{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T22:03:28Z","timestamp":1777586608065,"version":"3.51.4"},"reference-count":86,"publisher":"Walter de Gruyter GmbH","issue":"2","license":[{"start":{"date-parts":[[2021,4,1]],"date-time":"2021-04-01T00:00:00Z","timestamp":1617235200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2021,4,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    There has been an amplified focus on and benefit from the adoption of artificial intelligence (AI) in medical imaging applications. However, deep learning approaches involve training with massive amounts of annotated data in order to guarantee generalization and achieve high accuracies. Gathering and annotating large sets of training images require expertise which is both expensive and time-consuming, especially in the medical field. Furthermore, in health care systems where mistakes can have catastrophic consequences, there is a general mistrust in the black-box aspect of AI models. In this work, we focus on improving the performance of medical imaging applications when limited data is available while focusing on the interpretability aspect of the proposed AI model. This is achieved by employing a novel transfer learning framework,\n                    <jats:italic>progressive transfer learning<\/jats:italic>\n                    , an automated annotation technique and a correlation analysis experiment on the learned representations.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:italic>Progressive transfer learning<\/jats:italic>\n                    helps jump-start the training of deep neural networks while improving the performance by gradually transferring knowledge from two source tasks into the target task. It is empirically tested on the wrist fracture detection application by first training a general radiology network\n                    <jats:italic>RadiNet<\/jats:italic>\n                    and using its weights to initialize\n                    <jats:italic>\n                      RadiNet\n                      <jats:sub>wrist<\/jats:sub>\n                    <\/jats:italic>\n                    , that is trained on wrist images to detect fractures. Experiments show that\n                    <jats:italic>\n                      RadiNet\n                      <jats:sub>wrist<\/jats:sub>\n                    <\/jats:italic>\n                    achieves an accuracy of 87% and an AUC ROC of 94% as opposed to 83% and 92% when it is pre-trained on the ImageNet dataset.\n                  <\/jats:p>\n                  <jats:p>\n                    This improvement in performance is investigated within an\n                    <jats:italic>explainable AI<\/jats:italic>\n                    framework. More concretely, the learned deep representations of\n                    <jats:italic>\n                      RadiNet\n                      <jats:sub>wrist<\/jats:sub>\n                    <\/jats:italic>\n                    are compared to those learned by the baseline model by conducting a correlation analysis experiment. The results show that, when transfer learning is\n                    <jats:italic>gradually<\/jats:italic>\n                    applied, some features are learned earlier in the network. Moreover, the deep layers in the\n                    <jats:italic>progressive transfer learning<\/jats:italic>\n                    framework are shown to encode features that are not encountered when traditional transfer learning techniques are applied.\n                  <\/jats:p>\n                  <jats:p>\n                    In addition to the empirical results, a clinical study is conducted and the performance of\n                    <jats:italic>\n                      RadiNet\n                      <jats:sub>wrist<\/jats:sub>\n                    <\/jats:italic>\n                    is compared to that of an expert radiologist. We found that\n                    <jats:italic>\n                      RadiNet\n                      <jats:sub>wrist<\/jats:sub>\n                    <\/jats:italic>\n                    exhibited similar performance to that of radiologists with more than 20 years of experience.\n                  <\/jats:p>\n                  <jats:p>This motivates follow-up research to train on more data to feasibly surpass radiologists\u2019 performance, and investigate the interpretability of AI models in the healthcare domain where the decision-making process needs to be credible and transparent.<\/jats:p>","DOI":"10.2478\/jaiscr-2022-0007","type":"journal-article","created":{"date-parts":[[2022,2,26]],"date-time":"2022-02-26T05:54:32Z","timestamp":1645854872000},"page":"101-120","source":"Crossref","is-referenced-by-count":11,"title":["A Progressive and Cross-Domain Deep Transfer Learning Framework for Wrist Fracture Detection"],"prefix":"10.2478","volume":"12","author":[{"given":"Christophe","family":"Karam","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering s, American University of Beirut"}]},{"given":"Julia El","family":"Zini","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering s, American University of Beirut"}]},{"given":"Mariette","family":"Awad","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering s, American University of Beirut"}]},{"given":"Charbel","family":"Saade","sequence":"additional","affiliation":[{"name":"Department of Health Professions, Medical Imaging Sciences, Faculty of Health Sciences , American University of Beirut"}]},{"given":"Lena","family":"Naffaa","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine , American University of Beiru"}]},{"given":"Mohammad El","family":"Amine","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine , American University of Beiru"}]}],"member":"374","published-online":{"date-parts":[[2022,2,23]]},"reference":[{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_001","doi-asserted-by":"crossref","unstructured":"[1] W. Cooney, R. Bussey, J. Dobyns, and R. Linscheid, \u201cDifficult wrist fractures. perilunate fracture-dislocations of the wrist.,\u201d Clinical Orthopaedics and Related Research, no. 214, pp. 136\u2013147, 1987.10.1097\/00003086-198701000-00020","DOI":"10.1097\/00003086-198701000-00020"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_002","doi-asserted-by":"crossref","unstructured":"[2] R. Lindsey, A. Daluiski, S. Chopra, A. Lachapelle, M. Mozer, S. Sicular, et al., \u201cDeep neural network improves fracture detection by clinicians,\u201d Proceedings of the National Academy of Sciences, vol. 115, no. 45, pp. 11591\u201311596, 2018.","DOI":"10.1073\/pnas.1806905115"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_003","doi-asserted-by":"crossref","unstructured":"[3] C. M. Court-Brown and B. Caesar, \u201cEpidemiology of adult fractures: A review,\u201d Injury, vol. 37, pp. 691\u2013697, Aug 2006.10.1016\/j.injury.2006.04.13016814787","DOI":"10.1016\/j.injury.2006.04.130"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_004","doi-asserted-by":"crossref","unstructured":"[4] C. A. Goldfarb, Y. Yin, L. A. Gilula, A. J. Fisher, and M. I. Boyer, \u201cWrist fractures: What the clinician wants to know,\u201d Radiology, vol. 219, no. 1, pp. 11\u201328, 2001. PMID: 11274530.10.1148\/radiology.219.1.r01ap131111274530","DOI":"10.1148\/radiology.219.1.r01ap1311"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_005","doi-asserted-by":"crossref","unstructured":"[5] H. R. Guly, \u201cInjuries initially misdiagnosed as sprained wrist (beware the sprained wrist),\u201d Emergency Medicine Journal, vol. 19, no. 1, pp. 41\u201342, 2002.10.1136\/emj.19.1.41172578811777870","DOI":"10.1136\/emj.19.1.41"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_006","doi-asserted-by":"crossref","unstructured":"[6] B. Petinaux, R. Bhat, K. Boniface, and J. Aristizabal, \u201cAccuracy of radiographic readings in the emergency department,\u201d The American Journal of Emergency Medicine, vol. 29, pp. 18\u201325, Jan 2011.10.1016\/j.ajem.2009.07.01120825769","DOI":"10.1016\/j.ajem.2009.07.011"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_007","doi-asserted-by":"crossref","unstructured":"[7] G. Litjens, T. Kooi, B. E. Bejnordi, A. A. A. Setio, F. Ciompi, M. Ghafoorian, et al., \u201cA survey on deep learning in medical image analysis,\u201d Medical image analysis, vol. 42, pp. 60\u201388, 2017.10.1016\/j.media.2017.07.00528778026","DOI":"10.1016\/j.media.2017.07.005"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_008","doi-asserted-by":"crossref","unstructured":"[8] D. Kim and T. MacKinnon, \u201cArtificial intelligence in fracture detection: Transfer learning from deep convolutional neural networks,\u201d Clinical Radiology, vol. 73, 12 2017.10.1016\/j.crad.2017.11.01529269036","DOI":"10.1016\/j.crad.2017.11.015"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_009","doi-asserted-by":"crossref","unstructured":"[9] J. Olczak, N. Fahlberg, A. Maki, A. Razavian, A. Jilert, A. Stark, et al., \u201cArtificial intelligence for analyzing orthopedic trauma radiographs: Deep learning algorithms\u2014are they on par with humans for diagnosing fractures?,\u201d Acta Orthopaedica, vol. 88, pp. 1\u20136, 07 2017.10.1080\/17453674.2017.1344459569480028681679","DOI":"10.1080\/17453674.2017.1344459"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_010","doi-asserted-by":"crossref","unstructured":"[10] R. Lindsey, A. Daluiski, S. Chopra, A. Lachapelle, M. Mozer, S. Sicular, et al., \u201cDeep neural network improves fracture detection by clinicians,\u201d Proceedings of the National Academy of Sciences, vol. 115, no. 45, pp. 11591\u201311596, 2018.","DOI":"10.1073\/pnas.1806905115"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_011","doi-asserted-by":"crossref","unstructured":"[11] D. Soekhoe, P. van der Putten, and A. Plaat, \u201cOn the impact of data set size in transfer learning using deep neural networks,\u201d Lecture Notes in Computer Science Advances in Intelligent Data Analysis XV, pp. 50\u201360, 2016.10.1007\/978-3-319-46349-0_5","DOI":"10.1007\/978-3-319-46349-0_5"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_012","doi-asserted-by":"crossref","unstructured":"[12] H.-C. Shin, H. R. Roth, M. Gao, L. Lu, Z. Xu, I. Nogues, et al., \u201cDeep convolutional neural networks for computer-aided detection: Cnn architectures, dataset characteristics and transfer learning,\u201d IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1285\u20131298, 2016.","DOI":"10.1109\/TMI.2016.2528162"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_013","doi-asserted-by":"crossref","unstructured":"[13] B. Q. Huynh, H. Li, and M. L. Giger, \u201cDigital mammographic tumor classification using transfer learning from deep convolutional neural networks,\u201d Journal of Medical Imaging, vol. 3, no. 3, p. 034501, 2016.","DOI":"10.1117\/1.JMI.3.3.034501"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_014","doi-asserted-by":"crossref","unstructured":"[14] A. Van Opbroek, M. A. Ikram, M. W. Vernooij, and M. De Bruijne, \u201cTransfer learning improves supervised image segmentation across imaging protocols,\u201d IEEE transactions on medical imaging, vol. 34, no. 5, pp. 1018\u20131030, 2014.","DOI":"10.1109\/TMI.2014.2366792"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_015","unstructured":"[15] V. Christen, A. Gro\u00df, and E. Rahm, \u201cApproaches for annotating medical documents.,\u201d in LWDA, pp. 227\u2013232, 2016."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_016","unstructured":"[16] P. Klassen, F. Xia, and M. Yetisgen-Yildiz, \u201cAnnotating and detecting medical events in clinical notes,\u201d in Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC 2016), pp. 3417\u20133421, 2016."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_017","doi-asserted-by":"crossref","unstructured":"[17] J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, \u201cImagenet: A large-scale hierarchical image database,\u201d in 2009 IEEE conference on computer vision and pattern recognition, pp. 248\u2013255, Ieee, 2009.10.1109\/CVPR.2009.5206848","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_018","unstructured":"[18] C. Karam, J. El Zini, and M. Awad, \u201cX-ray wrist fracture classification,\u201d 2019."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_019","doi-asserted-by":"crossref","unstructured":"[19] Y. LeCun, Y. Bengio, and G. Hinton, \u201cDeep learning,\u201d nature, vol. 521, no. 7553, p. 436, 2015.","DOI":"10.1038\/nature14539"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_020","doi-asserted-by":"crossref","unstructured":"[20] P. Lakhani and B. Sundaram, \u201cDeep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks,\u201d Radiology, vol. 284, no. 2, pp. 574\u2013582, 2017.10.1148\/radiol.201716232628436741","DOI":"10.1148\/radiol.2017162326"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_021","doi-asserted-by":"crossref","unstructured":"[21] M. P. McBee, O. A. Awan, A. T. Colucci, C. W. Ghobadi, N. Kadom, A. P. Kansagra, et al., \u201cDeep learning in radiology,\u201d Academic radiology, vol. 25, no. 11, pp. 1472\u20131480, 2018.","DOI":"10.1016\/j.acra.2018.02.018"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_022","doi-asserted-by":"crossref","unstructured":"[22] J. H. Thrall, X. Li, Q. Li, C. Cruz, S. Do, K. Dreyer, et al., \u201cArtificial intelligence and machine learning in radiology: opportunities, challenges, pitfalls, and criteria for success,\u201d Journal of the American College of Radiology, vol. 15, no. 3, pp. 504\u2013508, 2018.10.1016\/j.jacr.2017.12.02629402533","DOI":"10.1016\/j.jacr.2017.12.026"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_023","doi-asserted-by":"crossref","unstructured":"[23] M. A. Mazurowski, M. Buda, A. Saha, and M. R. Bashir, \u201cDeep learning in radiology: An overview of the concepts and a survey of the state of the art with focus on mri,\u201d Journal of Magnetic Resonance Imaging, vol. 49, no. 4, pp. 939\u2013954, 2019.10.1002\/jmri.26534648340430575178","DOI":"10.1002\/jmri.26534"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_024","doi-asserted-by":"crossref","unstructured":"[24] A. S. Becker, M. Marcon, S. Ghafoor, M. C. Wurnig, T. Frauenfelder, and A. Boss, \u201cDeep learning in mammography: diagnostic accuracy of a multipurpose image analysis software in the detection of breast cancer,\u201d Investigative radiology, vol. 52, no. 7, pp. 434\u2013440, 2017.10.1097\/RLI.000000000000035828212138","DOI":"10.1097\/RLI.0000000000000358"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_025","doi-asserted-by":"crossref","unstructured":"[25] J. Wang, X. Yang, H. Cai, W. Tan, C. Jin, and L. Li, \u201cDiscrimination of breast cancer with microcalcifications on mammography by deep learning,\u201d Scientific reports, vol. 6, p. 27327, 2016.","DOI":"10.1038\/srep27327"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_026","doi-asserted-by":"crossref","unstructured":"[26] D. Ribli, A. Horv\u00e1th, Z. Unger, P. Pollner, and I. Csabai, \u201cDetecting and classifying lesions in mammograms with deep learning,\u201d Scientific reports, vol. 8, no. 1, p. 4165, 2018.10.1038\/s41598-018-22437-z585466829545529","DOI":"10.1038\/s41598-018-22437-z"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_027","doi-asserted-by":"crossref","unstructured":"[27] M. Araya-Polo, J. Jennings, A. Adler, and T. Dahlke, \u201cDeep-learning tomography,\u201d The Leading Edge, vol. 37, no. 1, pp. 58\u201366, 2018.10.1190\/tle37010058.1","DOI":"10.1190\/tle37010058.1"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_028","unstructured":"[28] K.-L. Hua, C.-H. Hsu, S. C. Hidayati, W.-H. Cheng, and Y.-J. Chen, \u201cComputer-aided classification of lung nodules on computed tomography images via deep learning technique,\u201d OncoTargets and therapy, vol. 8, 2015."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_029","doi-asserted-by":"crossref","unstructured":"[29] T. W\u00fcrfl, F. C. Ghesu, V. Christlein, and A. Maier, \u201cDeep learning computed tomography,\u201d in International conference on medical image computing and computer-assisted intervention, pp. 432\u2013440, Springer, 2016.10.1007\/978-3-319-46726-9_50","DOI":"10.1007\/978-3-319-46726-9_50"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_030","unstructured":"[30] H. Zhang, L. Li, K. Qiao, L. Wang, B. Yan, L. Li, et al., \u201cImage prediction for limited-angle tomography via deep learning with convolutional neural network,\u201d arXiv preprint arXiv:1607.08707, 2016."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_031","doi-asserted-by":"crossref","unstructured":"[31] M. H. Yap, G. Pons, J. Mart\u00ed, S. Ganau, M. Sent\u00eds, R. Zwiggelaar, et al., \u201cAutomated breast ultrasound lesions detection using convolutional neural networks,\u201d IEEE journal of biomedical and health informatics, vol. 22, no. 4, pp. 1218\u20131226, 2017.","DOI":"10.1109\/JBHI.2017.2731873"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_032","doi-asserted-by":"crossref","unstructured":"[32] K. Lekadir, A. Galimzianova,\u00c0. Betriu, M. del Mar Vila, L. Igual, D. L. Rubin, et al., \u201cA convolutional neural network for automatic characterization of plaque composition in carotid ultrasound,\u201d IEEE journal of biomedical and health informatics, vol. 21, no. 1, pp. 48\u201355, 2016.10.1109\/JBHI.2016.2631401529362227893402","DOI":"10.1109\/JBHI.2016.2631401"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_033","doi-asserted-by":"crossref","unstructured":"[33] P. Burlina, S. Billings, N. Joshi, and J. Albayda, \u201cAutomated diagnosis of myositis from muscle ultrasound: Exploring the use of machine learning and deep learning methods,\u201d PloS one, vol. 12, no. 8, p. e0184059, 2017.10.1371\/journal.pone.0184059557667728854220","DOI":"10.1371\/journal.pone.0184059"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_034","doi-asserted-by":"crossref","unstructured":"[34] P. H. Kalmet, S. Sanduleanu, S. Primakov, G. Wu, A. Jochems, T. Refaee, A. Ibrahim, L. v. Hulst, P. Lambin, and M. Poeze, \u201cDeep learning in fracture detection: a narrative review,\u201d Acta orthopaedica, vol. 91, no. 2, pp. 215\u2013220, 2020.10.1080\/17453674.2019.1711323714427231928116","DOI":"10.1080\/17453674.2019.1711323"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_035","doi-asserted-by":"crossref","unstructured":"[35] R. M. Jones, A. Sharma, R. Hotchkiss, J. W. Sperling, J. Hamburger, C. Ledig, R. O\u2019Toole, M. Gardner, S. Venkatesh, M. M. Roberts, et al., \u201cAssessment of a deep-learning system for fracture detection in musculoskeletal radiographs,\u201d NPJ digital medicine, vol. 3, no. 1, pp. 1\u20136, 2020.10.1038\/s41746-020-00352-w759920833145440","DOI":"10.1038\/s41746-020-00352-w"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_036","doi-asserted-by":"crossref","unstructured":"[36] A. M. Raisuddin, E. Vaattovaara, M. Nevalainen, M. Nikki, E. J\u00a8arvenp\u00a8a\u00a8a, K. Makkonen, P. Pinola, T. Palsio, A. Niemensivu, O. Tervonen, et al., \u201cCritical evaluation of deep neural networks for wrist fracture detection,\u201d Scientific reports, vol. 11, no. 1, pp. 1\u201311, 2021.10.1038\/s41598-021-85570-2797104833727668","DOI":"10.1038\/s41598-021-85570-2"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_037","doi-asserted-by":"crossref","unstructured":"[37] B. Guan, G. Zhang, J. Yao, X. Wang, and M. Wang, \u201cArm fracture detection in x-rays based on improved deep convolutional neural network,\u201d Computers & Electrical Engineering, vol. 81, p. 106530, 2020.","DOI":"10.1016\/j.compeleceng.2019.106530"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_038","doi-asserted-by":"crossref","unstructured":"[38] S. J. Pan and Q. Yang, \u201cA survey on transfer learning,\u201d IEEE Trans. knowledge and data engineering, vol. 22, no. 10, pp. 1345\u20131359, 2010.","DOI":"10.1109\/TKDE.2009.191"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_039","doi-asserted-by":"crossref","unstructured":"[39] R. Raina, A. Battle, H. Lee, B. Packer, and A. Y. Ng, \u201cSelf-taught learning: transfer learning from unlabeled data,\u201d in Proceedings of the 24th international conference on Machine learning, pp. 759\u2013766, ACM, 2007.10.1145\/1273496.1273592","DOI":"10.1145\/1273496.1273592"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_040","doi-asserted-by":"crossref","unstructured":"[40] H. Ravishankar, P. Sudhakar, R. Venkataramani, S. Thiruvenkadam, P. Annangi, N. Babu, et al., \u201cUnderstanding the mechanisms of deep transfer learning for medical images,\u201d in Deep Learning and Data Labeling for Medical Applications, pp. 188\u2013196, Springer, 2016.10.1007\/978-3-319-46976-8_20","DOI":"10.1007\/978-3-319-46976-8_20"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_041","doi-asserted-by":"crossref","unstructured":"[41] N. Tajbakhsh, J. Y. Shin, S. R. Gurudu, R. T. Hurst, C. B. Kendall, M. B. Gotway, et al., \u201cConvolutional neural networks for medical image analysis: Full training or fine tuning?,\u201d IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1299\u20131312, 2016.","DOI":"10.1109\/TMI.2016.2535302"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_042","doi-asserted-by":"crossref","unstructured":"[42] B. J. Erickson, P. Korfiatis, Z. Akkus, and T. L. Kline, \u201cMachine learning for medical imaging,\u201d Radiographics, vol. 37, no. 2, pp. 505\u2013515, 2017.10.1148\/rg.2017160130537562128212054","DOI":"10.1148\/rg.2017160130"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_043","unstructured":"[43] A. Krizhevsky, I. Sutskever, and G. E. Hinton, \u201cImagenet classification with deep convolutional neural networks,\u201d in Advances in neural information processing systems, pp. 1097\u20131105, 2012."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_044","doi-asserted-by":"crossref","unstructured":"[44] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, \u201cInception-v4, inception-resnet and the impact of residual connections on learning,\u201d in Thirty-First AAAI Conference on Artificial Intelligence, 2017.10.1609\/aaai.v31i1.11231","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_045","unstructured":"[45] S. Ren, K. He, R. Girshick, and J. Sun, \u201cFaster rcnn: Towards real-time object detection with region proposal networks,\u201d in Advances in neural information processing systems, pp. 91\u201399, 2015."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_046","doi-asserted-by":"crossref","unstructured":"[46] T. Urakawa, Y. Tanaka, H. Matsuzawa, K. Watanabe, and N. Endo, \u201cDetecting intertrochanteric hip fractures with orthopedist-level accuracy using a deep convolutional neural network,\u201d Journal of the International Skeletal Society A Journal of Radiology, Pathology and Orthopedics, vol. 42, pp. 239\u2013244, 2019.10.1007\/s00256-018-3016-329955910","DOI":"10.1007\/s00256-018-3016-3"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_047","unstructured":"[47] P. Rajpurkar, J. Irvin, A. Bagul, D. Ding, T. Duan, H. Mehta, et al., \u201cMura dataset: Towards radiologist-level abnormality detection in musculoskeletal radiographs,\u201d 2017. cite arxiv:1712.06957."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_048","doi-asserted-by":"crossref","unstructured":"[48] K. Gan, D. Xu, Y. Lin, Y. Shen, T. Zhang, K. Hu, et al., \u201cArtificial intelligence detection of distal radius fractures: a comparison between the convolutional neural network and professional assessments,\u201d Acta orthopaedica, pp. 1\u201312, 2019.10.1080\/17453674.2019.1600125671819030942136","DOI":"10.1080\/17453674.2019.1600125"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_049","doi-asserted-by":"crossref","unstructured":"[49] J. de Matos, A. de Souza Britto Jr., L. E. S. Oliveira, and A. L. Koerich, \u201cDouble transfer learning for breast cancer histopathologic image classification,\u201d CoRR, vol. abs\/1904.07834, 2019.","DOI":"10.1109\/IJCNN.2019.8852092"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_050","unstructured":"[50] S. Christodoulidis, M. Anthimopoulos, L. Ebner, A. Christe, and S. G. Mougiakakou, \u201cMulti-source transfer learning with convolutional neural networks for lung pattern analysis,\u201d CoRR, vol. abs\/1612.02589, 2016."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_051","doi-asserted-by":"crossref","unstructured":"[51] J. Li, W. Wu, D. Xue, and P. Gao, \u201cMulti-source deep transfer neural network algorithm,\u201d Sensors (Basel, Switzerland), vol. 19, p. 3992, Sep 2019. 31527437[pmid].10.3390\/s19183992676784731527437","DOI":"10.3390\/s19183992"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_052","unstructured":"[52] R. Gupta and L.-A. Ratinov, \u201cText categorization with knowledge transfer from heterogeneous data sources,\u201d in AAAI, pp. 842\u2013847, 2008."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_053","doi-asserted-by":"crossref","unstructured":"[53] Z. Yu, Z. Jin, L. Wei, J. Guo, J. Huang, D. Cai, X. He, and X.-S. Hua, \u201cProgressive transfer learning for person re-identification,\u201d Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, Aug 2019.10.24963\/ijcai.2019\/586","DOI":"10.24963\/ijcai.2019\/586"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_054","unstructured":"[54] W. Hu, Y. Jin, X. Wu, and J. Chen, \u201cProgressive transfer learning for low frequency data prediction in full waveform inversion,\u201d 2019."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_055","doi-asserted-by":"crossref","unstructured":"[55] Y. Gu, Z. Ge, C. P. Bonnington, and J. Zhou, \u201cProgressive transfer learning and adversarial domain adaptation for cross-domain skin disease classification,\u201d IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 5, pp. 1379\u20131393, 2020.","DOI":"10.1109\/JBHI.2019.2942429"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_056","unstructured":"[56] J. Antol\u00edk, \u201cAutomatic annotation of medical records,\u201d Studies in health technology and informatics, vol. 116, pp. 817\u2013822, 2005."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_057","doi-asserted-by":"crossref","unstructured":"[57] C. Ganoe, W. Wu, P. Barr, W. Haslett, M. Dannenberg, K. Bonasia, J. Finora, J. Schoonmaker, W. Onsando, J. Ryan, et al., \u201cNatural language processing for automated annotation of medication mentions in primary care visit conversations,\u201d medRxiv, 2021.10.1101\/2021.03.29.21254488","DOI":"10.1101\/2021.03.29.21254488"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_058","doi-asserted-by":"crossref","unstructured":"[58] H. Li, B. Zhang, Y. Zhang, W. Liu, Y. Mao, J. Huang, and L. Wei, \u201cA semi-automated annotation algorithm based on weakly supervised learning for medical images,\u201d Biocybernetics and Biomedical Engineering, vol. 40, no. 2, pp. 787\u2013802, 2020.10.1016\/j.bbe.2020.03.005","DOI":"10.1016\/j.bbe.2020.03.005"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_059","doi-asserted-by":"crossref","unstructured":"[59] R. Bouslimi and J. Akaichi, \u201cNew approach for automatic medical image annotation using the bagof-words model,\u201d in 2015 6th IEEE International Conference on Software Engineering and Service Science (ICSESS), pp. 1088\u20131093, 2015.","DOI":"10.1109\/ICSESS.2015.7339241"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_060","doi-asserted-by":"crossref","unstructured":"[60] T. Gong, S. Li, J. Wang, C. L. Tan, B. Pang, T. Lim, C. Lee, Q. Tian, and Z. Zhang, \u201cAutomatic labeling and classification of brain ct images,\u201d pp. 1581\u20131584, 09 2011.","DOI":"10.1109\/ICIP.2011.6115751"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_061","doi-asserted-by":"crossref","unstructured":"[61] A. R. Aronson and F.-M. Lang, \u201cAn overview of metamap: historical perspective and recent advances,\u201d Journal of the American Medical Informatics Association : JAMIA, vol. 17, no. 3, pp. 229\u2013236, 2010. PMC2995713[pmcid].10.1136\/jamia.2009.002733299571320442139","DOI":"10.1136\/jamia.2009.002733"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_062","unstructured":"[62] M. D. Zeiler and R. Fergus, \u201cVisualizing and understanding convolutional networks,\u201d CoRR, vol. abs\/1311.2901, 2013."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_063","unstructured":"[63] M. Sundararajan, A. Taly, and Q. Yan, \u201cAxiomatic attribution for deep networks,\u201d arXiv preprint arXiv:1703.01365, 2017."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_064","doi-asserted-by":"crossref","unstructured":"[64] M. T. Ribeiro, S. Singh, and C. Guestrin, \u201cAnchors: High-precision model-agnostic explanations,\u201d in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 32, 2018.10.1609\/aaai.v32i1.11491","DOI":"10.1609\/aaai.v32i1.11491"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_065","unstructured":"[65] B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, \u201cObject detectors emerge in deep scene cnns,\u201d 2015."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_066","doi-asserted-by":"crossref","unstructured":"[66] B. Zhou, A. Khosla, L. A., A. Oliva, and A. Torralba, \u201cLearning deep features for discriminative localization.,\u201d CVPR, 2016.10.1109\/CVPR.2016.319","DOI":"10.1109\/CVPR.2016.319"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_067","doi-asserted-by":"crossref","unstructured":"[67] A. Chattopadhay, A. Sarkar, P. Howlader, and V. N. Balasubramanian, \u201cGrad-cam++: Generalized gradient-based visual explanations for deep convolutional networks,\u201d 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), Mar 2018.10.1109\/WACV.2018.00097","DOI":"10.1109\/WACV.2018.00097"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_068","doi-asserted-by":"crossref","unstructured":"[68] B. N. Patro, M. Lunayach, S. Patel, and V. P. Namboodiri, \u201cU-cam: Visual explanation using uncertainty based class activation maps,\u201d in Proceedings of the IEEE International Conference on Computer Vision, pp. 7444\u20137453, 2019.","DOI":"10.1109\/ICCV.2019.00754"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_069","doi-asserted-by":"crossref","unstructured":"[69] M. T. Ribeiro, S. Singh, and C. Guestrin, \u201c\u201dwhy should i trust you?\u201d: Explaining the predictions of any classifier,\u201d 2016.10.1145\/2939672.2939778","DOI":"10.1145\/2939672.2939778"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_070","doi-asserted-by":"crossref","unstructured":"[70] D. Bau, B. Zhou, A. Khosla, A. Oliva, and A. Torralba, \u201cNetwork dissection: Quantifying interpretability of deep visual representations,\u201d in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 6541\u20136549, 2017.","DOI":"10.1109\/CVPR.2017.354"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_071","doi-asserted-by":"crossref","unstructured":"[71] R. Fong and A. Vedaldi, \u201cNet2vec: Quantifying and explaining how concepts are encoded by filters in deep neural networks,\u201d in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 8730\u20138738, 2018.","DOI":"10.1109\/CVPR.2018.00910"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_072","doi-asserted-by":"crossref","unstructured":"[72] K. Leino, S. Sen, A. Datta, M. Fredrikson, and L. Li, \u201cInfluence-directed explanations for deep convolutional networks,\u201d in 2018 IEEE International Test Conference (ITC), pp. 1\u20138, IEEE, 2018.10.1109\/TEST.2018.8624792","DOI":"10.1109\/TEST.2018.8624792"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_073","doi-asserted-by":"crossref","unstructured":"[73] A. Mahendran and A. Vedaldi, \u201cUnderstanding deep image representations by inverting them,\u201d in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 5188\u20135196, 2015.","DOI":"10.1109\/CVPR.2015.7299155"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_074","doi-asserted-by":"crossref","unstructured":"[74] A. Dosovitskiy and T. Brox, \u201cInverting visual representations with convolutional networks,\u201d in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4829\u20134837, 2016.","DOI":"10.1109\/CVPR.2016.522"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_075","doi-asserted-by":"crossref","unstructured":"[75] S. Bach, A. Binder, G. Montavon, F. Klauschen, K. M\u00fcller, and W. Samek, \u201cOn pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation,\u201d PLoS ONE, vol. 10, 2015.10.1371\/journal.pone.0130140449875326161953","DOI":"10.1371\/journal.pone.0130140"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_076","doi-asserted-by":"crossref","unstructured":"[76] M. B\u00f6hle, F. Eitel, M. Weygandt, and K. Ritter, \u201cLayer-wise relevance propagation for explaining deep neural network decisions in mri-based alzheimer\u2019s disease classification,\u201d Frontiers in aging neuroscience, vol. 11, pp. 194\u2013194, Jul 2019. 31417397[pmid].10.3389\/fnagi.2019.00194668508731417397","DOI":"10.3389\/fnagi.2019.00194"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_077","doi-asserted-by":"crossref","unstructured":"[77] F. Eitel, E. Soehler, J. Bellmann-Strobl, A. U. Brandt, K. Ruprecht, R. M. Giess, J. Kuchling, S. Asseyer, M. Weygandt, J.-D. Haynes, M. S. l, F. Paul, and K. Ritter, \u201cUncovering convolutional neural network decisions for diagnosing multiple sclerosis on conventional mri using layer-wise relevance propagation,\u201d NeuroImage: Clinical, vol. 24, p. 102003, 2019.","DOI":"10.1016\/j.nicl.2019.102003"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_078","doi-asserted-by":"crossref","unstructured":"[78] D. R. Hardoon, S. Szedmak, and J. Shawe-Taylor, \u201cCanonical correlation analysis: An overview with application to learning methods,\u201d Neural computation, vol. 16, no. 12, pp. 2639\u20132664, 2004.","DOI":"10.1162\/0899766042321814"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_079","doi-asserted-by":"crossref","unstructured":"[79] D. Sussillo, M. M. Churchland, M. T. Kaufman, and K. V. Shenoy, \u201cA neural network that finds a naturalistic solution for the production of muscle activity,\u201d Nature neuroscience, vol. 18, no. 7, pp. 1025\u20131033, 2015.","DOI":"10.1038\/nn.4042"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_080","doi-asserted-by":"crossref","unstructured":"[80] M. Faruqui and C. Dyer, \u201cImproving vector space word representations using multilingual correlation,\u201d in Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics, pp. 462\u2013471, 2014.10.3115\/v1\/E14-1049","DOI":"10.3115\/v1\/E14-1049"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_081","unstructured":"[81] M. Raghu, J. Gilmer, J. Yosinski, and J. Sohl-Dickstein, \u201cSvcca: Singular vector canonical correlation analysis for deep learning dynamics and interpretability,\u201d in Advances in Neural Information Processing Systems, pp. 6076\u20136085, 2017."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_082","doi-asserted-by":"crossref","unstructured":"[82] Y. Bengio, A. Courville, and P. Vincent, \u201cRepresentation learning: A review and new perspectives,\u201d IEEE transactions on pattern analysis and machine intelligence, vol. 35, no. 8, pp. 1798\u20131828, 2013.","DOI":"10.1109\/TPAMI.2013.50"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_083","unstructured":"[83] J. Yosinski, J. Clune, Y. Bengio, and H. Lipson, \u201cHow transferable are features in deep neural networks?,\u201d in Advances in neural information processing systems, pp. 3320\u20133328, 2014."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_084","doi-asserted-by":"crossref","unstructured":"[84] C. Castillo, T. Steffens, L. Sim, and L. Caffery, \u201cThe effect of clinical information on radiology reporting: A systematic review,\u201d Journal of Medical Radiation Sciences, vol. 68, no. 1, pp. 60\u201374, 2021.10.1002\/jmrs.424789092332870580","DOI":"10.1002\/jmrs.424"},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_085","unstructured":"[85] Theano Development Team, \u201cTheano: A Python framework for fast computation of mathematical expressions,\u201d arXiv e-prints, vol. abs\/1605.02688, May 2016."},{"key":"2026042814174551871_j_jaiscr-2022-0007_ref_086","unstructured":"[86] G. Huang, Z. Liu, and K. Q. Weinberger, \u201cDensely connected convolutional networks,\u201d CoRR, vol. abs\/1608.06993, 2016."}],"container-title":["Journal of Artificial Intelligence and Soft Computing Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/reference-global.com\/pdf\/10.2478\/jaiscr-2022-0007","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T20:07:07Z","timestamp":1777406827000},"score":1,"resource":{"primary":{"URL":"https:\/\/reference-global.com\/article\/10.2478\/jaiscr-2022-0007"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,4,1]]},"references-count":86,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,2,23]]},"published-print":{"date-parts":[[2021,4,1]]}},"alternative-id":["10.2478\/jaiscr-2022-0007"],"URL":"https:\/\/doi.org\/10.2478\/jaiscr-2022-0007","relation":{},"ISSN":["2449-6499"],"issn-type":[{"value":"2449-6499","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,1]]}}}