{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T15:10:42Z","timestamp":1778598642091,"version":"3.51.4"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T00:00:00Z","timestamp":1778544000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T00:00:00Z","timestamp":1778544000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100002551","name":"Seoul National University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100002551","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>To address diagnostic delays in pediatric abdominal emergencies, this study aimed to develop and validate multi-institutional deep learning models for detecting intussusception and splenomegaly on abdominal radiographs, thereby evaluating their potential to enhance clinical triage. This retrospective study included 26,552 radiographs from seven tertiary hospitals (2012\u20132022). Models were trained with EfficientDet-B2, selected after an ablation study against ResNet-50, DenseNet-121, and other EfficientDet variants. Three strategies were compared: independent binary classifiers, a multiclass model, and transfer learning (fine-tuning from intussusception to splenomegaly). External performance was primarily assessed using a locked external test set, with leave-one-institution-out validation performed as a secondary robustness analysis. Subgroup analyses were performed by institution and body weight (&lt;\u200910\u00a0kg, 10\u2013\u2009&lt;\u200920\u00a0kg, 20-\u2009&lt;\u200930\u00a0kg), and visualized using area under curve (AUC) with 95% confidence intervals. An ablation study identified EfficientDet-B2 as the best-performing backbone among ResNet-50, DenseNet-121, and other EfficientDet variants. The EfficientDet-B2 models achieved internal AUCs of 0.851 for intussusception and 0.834 for splenomegaly, with locked external AUCs of 0.818 and 0.806, respectively. Secondary leave-one-institution-out (LOIO) AUCs ranged from 0.832 to 0.912, while multiclass training and transfer learning improved intussusception and splenomegaly performance, respectively. Subgroup analyses likewise showed stable AUC across institutions, with reduced sensitivity in the\u2009&lt;\u200910\u00a0kg group but maintained performance in those weighing 10\u2013\u2009&lt;\u200930\u00a0kg. Multiclass training improved intussusception detection, whereas transfer learning enhanced splenomegaly recognition. With robust performance across institutions and clinically interpretable subgroup analyses, these models show promise for supporting triage of pediatric abdominal radiographs. Prospective validation in real-world workflows is warranted to confirm utility.<\/jats:p>","DOI":"10.1007\/s10278-026-01994-0","type":"journal-article","created":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T14:34:30Z","timestamp":1778596470000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Transfer learning for Multi-institutional Classification of Intussusception and Splenomegaly in Pediatric Abdominal Radiographs"],"prefix":"10.1007","author":[{"given":"Minsoo","family":"Shin","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sungwon","family":"Ham","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9521-3575","authenticated-orcid":false,"given":"Yoon","family":"Lee","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6449-7819","authenticated-orcid":false,"given":"Jung Ok","family":"Shim","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,5,12]]},"reference":[{"key":"1994_CR1","doi-asserted-by":"publisher","first-page":"867","DOI":"10.1002\/bjs.1800790906","volume":"79","author":"MD Stringer","year":"1992","unstructured":"1 Stringer, M. D., Pablot, S. M. & Brereton, R. J. Paediatric intussusception. Br J Surg 79, 867-876 (1992). https:\/\/doi.org\/10.1002\/bjs.1800790906","journal-title":"Br J Surg"},{"key":"1994_CR2","doi-asserted-by":"publisher","first-page":"861","DOI":"10.1007\/s00595-009-3979-8","volume":"39","author":"A Demirkan","year":"2009","unstructured":"2 Demirkan, A. et al. Intussusception in adult and pediatric patients: two different entities. Surg Today 39, 861-865 (2009). https:\/\/doi.org\/10.1007\/s00595-009-3979-8","journal-title":"Surg Today"},{"key":"1994_CR3","doi-asserted-by":"publisher","unstructured":"Gluckman, S., Karpelowsky, J., Webster, A. C. & McGee, R. G. Management for intussusception in children. Cochrane Database Syst Rev 6, Cd006476 (2017). https:\/\/doi.org\/10.1002\/14651858.CD006476.pub3","DOI":"10.1002\/14651858.CD006476.pub3"},{"key":"1994_CR4","doi-asserted-by":"publisher","first-page":"1101","DOI":"10.1007\/s00247-017-3878-x","volume":"47","author":"EA Edwards","year":"2017","unstructured":"4 Edwards, E. A. et al. Intussusception: past, present and future. Pediatr Radiol 47, 1101-1108 (2017). https:\/\/doi.org\/10.1007\/s00247-017-3878-x","journal-title":"Pediatr Radiol"},{"key":"1994_CR5","doi-asserted-by":"publisher","first-page":"521","DOI":"10.1016\/j.acra.2017.01.002","volume":"24","author":"AG Carroll","year":"2017","unstructured":"5 Carroll, A. G. et al. Comparative Effectiveness of Imaging Modalities for the Diagnosis and Treatment of Intussusception: A Critically Appraised Topic. Acad Radiol 24, 521-529 (2017). https:\/\/doi.org\/10.1016\/j.acra.2017.01.002","journal-title":"Acad Radiol"},{"key":"1994_CR6","doi-asserted-by":"publisher","first-page":"704635","DOI":"10.3389\/fped.2021.704635","volume":"9","author":"M Suttorp","year":"2021","unstructured":"6 Suttorp, M. & Classen, C. F. Splenomegaly in Children and Adolescents. Front Pediatr 9, 704635 (2021). https:\/\/doi.org\/10.3389\/fped.2021.704635","journal-title":"Front Pediatr"},{"key":"1994_CR7","doi-asserted-by":"publisher","first-page":"119","DOI":"10.2214\/ajr.157.1.2048509","volume":"157","author":"HK Rosenberg","year":"1991","unstructured":"7 Rosenberg, H. K. et al. Normal splenic size in infants and children: sonographic measurements. AJR Am J Roentgenol 157, 119-121 (1991). https:\/\/doi.org\/10.2214\/ajr.157.1.2048509","journal-title":"AJR Am J Roentgenol"},{"key":"1994_CR8","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1177\/028418519403500210","volume":"35","author":"P Prassopoulos","year":"1994","unstructured":"8 Prassopoulos, P. & Cavouras, D. CT assessment of normal splenic size in children. Acta Radiol 35, 152-154 (1994).","journal-title":"Acta Radiol"},{"key":"1994_CR9","doi-asserted-by":"crossref","unstructured":"Grover, S. A., Barkun, A. N. & Sackett, D. L. The rational clinical examination. Does this patient have splenomegaly? Jama 270, 2218\u20132221 (1993).","DOI":"10.1001\/jama.1993.03510180088040"},{"key":"1994_CR10","doi-asserted-by":"publisher","first-page":"105","DOI":"10.1016\/j.blre.2008.10.001","volume":"23","author":"AL Pozo","year":"2009","unstructured":"10 Pozo, A. L., Godfrey, E. M. & Bowles, K. M. Splenomegaly: investigation, diagnosis and management. Blood Rev 23, 105-111 (2009). https:\/\/doi.org\/10.1016\/j.blre.2008.10.001","journal-title":"Blood Rev"},{"key":"1994_CR11","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","volume":"521","author":"Y LeCun","year":"2015","unstructured":"11 LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436-444 (2015). https:\/\/doi.org\/10.1038\/nature14539","journal-title":"Nature"},{"key":"1994_CR12","doi-asserted-by":"crossref","unstructured":"Mathew, Amitha, P. Amudha, and S. Sivakumari. \"Deep learning techniques: an overview.\" International conference on advanced machine learning technologies and applications. Singapore: Springer Singapore, 2020.","DOI":"10.1007\/978-981-15-3383-9_54"},{"key":"1994_CR13","doi-asserted-by":"publisher","unstructured":"Alapat, D. J., Menon, M. V. & Ashok, S. A Review on Detection of Pneumonia in Chest X-ray Images Using Neural Networks. J Biomed Phys Eng 12, 551\u2013558 (2022). https:\/\/doi.org\/10.31661\/jbpe.v0i0.2202-1461","DOI":"10.31661\/jbpe.v0i0.2202-1461"},{"key":"1994_CR14","doi-asserted-by":"publisher","first-page":"14917","DOI":"10.1038\/s41598-024-65703-z","volume":"14","author":"CF Chen","year":"2024","unstructured":"14 Chen, C. F. et al. A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography. Sci Rep 14, 14917 (2024). https:\/\/doi.org\/10.1038\/s41598-024-65703-z","journal-title":"Sci Rep"},{"key":"1994_CR15","doi-asserted-by":"crossref","unstructured":"Koonce, Brett. \"ResNet 50.\"\u00a0Convolutional neural networks with swift for tensorflow: image recognition and dataset categorization. Berkeley, CA: Apress, 2021. 63\u201372.","DOI":"10.1007\/978-1-4842-6168-2_6"},{"key":"1994_CR16","doi-asserted-by":"crossref","unstructured":"Huang, Gao, et al. \"Densely connected convolutional networks.\"\u00a0Proceedings of the IEEE conference on computer vision and pattern recognition. 2017.","DOI":"10.1109\/CVPR.2017.243"},{"key":"1994_CR17","unstructured":"Tan, Mingxing, and Quoc Le. \"Efficientnet: Rethinking model scaling for convolutional neural networks.\"\u00a0International conference on machine learning. PMLR, 2019."},{"key":"1994_CR18","doi-asserted-by":"crossref","unstructured":"Tan, Mingxing, Ruoming Pang, and Quoc V. Le. \"Efficientdet: Scalable and efficient object detection.\"\u00a0Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition. 2020.","DOI":"10.1109\/CVPR42600.2020.01079"},{"key":"1994_CR19","doi-asserted-by":"crossref","unstructured":"Lin, Tsung-Yi, et al. \"Microsoft coco: Common objects in context.\"\u00a0European conference on computer vision. Cham: Springer International Publishing, 2014.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"1994_CR20","unstructured":"Liu, Yifan, et al. \"Auxiliary learning for deep multi-task learning.\"\u00a0arXiv preprint arXiv:1909.02214\u00a0(2019)."},{"key":"1994_CR21","doi-asserted-by":"crossref","unstructured":"Llugsi, Ricardo, et al. \"Comparison between Adam, AdaMax and Adam W optimizers to implement a Weather Forecast based on Neural Networks for the Andean city of Quito.\"\u00a02021 IEEE Fifth Ecuador Technical Chapters Meeting (ETCM). IEEE, 2021.","DOI":"10.1109\/ETCM53643.2021.9590681"},{"key":"1994_CR22","unstructured":"Loshchilov, Ilya, and Frank Hutter. \"Decoupled weight decay regularization.\"\u00a0arXiv preprint arXiv:1711.05101\u00a0(2017)."},{"key":"1994_CR23","unstructured":"Mao, Anqi, Mehryar Mohri, and Yutao Zhong. \"Cross-entropy loss functions: Theoretical analysis and applications.\"\u00a0International conference on Machine learning. pmlr, 2023."},{"key":"1994_CR24","doi-asserted-by":"crossref","unstructured":"Hossin, Mohammad, and Md Nasir Sulaiman. \"A review on evaluation metrics for data classification evaluations.\"\u00a0International journal of data mining & knowledge management process\u00a05.2 (2015): 1.","DOI":"10.5121\/ijdkp.2015.5201"},{"issue":"6","key":"1994_CR25","doi-asserted-by":"publisher","first-page":"599","DOI":"10.14569\/IJACSA.2021.0120670","volume":"12","author":"\u017deljko Vujovi\u0107","year":"2021","unstructured":"25 Vujovi\u0107, \u017deljko. \"Classification model evaluation metrics.\"\u00a0International Journal of Advanced Computer Science and Applications\u00a012.6 (2021): 599-606.","journal-title":"International Journal of Advanced Computer Science and Applications"},{"key":"1994_CR26","doi-asserted-by":"crossref","unstructured":"DeLong, Elizabeth R., David M. DeLong, and Daniel L. Clarke-Pearson. \"Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach.\"\u00a0Biometrics\u00a0(1988): 837\u2013845.","DOI":"10.2307\/2531595"}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-01994-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-01994-0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-01994-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T14:34:45Z","timestamp":1778596485000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10278-026-01994-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,12]]},"references-count":26,"alternative-id":["1994"],"URL":"https:\/\/doi.org\/10.1007\/s10278-026-01994-0","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,12]]},"assertion":[{"value":"16 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 April 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 April 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 May 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This retrospective multicentre study was approved by the Institutional Review Boards of all participating institutions: Hospital A (No. 2022AS0278), Hospital B (No. SMC2022\u201402\u2014093\u2014002), Hospital C (No. 2022\u20140418\u20140002), Hospital D (No. AJOUIRB-MDB-2022\u2014192), Hospital E (No. CBNUH2022\u201402\u2014015\u2014002), Hospital F (No. CHAMC2022\u201404\u2014031\u2014003), and Hospital G (No. 2021\u201408\u2014063, 2022\u201405\u2014010). The requirement for informed consent was waived owing to the retrospective design. The study complied with the Declaration of Helsinki.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical Approval"}},{"value":"The requirement for informed consent was waived by the Institutional Review Boards because of the retrospective study design.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to Participate"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to Publish"}},{"value":"The authors declare no competing interests.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"The authors declare no competing interests.","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}]}}