiographs. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by van Ginneken in this issue.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The clinical acceptance of AI techniques would be aided by multi-modality and information fusion of both imaging and non-imaging data [34,170].”
From Survey of explainable artificial intelligence techniques for biomedical imaging with deep neural networks (2023) · cites this paper as [170]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“deep learning models have demonstrated diagnostic accuracy comparable to that of human experts in narrow clinical tasks for several medical domains and imaging modalities, including chest and extremity X-rays8–10, computed tomography (CT)11, magnetic resonance imaging (MRI)12, whole slide images (WSI)13,14, and dermatology images15.”
From Self-supervised learning for medical image classification: a systematic review and implementation guidelines (2023) · cites this paper as [8]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The current state-of-the-art deep learning models for radiology applications consider only pixel-value information without data informing clinical context.”
From Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines (2020) · cites this paper as [13]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Early applications of CNNs for image analysis in medicine include diabetic retinopathy, skin cancer, and chest X-rays11–18.”
From Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines (2020) · cites this paper as [13]
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 [79]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 3×
“When we moved from AlexNet (18) to ResNet-18 (19), we observed an 11-point improvement in negative class F1, mostly driven by higher recall.”
MethodLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
DeLong, E. R., DeLong, D. M., & Clarke-Pearson, D. L. (1988) · Biometrics · doi:10.2307/2531595 · Cited by 22,649 · cited 1×
“The DeLong nonparametric statistical test (24) implemented in the Daim (version 1.1.0) R library was used to assess statistical differences among AUC values, wherein P values less than .05 were considered to indicate a significant difference.”
MethodInfluentialGulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., et al. (2016) · JAMA · doi:10.1001/jama.2016.17216 · Cited by 7,697 · cited 1×
“These scalability results are in line with those of Gulshan et al (27), who found that CNN performance for classification of retinal fundus photographs reached a plateau after approximately 60 000 images.”
ResultHanley, J. A., & McNeil, B. J. (1983) · Radiology · doi:10.1148/radiology.148.3.6878708 · Cited by 7,016 · cited 1×
“The method described by Hanley and McNeil (25) was used to compute 95% confidence intervals for AUC values.”
MethodLakhani, P., & Sundaram, B. (2017) · Radiology · doi:10.1148/radiol.2017162326 · Cited by 1,775
Park, S. H., & Han, K. (2018) · Radiology · doi:10.1148/radiol.2017171920 · Cited by 894
Snorkel: rapid training data creation with weak supervision
Ratner, A., Bach, S. H., Ehrenberg, H., Fries, J., Wu, S., & Ré, C. (2017) · Proceedings of the VLDB Endowment · doi:10.14778/3157794.3157797 · Cited by 777
Self-citationEfficient Additive Kernels via Explicit Feature Maps
Vedaldi, A., & Zisserman, A. (2012) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2011.153 · Cited by 688
Diagnostic Accuracy of Digital Screening Mammography With and Without Computer-Aided Detection
Lehman, C. D., Wellman, R. D., Buist, D. S. M., Kerlikowske, K., Tosteson, A. N. A., & Miglioretti, D. L. (2015) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2015.5231 · Cited by 678 · cited 1×
“Further, this method of disease-specific computer-assisted diagnosis may not ultimately be beneficial to the interpreting clinician (16).”
Method
Research works printed but not linked: 10
- Rosman DA, Nshizirungu JJ, Rudakemwa E, et al. Imaging in the land of 1000 hills: Rwanda radiology country report. J Glob Radiol 2015;1(1):6. [Google Scholar]
- Ali FS, Harrington SG, Kennedy SB, Hussain S. Diagnostic radiology in Liberia: a country report. J Glob Radiol 2015;1(1):6. [Google Scholar]
- Deng J, Dong W, Socher R, Li LJ, Li K, Li F. ImageNet: a large-scale hierarchical image database. In: IEEE CVPR 2009; 248–255. [Google Scholar]
- Wang X, Peng Y, Lu L, Lu Z, Bagheri M, Summers RM. ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases. In: IEEE CVPR 2017; 3462–3471. [Google Scholar]
- Davies ER. Computer and machine vision: theory, algorithms, practicalities. Waltham, Mass: Academic Press/Elsevier, 2012. [Google Scholar]
- He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: IEEE CVPR 2016; 770–778. [Google Scholar]
- Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. In: IEEE CVPR 2017;3:4700–4708. [Google Scholar]
- Paszke A, Gross S, Chintala S, et al. Automatic differentiation in PyTorch. Neural Inf Process Syst Autodiff Work, 2017. https://openreview.net/forum?id=BJJsrmfCZ. Accessed May 8, 2018.
- Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A. Learning deep features for discriminative localization. In: IEEE CVPR 2016; 2921–2929. [Google Scholar]
- Kingma DP, Ba JL. Adam: a method for stochastic optimization. In: International Conference on Learning Representations 2015. https://pdfs.semanticscholar.org/aeea/02a93d674e0a044a8715e767f3a372582604.pdf. Published 2015. Accessed March 31, 2018. [Google Scholar]
Cited by
Items 101–200 of 206
Chest X-Ray Investigation: A Convolutional Neural Network Approach
Das, T. K., Chowdhary, C. L., & Gao, X. (2020) · Journal of biomimetics, biomaterials and biomedical engineering · doi:10.4028/www.scientific.net/jbbbe.45.57 · Cited by 28
Hossain, I., Puppala, S., & Talukder, S. K. (2023) · IEEE International Conference on AI in Cybersecurity (ICAIC) · doi:10.1109/icaic57335.2023.10044122 · Cited by 29
“Medical images are processed with DNNs for faster diagnosis of skin disease [2], lung cancer [3], or diabetic retinopathy (DR) [4].”
BackgroundFilice, R. W. (2019) · Journal of the American College of Radiology · doi:10.1016/j.jacr.2019.05.010 · Cited by 22
Lee, R. S., Dunnmon, J., He, A., Tang, S., Ré, C., & Rubin, D. L. (2020) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2020.103656 · Cited by 27
Self-citationSearching for pneumothorax in x-ray images using autoencoded deep features
Sze-To, A., Riasatian, A., & Tizhoosh, H. R. (2021) · Scientific Reports · doi:10.1038/s41598-021-89194-4 · Cited by 24
A survey on artificial intelligence in pulmonary imaging
Saha, P. K., Nadeem, S. A., & Comellas, A. P. (2023) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.1510 · Cited by 26
Yi, C., Tang, Y., Ouyang, R., Zhang, Y., Cao, Z., Yang, Z., Wu, S., Han, M., Xiao, J. Z., Chang, P., et al. (2021) · European Radiology · doi:10.1007/s00330-021-08275-0 · Cited by 20
End‐to‐end deep learning classification of vocal pathology using stacked vowels
Liu, G. S., Hodges, J. M., Yu, J., Sung, C. K., Erickson, E., & Doyle, P. C. (2023) · Laryngoscope Investigative Otolaryngology · doi:10.1002/lio2.1144 · Cited by 24
Chest Disease Image Classification Based on Spectral Clustering Algorithm
Jiang, S., Gu, Y., & Kumar, E. (2023) · Research Reports on Computer Science · doi:10.37256/rrcs.2120232742 · Cited by 23
“it is also difficult to find the most representative features for fine-grained classification of chest diseases [13, 14].”
BackgroundGefter, W. B., Prokop, M., Seo, J. B., Raoof, S., Langlotz, C. P., & Hatabu, H. (2024) · Radiology · doi:10.1148/radiol.232778 · Cited by 24
Self-citationHan, P., Jiang, L., Cheng, J.-L., Shi, K., Huang, S., Jiang, Y., Jiang, L., Xia, Q., Li, Y.-Y., Zhu, M., et al. (2023) · European Journal of Radiology · doi:10.1016/j.ejrad.2023.111277 · Cited by 22
Transfer learning with NASNet-Mobile for Pneumonia X-ray classification
Naskinova, I. (2022) · Asian-European Journal of Mathematics · doi:10.1142/s1793557122502400 · Cited by 22
Hopson, J. B., Neji, R., Dunn, J., McGinnity, C. J., Flaus, A., Reader, A. J., & Hammers, A. (2022) · IEEE Transactions on Radiation and Plasma Medical Sciences · doi:10.1109/trpms.2022.3231702 · Cited by 19
Articles That Use Artificial Intelligence for Ultrasound: A Reader’s Guide
Kuang, M., Hu, H.-T., Li, W., Chen, S., & Lü, X. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.631813 · Cited by 17
“(34) in the research reporting that the AI performance benefited little after a certain number of images were used for training.”
BackgroundKim, W., Lee, S. M., Kim, J. I., Ahn, Y., Park, S., Choe, J., & Seo, J. B. (2021) · American Journal of Roentgenology · doi:10.2214/ajr.21.26682 · Cited by 19
Kim, H., Park, C. M., & Goo, J. M. (2020) · European Radiology · doi:10.1007/s00330-019-06589-8 · Cited by 19
Chatterjee, D., Kanhere, A., Doo, F. X., Zhao, J., Chan, A., Welsh, A., Kulkarni, P., Trang, A., Parekh, V. S., & Yi, P. H. (2024) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-024-01273-w · Cited by 26
Artificial intelligence in traumatology
Breu, R., Avelar, C., Bertalan, Z., Grillari, J., Redl, H. R., Ljuhar, R., Quadlbauer, S., & Hausner, T. (2024) · Bone and Joint Research · doi:10.1302/2046-3758.1310.bjr-2023-0275.r3 · Cited by 15
Nguyen, Q., Pham, H., Wong, K., Nguyen, P. L., Nguyen, T. T., & Do, M. N. (2023) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2023.3314066 · Cited by 18
“Consequently, the availability of the on-device memory constraints the size of the neural networks model in training [18], [19].”
BackgroundKent, M., Vasconcelos, L., Ansari, S., Ghanbari, H., & Nenadic, I. (2023) · Journal of Electrocardiology · doi:10.1016/j.jelectrocard.2023.04.004 · Cited by 15
Validation of a Deep Learning Model for Detecting Chest Pathologies from Digital Chest Radiographs
Ajmera, P., Onkar, P. M., Desai, S., Pant, R., Seth, J., Gupte, T., Kulkarni, V., Kharat, A. T., Passi, N., Khaladkar, S. M., et al. (2023) · Diagnostics · doi:10.3390/diagnostics13030557 · Cited by 12
Automatic ARDS surveillance with chest X-ray recognition using convolutional neural networks
Ye, R. Z., Lipatov, K., Diedrich, D. A., Bhattacharyya, A., Erickson, B. J., Pickering, B. W., & Herasevich, V. (2024) · Journal of Critical Care · doi:10.1016/j.jcrc.2024.154794 · Cited by 17
A ensemble methodology for automatic classification of chest X-rays using deep learning
Vogado, L. H. S., Araújo, F. H. D., Neto, P. D. A. D. S., de Almeida, J. D. S., Tavares, J. M. R. S., & Veras, R. D. M. S. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105442 · Cited by 12
“[17] 2019 DenseNet121 216,431 Private Acc of 91%”
MethodDeep Learning for Triage of Chest Radiographs: Should Every Institution Train Its Own System?
van Ginneken, B. (2018) · Radiology · doi:10.1148/radiol.2018182318 · Cited by 13
Multimodal Data Fusion Integrating Text and Medical Imaging Data in Electronic Health Records
Rele, M., Julian, A., Patil, D., & Udaya Krishnan, M. (2024) · Information systems engineering and management · doi:10.1007/978-3-031-69197-3_27 · Cited by 15
Liu, G. S., Shenson, J. A., Farrell, J., & Blevins, N. H. (2023) · Journal of Biomedical Optics · doi:10.1117/1.jbo.28.1.016004 · Cited by 11
“29,30 Again, the inputs to ARRInet-W and ARRInet-M were image patches with spatial dimensions of 32 × 32 pixels and spectral dimension of either three channels for inputs to ARRInet-W (three RGB channels for broad white light illumination only) or 18 channels for inputs to ARRInet-M (three RGB…”
MethodDunnmon, J. (2021) · Radiologic Clinics of North America · doi:10.1016/j.rcl.2021.07.006 · Cited by 10
Self-citationKumar, G., Sharma, N., & Paul, A. (2023) · Medical Physics · doi:10.1002/mp.16722 · Cited by 10
Humphries, M. P., Kaye, D., Stankeviciute, G., Halliwell, J., Wright, A. I., Bansal, D., Brettle, D. S., & Treanor, D. E. (2024) · The Journal of Pathology · doi:10.1002/path.6326 · Cited by 12
Kougia, V., Pavlopoulos, J., Papapetrou, P., & Gordon, M. (2021) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocab046 · Cited by 10
“Many works perform binary classification by employing pretrained convolutional neural networks (CNNs), eg, DenseNet-121(19) and VGG-19,(20) and report high scores.(21,22) As they mention, these methods can also be used for exam prioritization.”
MethodEllis, R., Ellestad, E., Elicker, B. M., Hope, M. D., & Tosun, D. (2020) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2020.103699 · Cited by 11
Mallon, D. H., McNamara, C., Rahmani, G., O’Regan, D. P., & Amiras, D. (2022) · Clinical Radiology · doi:10.1016/j.crad.2022.06.011 · Cited by 11
Sex estimation from Thai hand radiographs using convolutional neural networks
Nonthasaen, P., Mahikul, W., Chobpenthai, T., & Achararit, P. (2023) · Forensic Science International Reports · doi:10.1016/j.fsir.2023.100332 · Cited by 11
Ghosh, T., Tanwar, S., Chumber, S., & Vani, K. (2021) · The Egyptian Journal of Radiology and Nuclear Medicine · doi:10.1186/s43055-021-00499-w · Cited by 8
“While there has been considerable interest in the application of convolutional neural networks and other forms of machine learning for classification of chest radio-graphs into various pathologies, the underlying technology utilized in all these studies remain exclusionary [5, 24 – 26].”
BackgroundAugmenting existing deterioration indices with chest radiographs to predict clinical deterioration
Mu, E., Jabbour, S., Dalca, A. V., Guttag, J. V., Wiens, J., & Sjoding, M. W. (2022) · PLoS ONE · doi:10.1371/journal.pone.0263922 · Cited by 8
Kim, M., Park, S.-K., Kubota, Y., Lee, S., Park, K., & Kong, D. (2022) · PLoS ONE · doi:10.1371/journal.pone.0276378 · Cited by 7
Mosquera, C., Binder, F., Díaz, F. N., Seehaus, A., Ducrey, G., Ocantos, J., Aineseder, M., Rubin, L., Rabinovich, D., Quiroga, A. E., et al. (2021) · Intelligence-Based Medicine · doi:10.1016/j.ibmed.2021.100039 · Cited by 7
High performance on-demand de-identification of a petabyte-scale medical imaging data lakePreprint
Mesterhazy, J., Olson, G., & Datta, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.01827 · Cited by 16
Vasconcelos, L., Martínez, B. O. P., Kent, M., Ansari, S., Ghanbari, H., & Nenadic, I. (2023) · Journal of Electrocardiology · doi:10.1016/j.jelectrocard.2023.09.010 · Cited by 6
Oyekan, B. (2024) · Journal of Knowledge Learning and Science Technology · doi:10.60087/jklst.vol3.n3.p139-164 · Cited by 7
Preservation of High Frequency Content for Deep Learning-Based Medical Image Classification
McIntosh, D., Marques, T. P., & Albu, A. B. (2021) · Canadian Conference on Computer and Robot Vision · doi:10.1109/crv52889.2021.00010 · Cited by 7
“A number of recent works indicate the growing interest in performing the automatic diagnosis of radiographs using computer vision [5]–[10].”
BackgroundMethodHanif, A., Potter, İ. Y., Tian, P., Kalkanlı, B., Deniz Erdogmus, B., Ioannidis, S., Dy, J., Kalpathy–Cramer, J., Ostmo, S. R., Jonas, K. E., et al. (2022) · Ophthalmology Science · doi:10.1016/j.xops.2022.100122 · Cited by 8
Artificial Intelligence Tracking of Otologic Instruments in Mastoidectomy Videos
Liu, G. S., Parulekar, S., Lee, M. C., Chemaly, T. E., Diop, M., Park, R., & Blevins, N. H. (2024) · Otology & Neurotology · doi:10.1097/mao.0000000000004330 · Cited by 6
Li, Q., Chen, M., Geng, J., Adamu, M. J., & Guan, X. (2023) · Diagnostics · doi:10.3390/diagnostics13132165 · Cited by 5
“However, the analysis and interpretation of the numerous chest radiographs generated worldwide heavily rely on human visual examination, which presents challenges such as the need for specialized skills, concentration, time consumption, high costs, potential operator bias, and the inability to leverage valuable inform…”
BackgroundArcher, H., Xia, S., Salzlechner, C., Götz, C., & Chhabra, A. B. (2024) · Seminars in Roentgenology · doi:10.1053/j.ro.2024.06.001 · Cited by 6
Bermúdez, A., Calderon-Ramirez, S., Thang, T., Tyrrell, P. N., Moemeni, A., Yang, S., & Torrents‐Barrena, J. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9206779 · Cited by 5
“We stress the importance of analysing statistically the obtained results in order to reach rigorous and accurate conclusions, as also seen in [12], [5].”
ResultTariq, A., Ramasamy, G., Patel, B. N., & Banerjee, I. (2024) · Journal of medical imaging · doi:10.1117/1.jmi.11.6.064003 · Cited by 7
Al-Azzawi, H. N. T., Ghandour, A., Haider, A., Azar, A. T., Althuniyan, N., Ibraheem, I. K., Hammadi, Y. I., Humaidi, A. J., Haider, Z., & Ahmed, S. (2025) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.9788 · Cited by 4
Wu, Y., Fong, S. J., & Liu, L. (2025) · Soft Computing · doi:10.1007/s00500-025-10644-6 · Cited by 7
“Dunnmon et al. (2019) and Rajpurkar et al. (2018) both used class activation mapping (CAM) to validate their deep learning models developed for interpreting chest radio-graphs.”
Rhee, C., Hong, K. J., Kim, K. H., Goo, J. M., & Hwang, E. J. (2025) · Korean Journal of Radiology · doi:10.3348/kjr.2025.0237 · Cited by 4
Chen, K., Kuo, M., Lee, C., Liao, H.-C., Tsai, D., Lin, S.-A., Hsiang, C.-W., Chang, C.-K., Ko, K., Hsu, Y., et al. (2025) · Journal of Medical Systems · doi:10.1007/s10916-025-02263-3 · Cited by 6
A Comparative Study of Deep Learning Models for Detecting Pulmonary Embolism
Varshney, A., Bansal, A., Agarwal, A., Mishra, V. K., & Badal, T. (2022) · Communications in computer and information science · doi:10.1007/978-3-030-95502-1_7 · Cited by 4
Sudarshan, V. K., Ramachandra, R. A., Tan, N. S. M., Ojha, S., & Tan, R. (2022) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.22715 · Cited by 5
Hong, W., Hwang, E. J., Park, C. M., & Goo, J. M. (2023) · Korean Journal of Radiology · doi:10.3348/kjr.2023.0255 · Cited by 5
“Owing to its excellent performance in experimental studies [1-6], AI is increasingly used in daily practice [7-9].”
BackgroundDetection of Tuberculosis Based on Hybridized Pre-Processing Deep Learning Method
Elashmawy, M. A., Elamvazuthi, I., Izhar, L. I., Paramasivam, S., & Su, S. W. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.0140808 · Cited by 4
“[17] did not rely on augmented photos to increase the size of the dataset, the high accuracy and AUC were probably caused by the numerous training images employed.”
BackgroundEnsemble-based deep meta learning for medical image segmentation
Ahmed, U., Lin, J. C., & Srivastava, G. (2021) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-219221 · Cited by 6
MSA-Net: multi-scale attention-based DenseNet for multi-label chest X-ray image classification
Chen, C., Isa, N. A. M., Liu, X., Ding, J., & Lu, L. C. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.109069 · Cited by 3
Wu, Y., Fong, S. J., & Liu, L. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05916-x · Cited by 6
“Dunnmon JA. et al. [17] trained a series of Convolutional Neural Networks, including the Resnet-18, DenseNet121 and AlexNet, for the classification of normal and abnormal of chest x-ray images.”
Jeong, H., Kim, H., Yoon, J., Go, K., & Gwak, J. (2022) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.22742 · Cited by 5
“They claimed that scratch train overperformed compared to the transfer learning using pre-trained model architecture although Tang et al. 56 and Dunnmon et al. 26 reported that the pre-training has better performance than scratch training.”
MethodApport de l’intelligence artificielle dans le post-traitement de l’imagerie thoracique
Greniér, P., Brun, A., & Mellot, F. (2023) · Revue des Maladies Respiratoires · doi:10.1016/j.rmr.2023.12.001 · Cited by 4
Adam, K., Mohd, I. I., & Ibrahim, Y. (2021) · International journal of electrical and computer engineering systems · doi:10.32985/ijeces.12.2.4 · Cited by 4
“In another study by Dunnmon et al. [26], three CNN models (Alexnet, ResNet, and DenseNet201) were used to classify chest radiographs into groups categorized as either normal or abnormal.”
MethodA Hybridized Pre-Processing Method for Detecting Tuberculosis using Deep Learning
Elashmawy, A. M. A., Elamvazuthi, I., Ali, S. S. A., Natarajan, E., & Paramasivam, S. (2021) · International Conference on Intelligent and Advanced Systems · doi:10.1109/icias49414.2021.9642622 · Cited by 3
Latif, J., Garcea, G., & Dennison, A. R. (2025) · Journal of Robotic Surgery · doi:10.1007/s11701-025-02617-6 · Cited by 2
“CNNs excel in analysing medical images, detecting patterns like tumours in radiology scans for applications such as malignancy prediction or diabetic retinopathy detection [16–20].”
YOLOv11-MFF: A multi-scale frequency-adaptive fusion network for enhanced CXR anomaly detection
Guan, L., Zhang, R., & Zhao, Y. M. (2025) · PLoS ONE · doi:10.1371/journal.pone.0334283 · Cited by 2
Li, B., Zhao, K., Liu, A., Xu, W., Tang, Y., Yang, K., Dai, L., Chen, X., Zhao, S., & Yan, C. (2025) · Open Heart · doi:10.1136/openhrt-2025-003519 · Cited by 2
Lee, H., Oh, J., Lee, J.-Y., Tae-Soo, L., & Park, H.-H. (2020) · Bangsaseon gisul gwahak/Journal of radiological science and technology · doi:10.17946/jrst.2020.43.2.65 · Cited by 2
Rocha, J., Mendonça, A. M., & Campilho, A. (2021) · U Porto Journal of Engineering · doi:10.24840/2183-6493_007.004_0002 · Cited by 2
“Current standard off-the-shelf CNN-based methods are frequently applied to detect abnormalities in CXR, and there are several papers which establish a comparison between well - known architectures, as illustrated in Tang et al. (2020) and Dunnmon et al. (2019).”
MethodArtificial Intelligence in Cardiopulmonary Imaging
Chan, J., & Auffermann, W. F. (2020) · Advances in Clinical Radiology · doi:10.1016/j.yacr.2020.04.006 · Cited by 2
Classificação de Distúrbios Pulmonares em Radiografias de Tórax Usando Redes Convolucionais
Machado, A., Araújo, L., & Veloso, L. R. (2021) · Anais de XXXV Simpósio Brasileiro de Telecomunicações e Processamento de Sinais · doi:10.14209/sbrt.2021.1570734174 · Cited by 3
Decomposition Method for Calculating the Weights of a Binary Neural Network
Litvinenko, A. E., Kucherov, D., & Glybovets, M. (2022) · Cybernetics and Systems Analysis · doi:10.1007/s10559-023-00522-0 · Cited by 2
“The problem of binary classification is not trivial in case where the objects are satellite [10] or X-ray images [13, 14].”
BackgroundSorace, L., Raju, N., O'Shaughnessy, J., Kachel, S., Jansz, K., Yang, N., & Lim, R. (2023) · Radiography · doi:10.1016/j.radi.2023.10.014 · Cited by 3
Automatic Detection of Tuberculosis from Chest X-Rays using Convolutional Neural Network
Satheeshkumar, K. G., & Arunachalam, V. (2020) · International Journal of Engineering and Advanced Technology · doi:10.35940/ijeat.e9292.069520 · Cited by 2
“…the quality metric scores of the proposed architecture with very recent architectures, it can be seen that the scores are comparable and in certain cases better than the average score(accuracy) of 89% obtained on three CNN architecture AlexNet, ResNet and DenseNet used by Jared A, et al ., [26].”
A deep learning approach to image quality assessment
Feng, Y. (2019) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/85493 · Cited by 2
Wolf, D., Payer, T., Lisson, C. S., Lisson, C. G., Beer, M., GOETZ, M. A., & Ropinski, T. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2308.06534 · Cited by 3
“Deep learning models trained on medical images have recently demonstrated diagnostic accuracy comparable to human experts for narrow clinical tasks 2–5 .”
Vipul.V.Bag (2024) · Journal of Electrical Systems · doi:10.52783/jes.1441 · Cited by 1
“[25 ] Now days the availability of infection detection testing kits for Corona virus disease 2019 is limited.”
BackgroundPatient-Facing Radiology Communication with LLMs: Calibration Deficit and the Metadata Paradox
Shin, C., Park, J. H., Kim, S. J., Lee, Y. H., & Lee, H. S. (2026) · Healthcare · doi:10.3390/healthcare14111490 · Cited by 2
“To ensure clinical generalizability, the dataset incorporated cases from outpatient, emergency, and inpatient settings, reflective of real-world practice [27].”
Li, F., Armato, S. G., Engelmann, R. M., Rhines, T., Crosby, J., Li, L., Giger, M. L., & MacMahon, H. (2021) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-021-00494-7 · Cited by 1
Mwaniki, P., Kamanu, T., Akech, S., & Eijkemans, M. J. C. (2022) · Wellcome Open Research · doi:10.12688/wellcomeopenres.17164.2 · Cited by 1
“Similar to other studies, all models were fitted using CXR images down-sampled to 224 by 224 pixels (Dunnmon et al ., 2018; Rajpurkar et al ., 2017; Wang & Xia, 2018).”
MethodThe role of artificial intelligence in the differential diagnosis of wheezing symptoms in children
Song, L., Zhu, Z., Hu, G., Sui, X., Song, W., & Jin, Z. (2022) · Radiology Science · doi:10.15212/radsci-2022-0002 · Cited by 1
“In deep learning-based medical image diagnosis, CAM and its variants have been widely used for visual interpretation through generating pixel-level diagnostic models for overall interpretability, thus greatly improving the diagnostic accuracy of AI models [61-65].”
MethodChest Disease Classification Using Convolutional Neural Networks
Kavitha, S., Thaarani, S., Singh, A. P., & Santhosh, G. (2023) · International Conference on Computer Communication and Informatics (ICCCI) · doi:10.1109/iccci56745.2023.10128179 · Cited by 2
“[5-9] presented the Deep Convolutional Neural Networks for Learning to Automatically Predict Multiple Diseases in Pediatric Chest Radiographs.”
BackgroundLesion-Aware Chest Radiography Abnormality Classification with Object Detection Framework
Pedrosa, J., Sousa, P., Silva, J., Mendonça, A. M., & Campilho, A. (2023) · Proceedings - IEEE Symposium on Computer-Based Medical Systems · doi:10.1109/cbms58004.2023.00324 · Cited by 2
Removing Radiographic Markers Using Deep Learning to Enable Image Sharing
Chang, K., & Li, M. (2023) · Radiology Artificial Intelligence · doi:10.1148/ryai.230369 · Cited by 1
Putra, A. E., Kartini, K., & Sari, A. P. (2024) · JASIEK (Jurnal Aplikasi Sains Informasi Elektronika dan Komputer) · doi:10.26905/jasiek.v6i1.11464 · Cited by 1
“Pengambilan ciri dilakukan melalui model CNN, yang terdiri dari tiga tahap konvolusi, satu tahap perataan (flatten), dan dua tahap lapisan terhubung sepenuhnya ( fully connected layer ) [12].”
Sonawane, S., & Puthran, S. (2024) · Innovations in Systems and Software Engineering · doi:10.1007/s11334-024-00592-z · Cited by 4
Intelligent Automation System for Vessels Recognition: Comparison of SIFT and SURF Methods
Musulin, J., Lorencin, I., Meštrić, H., & Car, Z. (2021) · Tehnicki vjesnik - Technical Gazette · doi:10.17559/tv-20200522115821 · Cited by 1
“Dunnmon et al. (2019) show the impact of dataset size on classification performance for Automated Classification of Chest Radiographs [29].”
BackgroundBenjamin, M., Engelhard, G., Aisen, A. M., Aradi, Y., & Benjamin, E. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.13781 · Cited by 3
“al Intelligence (AI) and Machine learning (ML) in radiology presents many challenges -3]. Development of effective algorithms requires [1 enormous amounts of data for training, testing and validation [4, 5, 6]. The best data is comprised of volumes of imaging studies collected from diverse sites, modality vendors, pro…”
BackgroundClass activation mapping (CAM)
Moore, C. (2019) · Radiopaedia.org · doi:10.53347/rid-72380 · Cited by 1
Self-supervised Learning for Chest CT - Training Strategies and Effect on Downstream ApplicationsPreprint
Tariq, A., Patel, B. N., & Banerjee, I. (2024) · medRxiv · doi:10.1101/2024.02.01.24302144 · Cited by 1
“D EEP learning models have been proposed and trained for high-dimensional medical imaging data, such as CT or MRI volumes, to target a variety of image interpretation tasks, and have demonstrated diagnostic accuracy comparable to that of human experts [11], [18], [21], [24].”
BackgroundReducing Labelled Data Requirement for Pneumonia Segmentation Using Image Augmentations
Seth, J., Lokwani, R., Kulkarni, V., Pant, A., & Kharat, A. T. (2022) · Lecture notes in networks and systems · doi:10.1007/978-981-16-5987-4_29
Luo, L., Chen, H., Xiao, Y., Zhou, Y., Wang, X., Vardhanabhuti, V., Wu, M., & Heng, P. (2021) · arXiv (Cornell University) · Cited by 1
Weakly Supervised Classification and Localization of Thorax Diseases on X-Ray images
Jose, A. (2021) · UVic’s Research and Learning Repository (University of Victoria) · Cited by 1
Thoracic Image Analysis Using Deep Learning
Wajgi, R. D., Tembhurne, J. V., & Wajgi, D. W. (2021) · Wiley eBooks · doi:10.1002/9781119818717.ch2 · Cited by 1
Mwaniki, P., Kamanu, T., Akech, S., & Eijkemans, M. J. C. (2021) · Wellcome Open Research · doi:10.12688/wellcomeopenres.17164.1
Classificação dos usos do solo no Douro através de redes neuronais e métodos probabilísticos
Matos, J., Fernandes, D., Ramos, L. C. V., & Lourenço, J. (2006) · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT)
Han, S. J., & Yi, P. H. (2023) · Journal of Thoracic Imaging · doi:10.1097/rti.0000000000000755
“Prior work has found that increasing training-set size beyond 20,000 images yields marginal bene fi t for DCNNs trained to detect abnormal CXRs, but data set size requirements for other less complex tasks like semantic labeling of images are unknown, 4,5 and which may be lower.”
BackgroundAN ALGEBRAIC METHOD FOR SYNTHESIZING ERROR-FREE BINARY NEURAL NETWORK
Litvinenko, A. L. (2024) · Kibernetyka ta Systemnyi Analiz · doi:10.34229/kca2522-9664.24.3.2
Detection and Classification of Chest Diseases using Machine Learning Algorithm
Aggarwal, R., Gupta, R., Jain, C., & Kumar, A. (2023) · International Conference on Information Management & Machine Intelligence (ICIMMI) · doi:10.1145/3647444.3647947
“During the classification stage, the COVID-19 status of the X-ray images was found by feeding the chosen features into a CNN. Dunnmon et al. (2019) evaluated CNN’s performance in automatic categorization of chest radiographs [34].”
MethodAn Algebraic Method for the Synthesis of Error-Free Binary Neural Network
Litvinenko, A. E. (2024) · Cybernetics and Systems Analysis · doi:10.1007/s10559-024-00676-5
Evaluating Local Explainable AI Techniques for the Classification of Chest X-Ray Images
Sciacca, E., Estatico, C., Verda, D., & Ferrari, E. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-63803-9_4 · Cited by 2
Pugh, S., Ruchkin, I., Weimer, J. E., & Lee, I. (2025) · ACM Transactions on Cyber-Physical Systems · doi:10.1145/3734695 · Cited by 1
“The medical case studies include five clinical alarm datasets derived from a 551-hour alarm labeling effort from Children’s Hospital of Philadelphia, and one radiography imaging reports dataset from [9].”
Metadata
- Authors
- Dunnmon, J. A., Yi, D., Langlotz, C. P., Ré, C., Rubin, D. L., & Lungren, M. P.
- Affiliations
- Stanford University (US)
- ORCID
- Venue
- Radiology
- Published
- 2019-02
- Volume
- 290
- Issue
- 2
- Pages
- 537-544
- ISSN
- 0033-8419 (ISSN-L)
- 1527-1315
- PMC
- PMC6358056
- OpenAlex
- W2901782819
- Semantic Scholar
- 3f3ecb0b178b2df70c3f1e90684895ef82a3f836
- Cited by
- 206
- References
- 24
- Citations by year
- 2026: 4 · 2025: 16 · 2024: 24 · 2023: 38 · 2022: 26 · 2021: 39 · 2020: 37 · 2019: 20
- FWCI
- 16.73
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Radiology (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Intelligence Community Postdoctoral Research Fellowship Program
- National Institutes of Health (1U01CA187947, 1U01CA190214, U01CA142555)
- National Cancer Institute
- Office of Naval Research (N000141712266)
- Defense Advanced Research Projects Agency (FA87501720095)
- 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
- convolutional neural networks, chest radiograph, automatic classification, normal vs abnormal classification, training data size, network architecture, activation visualization, error analysis, chest diseases
- MeSH terms
- Radiologists, Female, Humans, Lung, Male, Radiographic Image Interpretation, Computer-Assisted, Retrospective Studies, ROC Curve, Radiography, Thoracic, Neural Networks, Computer
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.