Radiomic versus Convolutional Neural Networks Analysis for Classification of Contrast-enhancing Lesions at Multiparametric Breast MRI
The DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Abstract
Purpose To compare the diagnostic performance of radiomic analysis (RA) and a convolutional neural network (CNN) to radiologists for classification of contrast agent-enhancing lesions as benign or malignant at multiparametric breast MRI. Materials and Methods Between August 2011 and August 2015, 447 patients with 1294 enhancing lesions (787 malignant, 507 benign; median size, 15 mm ± 20) were evaluated. Lesions were manually segmented by one breast radiologist. RA was performed by using L1 regularization and principal component analysis. CNN used a deep residual neural network with 34 layers. All algorithms were also retrained on half the number of lesions (n = 647). Machine interpretations were compared with prospective interpretations by three breast radiologists. Standard of reference was histologic analysis or follow-up. Areas under the receiver operating curve (AUCs) were used to compare diagnostic performance. Results CNN trained on the full cohort was superior to training on the half-size cohort (AUC, 0.88 vs 0.83, respectively; P = .01), but there was no difference for RA and L1 regularization (AUC, 0.81 vs 0.80, respectively; P = .76) or RA and principal component analysis (AUC, 0.78 vs 0.78, respectively; P = .93). By using the full cohort, CNN performance (AUC, 0.88; 95% confidence interval: 0.86, 0.89) was better than RA and L1 regularization (AUC, 0.81; 95% confidence interval: 0.79, 0.83; P < .001) and RA and principal component analysis (AUC, 0.78; 95% confidence interval: 0.76, 0.80; P < .001). However, CNN was inferior to breast radiologist interpretation (AUC, 0.98; 95% confidence interval: 0.96, 0.99; P < .001). Conclusion A convolutional neural network was superior to radiomic analysis for classification of enhancing lesions as benign or malignant at multiparametric breast MRI. Both approaches were inferior to radiologists' performance; however, more training data will further improve performance of convolutional neural network, but not that of radiomics algorithms. © RSNA, 2018 Online supplemental material is available for this article.
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 [104]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
The meaning and use of the area under a receiver operating characteristic (ROC) curve.
Hanley, J. A., & McNeil, B. J. (1982) · Radiology · doi:10.1148/radiology.143.1.7063747 · Cited by 21,856
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362
Radiomics: Images Are More than Pictures, They Are Data
Gillies, R. J., Kinahan, P. E., & Hricak, H. (2016) · Radiology · doi:10.1148/radiol.2015151169 · Cited by 8,421
Computational Radiomics System to Decode the Radiographic Phenotype
(2017) · doi:10.1158/0008-5472.can-17-0339 · Cited by 6,593
N4ITK: Improved N3 Bias Correction
Tustison, N. J., Avants, B. B., Cook, P. A., Zheng, Y., Egan, A., Yushkevich, P. A., & Gee, J. C. (2010) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2010.2046908 · Cited by 6,073
Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach
Aerts, H. J. W. L., Velazquez, E. R., Leijenaar, R. T. H., Parmar, C., Grossmann, P., Carvalho, S., Bussink, J., Monshouwer, R., Haibe-Kains, B., Rietveld, D., et al. (2014) · Nature Communications · doi:10.1038/ncomms5006 · Cited by 5,132
Python for scientific computing
Oliphant, T. E. (2007) · Computing in Science & Engineering · doi:10.1109/mcse.2007.58 · Cited by 3,303
Kriege, M., Brekelmans, C. T. M., Boetes, C., Besnard, P. E., Zonderland, H. M., Obdeijn, I. M., Manoliu, R. A., Kok, T., Peterse, H., Tilanus-Linthorst, M. M. A., et al. (2004) · New England Journal of Medicine · doi:10.1056/nejmoa031759 · Cited by 1,687
Research works printed but not linked: 1
- Goodfellow IJ, 2014, NIPS’14: Proceedings of the 27th International Conference on Neural Information Processing Systems, Volume 2, Montreal, Canada, December 8–13, 2014, 2672
Cited by
Items 101–200 of 296
Preoperative prediction of lymph node metastasis using deep learning-based features
Cattell, R., Ying, J., Lei, L., Ding, J., Chen, S., Sosa, M. S., & Huang, C. (2022) · Visual Computing for Industry Biomedicine and Art · doi:10.1186/s42492-022-00104-5 · Cited by 25
“Using MRI, CNN performance has been compared and shown to outperform radiomics for the purposes of breast lesion classification [23] and gene mutations in low grade gliomas [24].”
MethodGoto, M., Sakai, K., Toyama, Y., Nakai, Y., & Yamada, K. (2023) · Japanese Journal of Radiology · doi:10.1007/s11604-023-01435-w · Cited by 25
Zhou, H., Bai, H. X., Jiao, Z., Cui, B., Wu, J., Zheng, H., Yang, H., & Liao, W. (2023) · European Journal of Radiology · doi:10.1016/j.ejrad.2023.111136 · Cited by 24
Sun, K., Jiao, Z., Zhu, H., Chai, W., Yan, X., Fu, C., Cheng, J., Yan, F., & Shen, D. G. (2021) · Journal of Translational Medicine · doi:10.1186/s12967-021-03117-5 · Cited by 24
Multi‐marker quantitative radiomics for mass characterization in dedicated breast CT imaging
Caballo, M., Pangallo, D. R., Sanderink, W. B. G., Hernandez, A. M., Lyu, S. H., Molinari, F., Boone, J. M., Mann, R. M., & Sechopoulos, I. (2020) · Medical Physics · doi:10.1002/mp.14610 · Cited by 23
Machine Learning Based on Multi-Parametric MRI to Predict Risk of Breast Cancer
Tao, W., Lu, M., Zhou, X., Montemezzi, S., Bai, G., Yue, Y., Li, X., Zhao, L., Zhou, C., & Lu, G. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.570747 · Cited by 23
“used radiomics and deep learning to analyze breast mpMRI but the breast mpMRIs in their study only included T2WI and contrastenhanced subtracted or non-subtracted images (23).”
MethodLiu, C., Qiao, M., Jiang, F., Guo, Y., Jin, Z., & Wang, Y. (2021) · Medical Physics · doi:10.1002/mp.15172 · Cited by 22
Chen, Y., Wang, L., Luo, R., Wang, S., Wang, H., Gao, F., & Wang, D. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.943415 · Cited by 22
“Recently, multiparametric MRI has also been studied with DL (13, 20).”
BackgroundMethodChen, X., Zhang, Y., Zhou, J., Wang, X., Liu, X., Nie, K., Lin, X., He, W., Su, M., Cao, G., et al. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.991892 · Cited by 22
Artificial intelligence in breast imaging: potentials and challenges
Li, J.-W., Sheng, D., Chen, J., You, C., Liu, S., Xu, H., & Chang, C. (2023) · Physics in Medicine and Biology · doi:10.1088/1361-6560/acfade · Cited by 23
Use of artificial intelligence in breast surgery: a narrative review
Seth, I., Lim, B., Joseph, K., Gracias, D., Xie, Y., Ross, R. J., & Rozen, W. M. (2024) · Gland Surgery · doi:10.21037/gs-23-414 · Cited by 22
Alsharif, W. M. (2023) · Saudi Medical Journal · doi:10.15537/smj.2023.44.2.20220611 · Cited by 21
Jaggi, A., Conole, E. L. S., Raisi‐Estabragh, Z., Gkontra, P., McCracken, C., Szabó, L. E., Neubauer, S., Petersen, S. E., Cox, S. R., & Lekadir, K. (2024) · Imaging Neuroscience · doi:10.1162/imag_a_00063 · Cited by 21
“Since principal component analysis (PCA) is commonly used in radiomics to extract lower dimensional representations of the data ( Antonopoulos et al., 2021; Raisi-Estabragh et al., 2020; Truhn et al., 2019; Zhang et al., 2017), we performed PCA on the z-scored radiomics.”
MethodThe Long Route to Standardized Radiomics: Unraveling the Knot from the End
Kuhl, C. K., & Truhn, D. (2020) · Radiology · doi:10.1148/radiol.2020200059 · Cited by 22
Self-citationXiang, Y., Dong, X., Zeng, C., Liu, J., Liu, H., Hu, X., Feng, J., Du, S., Wang, J., Han, Y., et al. (2022) · Frontiers in Immunology · doi:10.3389/fimmu.2022.913703 · Cited by 20
“In general, machine learning can be divided into two major categories: radiomics, where image features are manually extracted, and deep learning (DL), where computers can automatically extract content without handcrafted features but require a larger pool of training images (16, 17).”
BackgroundCao, W., Hu, H., Guo, J., Qin, Q., Lian, Y., Li, J., Wu, Q., Chen, J., Wang, X., & Deng, Y. (2023) · Journal of Translational Medicine · doi:10.1186/s12967-023-04023-8 · Cited by 20
Yin, P., Wang, W., Wang, S., Liu, T., Sun, C., Liu, X., Chen, L., & Hong, N. (2023) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-22-1042 · Cited by 20
Yu, P., Wang, C., Zhang, H., Zheng, G., Jia, C., Liu, Z., Wang, Q., Mu, Y., Yang, X., Mao, N., et al. (2024) · Chinese Journal of Cancer Research · doi:10.21147/j.issn.1000-9604.2024.05.07 · Cited by 20
Huang, Y., Song, X., Chen, Y., Qiu, H., Zhou, T., Wang, S., Zhou, Y., Li, W., Lin, Y., Wang, Q., et al. (2025) · Radiology · doi:10.1148/radiol.243545 · Cited by 19
Perre, S. V., Duron, L., Milon, A., Bekhouche, A., Balvay, D., Cornélis, F., Fournier, L. S., & Thomassin‐Naggara, I. (2021) · European Radiology · doi:10.1007/s00330-020-07519-9 · Cited by 19
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
“radiomic analysis in certain tasks [15].”
Zhou, J., Liu, Y.-L., Zhang, Y., Chen, J., Combs, F. J., Parajuli, R., Mehta, R. S., Liu, H., Chen, Z., Zhao, Y., et al. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.728224 · Cited by 18
“2 lesions with a high accuracy (16, 17), but studies dedicated to the diagnosis of NME were rarely reported.”
BackgroundMethodResultInfluentialRashid, H. U., İbrikçі, T., Paydaş, S., Binokay, F., & Çevik, U. (2022) · Expert Systems · doi:10.1111/exsy.13018 · Cited by 18
“A CNN-based ResNet model was developed by Truhn et al. (2018).”
MethodHigh‐Resolution Breast MRI Reconstruction Using a Deep Convolutional Generative Adversarial Network
Sun, K., Qu, L., Lian, C., Pan, Y., Hu, D., Xia, B., Li, X., Chai, W., Yan, F., & Shen, D. G. (2020) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.27256 · Cited by 17
Germain, P., Vardazaryan, A., Padoy, N., Labani, A., Roy, C., Schindler, T. H., & Ghannudi, S. E. (2021) · Diagnostics · doi:10.3390/diagnostics12010069 · Cited by 17
“This has been demonstrated for renal cancer [24], subtyping different types of cerebral glioma [27], diagnosis of breast cancer [28] and predicting axillary lymph node metastasis of breast cancer [29].”
BackgroundLi, H., Zhao, Y., Duan, J., Gu, J., Liu, Z., Zhang, H., Zhang, Y., & Li, Z. (2024) · Displays · doi:10.1016/j.displa.2024.102698 · Cited by 16
Zhou, J., Zhang, Y., Miao, H., Yoon, G. Y., Wang, J., Lin, Y., Wang, H., Liu, Y., Chen, J., Pan, Z., et al. (2024) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.29447 · Cited by 16
Shen, L., Dai, B., Dou, S., Yan, F., Yang, T., & Wu, Y. (2025) · BMC Cancer · doi:10.1186/s12885-025-13424-5 · Cited by 16
Liu, Y., Jia, X., Zhao, J., Peng, Y., Yao, X., Hu, X., Cui, J., Chen, H., Chen, X., Wu, J., et al. (2023) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.29111 · Cited by 15
Qi, R., Wu, W., Lu, Z., & Liao, S. (2024) · IEEE Access · doi:10.1109/access.2024.3430983 · Cited by 15
“However, it requires a large amount of annotated data and computational resources and lacks interpretability [38].”
BackgroundVerburg, E., van Gils, C. H., van der Velden, B. H. M., Bakker, M. F., Pijnappel, R. M., Veldhuis, W. B., & Gilhuijs, K. G. A. (2022) · Investigative Radiology · doi:10.1097/rli.0000000000000934 · Cited by 15
Lee, H., Nguyen, A.-T., Ki, S. Y., Lee, J. E., Do, L.-N., Park, M. H., Lee, J. S., Kim, H. J., Park, I., & Lim, H. S. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.744460 · Cited by 15
Khader, F., Müller‐Franzes, G. A., Wang, T., Han, T., Arasteh, S. T., Haarburger, C., Stegmaier, J., Bressem, K. K., Kuhl, C. K., Nebelung, S., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.09162 · Cited by 16
Self-citationWang, W., & Wang, Y. (2023) · Diagnostics · doi:10.3390/diagnostics13091582 · Cited by 14
“They found that despite the adjustment of the ANN method, the accuracy of CNN was still higher than that of ANN [36].”
MethodWang, J., Song, P., Zhang, M., Liu, W., Zeng, X., Chen, N., Li, Y., & Wang, M. (2024) · Cancer Medicine · doi:10.1002/cam4.70046 · Cited by 14
“34,35 This consistency highlights the feasible and relative superiority of DL models in distinguishing MSI and MSS in patients with EC.”
BackgroundSalmanpour, M. R., Amiri, S., Gharibi, S., Shariftabrizi, A., Xu, Y., Weeks, W. B., Rahmim, A., & Hacihaliloglu, I. (2025) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-025-01585-5 · Cited by 14
“Recent studies emphasize the integration of radiological and biological data to enhance the interpretability and relevance of RFs, thereby improving the explainability and trust in AI models used in clinical settings [13, 40, 41, 43–45].”
Li, W., Li, Y., Gao, S., Huang, N. K., Kojima, I., Kusama, T., Ou, Y., Iikubo, M., & Niu, X. (2024) · BMC Cancer · doi:10.1186/s12885-024-12533-x · Cited by 13
“These models, which use self-attention mechanisms, are adept at handling the complexity and variability inherent in medical images [13].”
BackgroundLi, C., Guo, Y., Jia, L., Yao, M., Shao, S., Chen, J., Xu, Y., & Wu, R. (2022) · Frontiers in Physiology · doi:10.3389/fphys.2022.882648 · Cited by 13
Sakaida, M., Yoshimura, T., Tang, M., Ichikawa, S., & Sugimori, H. (2023) · Algorithms · doi:10.3390/a16100483 · Cited by 13
Preoperative Prediction Power of Radiomics for Breast Cancer: A Systemic Review and Meta-Analysis
Li, Z., Ye, J., Du, H., Cao, Y., Wang, Y., Liu, D., Zhu, F., & Shen, H. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.837257 · Cited by 13
“…and 1 radiomics feature Chen,2019 (12) Retrospective China 81 Conventional algorithm Functional image MMG, MRI 40 41 33 8 7 33 14 radiomic features Truhn, 2019 (26) Retrospective Germany 447 Conventional algorithm Functional image MRI 787 507 616 78 171 429 10 radiomic features Lei, 2019 (18)…”
BackgroundMethodMachine Learning for Hepatocellular Carcinoma Segmentation at MRI:RadiologyIn Training
Raman, A. G., Jones, C. K., & Weiss, C. R. (2022) · Radiology · doi:10.1148/radiol.212386 · Cited by 13
Prinzi, F., Militello, C., Zarcaro, C., Bartolotta, T. V., Gaglio, S., & Vitabile, S. (2025) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2024.108576 · Cited by 12
High-Performance Open-Source AI for Breast Cancer Detection and Localization in MRI
Hirsch, L., Sutton, E. J., Huang, Y., Kayis, B., Hughes, M. C., Martinez, D. F., Makse, H. A., & Parra, L. C. (2025) · Radiology Artificial Intelligence · doi:10.1148/ryai.240550 · Cited by 12
A fully-automated computer-aided breast lesion detection and classification system
Mutlu, F., Çetinel, G., & Gül, S. (2020) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2020.102157 · Cited by 12
Mammogram data analysis: Trends, challenges, and future directions
Karthikeyan, V., Karuppiah, G., Vadivel, S. R. S., & Joseph, D. R. V. (2024) · Elsevier eBooks · doi:10.1016/b978-0-443-13999-4.00016-x · Cited by 12
Wang, H., van der Velden, B. H. M., Ragusi, M. A. A., Veldhuis, W. B., Viergever, M. A., Verburg, E., & Gilhuijs, K. G. A. (2021) · Investigative Radiology · doi:10.1097/rli.0000000000000759 · Cited by 12
Lyu, Y., Chen, Y., Meng, L., Guo, J., Zhan, X., Chen, Z., Yan, W., Zhang, Y., Zhao, X., & Zhang, Y. (2023) · Frontiers in Oncology · doi:10.3389/fonc.2023.1074060 · Cited by 12
Rad-EfficientNet: Improving Breast MRI Diagnosis Through Integration of Radiomics and Deep Learning
Georgas, K., Vezakis, I., Kakkos, I., Douma, N., Panourgias, E. C., Moulopoulos, L. A., & Matsopoulos, G. K. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3551840 · Cited by 11
“Truhn et al. (2019) [23] found that combining deep learning with radiomic features is a promising way to improve classification performance.”
Deep learning model to predict Ki-67 expression of breast cancer using digital breast tomosynthesis
Oba, K., Adachi, M., Kobayashi, T., Takaya, E., Shimokawa, D., Fukuda, T., Takahashi, K., Yagishita, K., Ueda, T., & Tsunoda, H. S. (2024) · Breast Cancer · doi:10.1007/s12282-024-01549-7 · Cited by 11
“Currently, the emphasis is on the potential usefulness of applying deep learning (DL) models, especially convolutional neural networks (CNN), for clinical imaging, such as the benign and malignant classification of breast lesions using MRI [13].”
Li, H., Whitney, H. M., Ji, Y., Edwards, A. V., Papaioannou, J., Liu, P., & Giger, M. L. (2022) · Journal of medical imaging · doi:10.1117/1.jmi.9.3.034502 · Cited by 10
“Truhn et al. 46 also reported a multiparametric breast MRI study using both conventional radiomic and convolutional neural networks (CNN) analysis for the classification of contrast-enhancing lesions.”
BackgroundStrotzer, Q., Wagner, T., Angstwurm, P., Hense, K., Scheuermeyer, L., Noeva, E., Dinkel, J. G., Stroszczynski, C., Fellner, C., Riemenschneider, M. J., et al. (2024) · Neuro-Oncology Advances · doi:10.1093/noajnl/vdae060 · Cited by 10
“35,36 Good results in differentiating brain metastases from pathological lung cancer types were found for both approaches.”
Kudus, K., Wagner, M. W., Namdar, K., Bennett, J., Nobre, L. F., Tabori, U. Y., Hawkins, C. E., Ertl-Wagner, B. B., & Khalvati, F. (2024) · Scientific Reports · doi:10.1038/s41598-024-69870-x · Cited by 10
“However, CNNs have limitations of their own, namely, they need large datasets to learn from 20,23,24 .”
BackgroundGu, Q., Sun, H., Liu, P., Hu, X., Yang, J., Chen, Y., & Xing, Y. (2025) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2025.110770 · Cited by 10
Mutasa, S., Chang, P., Nemer, J. S., Sant, E. P. V., Sun, M., McIlvride, A., Siddique, M., & Ha, R. S. (2020) · Clinical Breast Cancer · doi:10.1016/j.clbc.2020.06.001 · Cited by 10
Xu, P., Yu, H., Xing, W., Zhang, S., Hu, H., Li, W., Jia, D. T., Zhi, S., & Peng, X. (2025) · Frontiers in Oncology · doi:10.3389/fonc.2025.1572720 · Cited by 9
Guan, X., Liu, J., Xu, L., Jiang, W., & Wang, C. (2025) · Annals of Surgical Oncology · doi:10.1245/s10434-025-17748-1 · Cited by 9
“11,12 Despite these advances, few studies have explored the application of deep learning radiomics in predicting ER in resectable PC, likely owing to the limited availability of annotated datasets.”
Tang, W., Zhang, M., Xu, C., Shao, Y., Tang, J., Gong, S., Dong, H., & Sheng, M. (2023) · PeerJ Computer Science · doi:10.7717/peerj-cs.1460 · Cited by 9
Zheng, Y.-M., Pang, J., Liu, Z.-J., Yuan, M.-G., Li, J., Wu, Z., Jiang, Y., & Dong, C. (2023) · Academic Radiology · doi:10.1016/j.acra.2023.06.026 · Cited by 9
Tong, Y., Hu, Z., Wang, H., Huang, J., Zhan, Y., Chai, W., Deng, Y., Yuan, Y., Shen, K., Wang, Y., et al. (2024) · European Radiology · doi:10.1007/s00330-024-10609-7 · Cited by 9
“Data augmentation methods are often applied to solve these problems, including flip, rotate, crop, etc. [26].”
MethodHuang, X., Huang, Y., Liu, K., Zhang, F., Zhu, Z., Xu, K., & Li, P. (2024) · npj Precision Oncology · doi:10.1038/s41698-024-00688-6 · Cited by 9
Feng, B., Liu, Z., Liu, Y., Chen, Y., Zhou, H., Cui, E., Li, X., Chen, X., Li, R., Yu, T., et al. (2022) · Frontiers in Oncology · doi:10.3389/fonc.2022.890659 · Cited by 9
Artificial Intelligence in Breast Imaging
Wang, X., Moriakov, N., Gao, Y., Zhang, T., Han, L., & Mann, R. M. (2022) · Medical radiology · doi:10.1007/978-3-030-94918-1_20 · Cited by 9
Application of Machine Learning to Breast MR Imaging
Gullo, R. L., van Veldhuizen, V., Roa, T., Kapetas, P., Teuwen, J., & Pinker, K. (2025) · Magnetic Resonance in Medical Sciences · doi:10.2463/mrms.rev.2025-0021 · Cited by 8
Role of Radiology in the Diagnosis and Treatment of Breast Cancer in Women: A Comprehensive Review
Arslan, M., Asim, M., Sattar, H., Khan, A., Ali, F. T., Zehra, M., & Talluri, K. (2024) · Cureus · doi:10.7759/cureus.70097 · Cited by 8
“Researchers demonstrated AUC values ranging from 0.84 to 0.92 in a study that used the three RGB channels and the inclusion of many post-contrast slices, namely slices [33,34].”
InfluentialYang, Y., Cheng, J. Z., Cui, C., Huang, H., Cheng, M., Wang, J., & Zuo, M. (2025) · Acta Oncologica · doi:10.2340/1651-226x.2025.42982 · Cited by 8
“Radiomics has become an emerging field of precision medicine [13].”
Yin, P., Sun, C., Wang, S., Chen, L., & Hong, N. (2021) · Frontiers in Oncology · doi:10.3389/fonc.2021.752672 · Cited by 8
“Some previous studies have also shown that deep learning performs better than RN in tumor classification and prognosis (33, 34).”
BackgroundShah, U., Biswas, M. R., Alzubaidi, M. S., Ali, H., Alam, T., Househ, M. S., & Shah, Z. (2022) · Studies in health technology and informatics · doi:10.3233/shti210911 · Cited by 8
“However, CNN were the prominent feature extraction techniques regardless of the imaging modility [10].”
MethodLiu, Y., Tong, Y., Wan, Y., Xia, Z., Yao, G., Shang, X., Huang, Y., Chen, L., Chen, D. Q., & Liu, B. (2023) · Frontiers in Oncology · doi:10.3389/fonc.2023.1119743 · Cited by 8
Meng, N., Feng, P., Yu, X., Wu, Y., Fu, F., Li, Z., Luo, Y., Tan, H., Yuan, J., Yang, Y., et al. (2023) · European Radiology · doi:10.1007/s00330-023-09978-2 · Cited by 8
“…while radiomics features based on different imaging methods have their own characteristics, and different diseases also possess relatively specific patho-physiological features, which together lead to a single classifier that may have difficulty meeting the needs of different scenarios [45].”
Parmar, A., Qazi, A. A., Stundzia, A., Sim, H., Lewin, J. H., Metser, U., O'Malley, M. E., & Hansen, A. R. (2021) · Canadian Urological Association Journal · doi:10.5489/cuaj.7294 · Cited by 8
Interpretable deep learning for multicenter gastric cancer T staging from CT images
Zheng, G., Wang, H., Chai, X., Xin, X., Li, F., Li, H., Ban, Y., Wang, J., Qi, X., Li, Y., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02002-5 · Cited by 7
“Furthermore, successful multi-center AI implementations in other medical domains — such as breast MRI 9 , lung cancer EGFR status prediction 10 , and peritoneal metastasis detection in colorectal cancer 11 — demonstrate that models combining large-scale heterogeneous data with modern convolutional…”
InfluentialHu, Y., Cai, Z., AiErken, N., Liu, Y., Shao, N., Shi, Y., Zhang, M., Hu, Y., Zhang, X., & Lin, Y. (2025) · Radiation Oncology · doi:10.1186/s13014-025-02605-y · Cited by 7
Oladimeji, O. O., Imran, A.-A.-Z., Wang, X., & Unnikrishnan, S. (2025) · Image and Vision Computing · doi:10.1016/j.imavis.2025.105601 · Cited by 8
Characterization of Breast Tumors from MR Images Using Radiomics and Machine Learning Approaches
Faraz, K., Dauce, G., Bouhamama, A., Leporq, B., Sasaki, H., Bito, Y., Beuf, O., & Pilleul, F. (2023) · Journal of Personalized Medicine · doi:10.3390/jpm13071062 · Cited by 7
“Furthermore, instead of extracting the radiomics, the deep learning approaches have also been directly applied to the image data [21].”
MethodAI in Healthcare: Breaking New Ground in the Management and Treatment of Cancer
Nizamullah, F. N. U., Zeb, S., Abbasi, N., Qayyum, M. U., & Fahad, M. (2024) · Asian Journal of Engineering Social and Health · doi:10.46799/ajesh.v3i10.453 · Cited by 6
“AI-driven electronic health record (EHR) systems, for instance, are able to analyze patient data in order to facilitate clinical decision-making, lower errors, and guarantee that patients receive the right care (Truhn et al., 2019).”
BackgroundChen, J., Liu, S., Lin, Y., Hu, W., Shi, H., Liao, N., Zhou, M., Gao, W., Chen, Y., & Shi, P. (2024) · Academic Radiology · doi:10.1016/j.acra.2024.11.065 · Cited by 6
Liu, Q., Qu, T., Xu, X., Li, X., Lin, D., Liu, H., Liu, Q., Qi, K., Fan, J., Zhou, L., et al. (2025) · European Journal of Radiology · doi:10.1016/j.ejrad.2025.112004 · Cited by 6
Sohn, Y., & Song, E. J. (2025) · Journal of Breast Cancer · doi:10.4048/jbc.2025.0123 · Cited by 6
“AI models for breast MRI primarily focus on the detection, segmentation, characterization, and classification of breast lesions [81].”
Liu, Z., Deyer, L., Yang, A., Liu, S. H., Gong, J. Q. X., Yang, Y., Huang, M., Doshi, A. H., Lu, M., Lee, D. T., et al. (2022) · IEEE International Conference on Bioinformatics and Bioengineering (BIBE) · doi:10.1109/bibe55377.2022.00013 · Cited by 6
“Radiomics and convolutional neural networks were compared by Truhn et al. [14] in the classification of breast cancer malignancy.”
MethodWang, G., Guo, Q., Shi, D., Zhai, H., Luo, W., Zhang, H., Ren, Z., Yan, G., & Ren, K. (2023) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.29150 · Cited by 6
“The results have exhibited variability, with some studies reporting, lower, 22,36 similar, 18,21 or higher 13,20,37 classi-fi cation performance for radiomics.”
Wang, H., Zhang, K., & Mainardi, L. T. (2024) · IEEE Access · doi:10.1109/access.2024.3408752 · Cited by 6
“Even if our results show a slightly lower performance in respect to DL-models, as seen in other radiomics-DL comparisons [20], the pay-off is in a more explainable, trustworthy solution, which would enable models’ clinical validation.”
MethodResultMüller‐Franzes, G. A., Nebelung, S., Schock, J., Haarburger, C., Khader, F., Pedersoli, F., Schulze-Hagen, M. F., Kuhl, C. K., & Truhn, D. (2022) · Diagnostics · doi:10.3390/diagnostics12020247 · Cited by 6
“This finding is plausible as such approaches achieve better results given enough data [30], yet disallow comparisons with radiomic analyses as performed in the present study.”
ResultSelf-citationPillai, P. S., Holmes, D. R., Carter, R. E., Inoue, A., Cook, D. A., Karwoski, R. A., Fidler, J. L., Fletcher, J. G., Leng, S., Yu, L., et al. (2022) · Journal of medical imaging · doi:10.1117/1.jmi.9.5.055501 · Cited by 6
Radiomics and deep learning methods in expanding the use of screening breast MRI
Reig, B. (2021) · European Radiology · doi:10.1007/s00330-021-08056-9 · Cited by 6
“Studies have already shown that radiomics-style handcrafted feature extraction performs less accurately than deep learning techniques in classifying benign frommalignant lesions [9, 11].”
BackgroundYang, Z.-Q., Zhang, Y., Lu, F., Yang, T., Shan, J., Jiang, Q., Lim, G. H., Bertucci, F., Du, H., & Zhu, Y. (2025) · Gland Surgery · doi:10.21037/gs-2025-223 · Cited by 5
“For instance, a comparative study showed that convolutional neural network outperformed radiomics, achieving a significantly higher AUC in classifying BC lesions as benign or malignant on magnetic resonance imaging (0.88 vs. 0.81; P<0.001) (29).”
InfluentialPicone, C., Fusco, R., Granata, V., Ottaiano, A., Brunese, M. C., Mormile, R., & Petrillo, A. (2025) · Journal of Medical Imaging and Interventional Radiology · doi:10.1007/s44326-025-00085-8 · Cited by 5
“AI-assisted diag - nostics enhance teledermatology accuracy, enabling doctors prioritise urgent cases [19].”
Dong, X., Xiang, Y., Chen, S., Li, Y., Feng, J., & Li, C. (2023) · Displays · doi:10.1016/j.displa.2023.102548 · Cited by 5
Singh, S., & Healy, N. A. (2024) · Insights into Imaging · doi:10.1186/s13244-024-01869-4 · Cited by 5
Multiloss strategy for breast cancer subtype classification using digital breast tomosynthesis
Guo, W., Xiao-min, L., Gong, Z., Zhang, G., & Jiang, X. (2023) · International Journal of Imaging Systems and Technology · doi:10.1002/ima.22978 · Cited by 5
“19 – 26 Zhang et al. 25 proposed a nine-layer CNN with parametric ReLU and rank-based stochastic pooling for the detection of abnormal breasts and achieved an accuracy of 94.0%.”
BackgroundInfluentialLiu, M., Sui, L.-Y., Yin, X., Wang, J., Li, G., Song, J., & Ji, Q. (2023) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-22-967 · Cited by 5
“Therefore, developing a non-0 ResNet34(Ir1e-4)(weit1e-3) [2] ResNet34(lr1e-4)(weit1e-3)e350 [3] ResNet34(lrle-4) [4] ResNet34(Ir3e-4)(weit1e-3) [5] ResNet34(Ir3e-4) [6] ResNet34(lr7e-4)(weitle-3) [7] ResNet34(lr7e-4) [8] ResNet50 [9] ResNet101 [10] Swin-transformer [11]”
Zhou, X., Chen, M., Xu, D., Hu, J., Liu, Z., Song, C., Tang, M., Wang, J., Chen, Y., Luo, Y., et al. (2026) · European Journal of Surgical Oncology · doi:10.1016/j.ejso.2026.111450 · Cited by 4
Dai, W., Li, X., Yu, T., Zhao, D., Shen, J., & Cheng, K.-T. T. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-43990-2_15 · Cited by 4
Liang, Y., Tang, W., Zhang, J., Wang, T., Ng, W. W. Y., Chen, S., Jiang, K., Wei, X., Jiang, X., & Guo, Y. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3552072 · Cited by 5
“Some works, such as [26], compare neural networks with radiomics and find that the former is superior to the latter when the dataset size is large enough.”
A prediction method for radiation proctitis based on SAM-Med2D model
Zhang, N., Ling, H., Zhang, W., & Zhang, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-87409-6 · Cited by 4
Qiu, E., Vejdani‐Jahromi, M., Kaliaev, A., Fazelpour, S., Goodman, D., Ryoo, I., Arasa, V. C. A., Fujima, N., Buch, K. A., & Sakai, O. (2024) · American Journal of Otolaryngology · doi:10.1016/j.amjoto.2024.104357 · Cited by 4
Artificial intelligence in pancreatic disease
Chen, B. (2020) · Artificial Intelligence in Medical Imaging · doi:10.35711/aimi.v1.i1.19 · Cited by 4
Impact of Molecular Subtype Definitions on AI Classification of Breast Cancer at MRI
Bae, M. S. (2023) · Radiology · doi:10.1148/radiol.223041 · Cited by 4
Prinzi, F., Militello, C., Sollami, G., Toia, P., Grutta, L. L., & Vitabile, S. (2025) · Scientific Data · doi:10.1038/s41597-025-05743-w · Cited by 4
“Considering the higher capability of deep architectures to extract informative features compared to shallow methods 20,21 , it is possible to investigate the use of deep architecture for segmentation and classification.”
Shi, K., Chen, C., Fei, Y., Qiu, L., Yuan, J., Zhu, Y., Luo, J., Xu, W., Cao, Y., & Zhou, S. Z. (2025) · Insights into Imaging · doi:10.1186/s13244-025-02142-y · Cited by 4
“MRI serves as a standard clinical tool for the diagnosis and staging of nasopharyngeal carcinoma [3, 17].”
Ye, M., Mao, J., Jin, J., Liu, H., Guo, H., Xu, Y., Yang, P., & Ma, L. (2025) · Journal of Thoracic Disease · doi:10.21037/jtd-2025-1208 · Cited by 3
“Owing to continuous progress in methodology, features obtained through DL now serve as a valuable supplement to conventional handcrafted radiomics characteristics in medical imaging research (23).”
Metadata
- Authors
- Truhn, D., Schrading, S., Haarburger, C., Schneider, H., Merhof, D., & Kuhl, C.
- Affiliations
- RWTH Aachen University (DE)
- Venue
- Radiology
- Published
- 2019-02
- Volume
- 290
- Issue
- 2
- Pages
- 290-297
- ISSN
- 0033-8419 (ISSN-L)
- 1527-1315
- OpenAlex
- W2900955936
- Cited by
- 273
- References
- 22
- Citations by year
- 2026: 21 · 2025: 39 · 2024: 30 · 2023: 55 · 2022: 48 · 2021: 37 · 2020: 28 · 2019: 14
- FWCI
- 13.62
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Radiology (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › MRI in cancer diagnosis
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Keywords
- convolutional neural networks, breast MRI, benign versus malignant differentiation, multiparametric MRI, deep residual neural network, L1 regularization, principal component analysis, radiologist performance comparison, training data size, computer-aided diagnosis, breast cancer diagnosis, AUC comparison, breast lesion segmentation
- MeSH terms
- Breast, Breast Neoplasms, Female, Humans, Image Interpretation, Computer-Assisted, Magnetic Resonance Imaging, Prospective Studies, 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.