{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T23:22:13Z","timestamp":1781047333154,"version":"3.54.1"},"reference-count":43,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,7,19]],"date-time":"2023-07-19T00:00:00Z","timestamp":1689724800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,7,19]],"date-time":"2023-07-19T00:00:00Z","timestamp":1689724800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Non-accidental trauma (NAT) is deadly and difficult to predict. Transformer models pretrained on large datasets have recently produced state of the art performance on diverse prediction tasks, but the optimal pretraining strategies for diagnostic predictions are not known. Here we report the development and external validation of Pretrained and Adapted BERT for Longitudinal Outcomes (PABLO), a transformer-based deep learning model with multitask clinical pretraining, to identify patients who will receive a diagnosis of NAT in the next year. We develop a clinical interface to visualize patient trajectories, model predictions, and individual risk factors. In two comprehensive statewide databases, approximately 1% of patients experience NAT within one year of prediction. PABLO predicts NAT events with area under the receiver operating characteristic curve (AUROC) of 0.844 (95% CI 0.838\u20130.851) in the California test set, and 0.849 (95% CI 0.846\u20130.851) on external validation in Florida, outperforming comparator models. Multitask pretraining significantly improves model performance. Attribution analysis shows substance use, psychiatric, and injury diagnoses, in the context of age and racial demographics, as influential predictors of NAT. As a clinical decision support system, PABLO can identify high-risk patients and patient-specific risk factors, which can be used to target secondary screening and preventive interventions at the point-of-care.<\/jats:p>","DOI":"10.1038\/s41746-023-00875-y","type":"journal-article","created":{"date-parts":[[2023,7,19]],"date-time":"2023-07-19T19:06:06Z","timestamp":1689793566000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":18,"title":["Development and external validation of a pretrained deep learning model for the prediction of non-accidental trauma"],"prefix":"10.1038","volume":"6","author":[{"given":"David","family":"Huang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steven","family":"Cogill","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renee Y.","family":"Hsia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samuel","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0151-5121","authenticated-orcid":false,"given":"David","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,7,19]]},"reference":[{"key":"875_CR1","first-page":"697","volume":"144","author":"KK Christoffel","year":"1990","unstructured":"Christoffel, K. K. Violent death and injury in US children and adolescents. Am. J. Dis. Child. 144, 697\u2013706 (1990).","journal-title":"Am. J. Dis. Child."},{"key":"875_CR2","doi-asserted-by":"publisher","first-page":"693","DOI":"10.1016\/j.jpedsurg.2019.03.024","volume":"55","author":"EH Rosenfeld","year":"2020","unstructured":"Rosenfeld, E. H. et al. Understanding non-accidental trauma in the United States: a national trauma databank study. J. Pediatr. Surg. 55, 693\u2013697 (2020).","journal-title":"J. Pediatr. Surg."},{"key":"875_CR3","doi-asserted-by":"publisher","first-page":"1181","DOI":"10.1097\/AOG.0b013e3181de0194","volume":"115","author":"D Cheng","year":"2010","unstructured":"Cheng, D. & Horon, I. L. Intimate-partner homicide among pregnant and postpartum women. Obstet. Gynecol. 115, 1181\u20131186 (2010).","journal-title":"Obstet. Gynecol."},{"key":"875_CR4","doi-asserted-by":"publisher","first-page":"1032","DOI":"10.1089\/jwh.2019.8072","volume":"29","author":"ME Wallace","year":"2020","unstructured":"Wallace, M. E., Friar, N., Herwehe, J. & Theall, K. P. Violence as a direct cause of and indirect contributor to maternal death. J. Women\u2019s Health (Larchmt) 29, 1032\u20131038 (2020).","journal-title":"J. Women\u2019s Health (Larchmt)"},{"key":"875_CR5","doi-asserted-by":"publisher","first-page":"636","DOI":"10.1542\/peds.108.3.636","volume":"108","author":"KW Feldman","year":"2001","unstructured":"Feldman, K. W. et al. The cause of infant and toddler subdural hemorrhage: a prospective study. Pediatrics 108, 636\u2013646 (2001).","journal-title":"Pediatrics"},{"key":"875_CR6","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1007\/s11524-012-9744-0","volume":"90","author":"GK Singh","year":"2013","unstructured":"Singh, G. K., Azuine, R. E., Siahpush, M. & Kogan, M. D. All-cause and cause-specific mortality among US youth: socioeconomic and rural-urban disparities and international patterns. J. Urban Health 90, 388\u2013405 (2013).","journal-title":"J. Urban Health"},{"key":"875_CR7","doi-asserted-by":"publisher","first-page":"1931","DOI":"10.1001\/jama.2017.13317","volume":"318","author":"MC Mercado","year":"2017","unstructured":"Mercado, M. C., Holland, K., Leemis, R. W., Stone, D. M. & Wang, J. Trends in emergency department visits for nonfatal self-inflicted injuries among youth aged 10 to 24 years in the United States, 2001-2015. JAMA 318, 1931\u20131933 (2017).","journal-title":"JAMA"},{"key":"875_CR8","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1001\/jama.2017.19138","volume":"319","author":"CY Spong","year":"2018","unstructured":"Spong, C. Y. & Bianchi, D. W. Improving public health requires inclusion of underrepresented populations in research. JAMA 319, 337\u2013338 (2018).","journal-title":"JAMA"},{"key":"875_CR9","doi-asserted-by":"publisher","first-page":"1645","DOI":"10.1001\/jama.2018.14745","volume":"320","author":"KV Rhodes","year":"2018","unstructured":"Rhodes, K. V., Dichter, M. E. & Smith, K. L. Challenges and opportunities for studying routine screening for abuse. JAMA 320, 1645\u20131647 (2018).","journal-title":"JAMA"},{"key":"875_CR10","doi-asserted-by":"publisher","first-page":"1678","DOI":"10.1001\/jama.2018.14741","volume":"320","author":"SJ Curry","year":"2018","unstructured":"Curry, S. J. et al. Screening for intimate partner violence, elder abuse, and abuse of vulnerable adults: US preventive services task force final recommendation statement. JAMA 320, 1678\u20131687 (2018).","journal-title":"JAMA"},{"key":"875_CR11","doi-asserted-by":"publisher","first-page":"254","DOI":"10.1016\/j.jemermed.2019.05.030","volume":"57","author":"K Kwan","year":"2019","unstructured":"Kwan, K., Wiebe, D., Cerd\u00e1, M. & Goldman-Mellor, S. Repeat assault injury among adolescents utilizing emergency care: a statewide longitudinal study. J. Emerg. Med. 57, 254\u2013262 (2019).","journal-title":"J. Emerg. Med."},{"key":"875_CR12","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1186\/1471-2431-14-217","volume":"14","author":"KJ Deans","year":"2014","unstructured":"Deans, K. J., Thackeray, J., Groner, J. I., Cooper, J. N. & Minneci, P. C. Risk factors for recurrent injuries in victims of suspected non-accidental trauma: a retrospective cohort study. BMC Pediatr. 14, 217 (2014).","journal-title":"BMC Pediatr."},{"key":"875_CR13","doi-asserted-by":"publisher","first-page":"e164","DOI":"10.5435\/JAAOS-D-18-00502","volume":"28","author":"C Zhao","year":"2020","unstructured":"Zhao, C., Starke, M., Tompson, J. D. & Sabharwal, S. Predictors for nonaccidental trauma in a child with a fracture-a national inpatient database study. J. Am. Acad. Orthop. Surg. 28, e164\u2013e171 (2020).","journal-title":"J. Am. Acad. Orthop. Surg."},{"key":"875_CR14","doi-asserted-by":"publisher","first-page":"b3677","DOI":"10.1136\/bmj.b3677","volume":"339","author":"BY Reis","year":"2009","unstructured":"Reis, B. Y., Kohane, I. S. & Mandl, K. D. Longitudinal histories as predictors of future diagnoses of domestic abuse: modelling study. BMJ 339, b3677 (2009).","journal-title":"BMJ"},{"key":"875_CR15","doi-asserted-by":"publisher","first-page":"e271","DOI":"10.1016\/S2589-7500(19)30123-2","volume":"1","author":"X Liu","year":"2019","unstructured":"Liu, X. et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis. Lancet Digit Health 1, e271\u2013e297 (2019).","journal-title":"Lancet Digit Health"},{"key":"875_CR16","unstructured":"Vaswani, A. et al. Attention is all you need. In Proc. 31st International Conference on Neural Information Processing Systems, 6000\u20136010 (Curran Associates, 2017)."},{"key":"875_CR17","unstructured":"Wolf, T. et al. Transformers: state-of-the-art natural language processing. In Proc. of the 2020 conference on empirical methods in natural language processing: system demonstrations, 38\u201345 (Association for Computational Linguistics, 2020)."},{"key":"875_CR18","unstructured":"Dosovitskiy, A. et al. An image is worth 16x16 words: transformers for image recognition at scale. In Proc 9th International Conference on Learning Representations (2021)."},{"key":"875_CR19","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-62922-y","volume":"10","author":"Y Li","year":"2020","unstructured":"Li, Y. et al. BEHRT: transformer for electronic health records. Sci. Rep. 10, 7155 (2020).","journal-title":"Sci. Rep."},{"key":"875_CR20","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1038\/s41746-021-00455-y","volume":"4","author":"L Rasmy","year":"2021","unstructured":"Rasmy, L., Xiang, Y., Xie, Z., Tao, C. & Zhi, D. Med-BERT: pretrained contextualized embeddings on large-scale structured electronic health records for disease prediction. NPJ Digit. Med. 4, 86 (2021).","journal-title":"NPJ Digit. Med."},{"key":"875_CR21","unstructured":"Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. BERT: pre-training of deep bidirectional transformers for language understanding. In Proc. of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics, 4171-4186 (Association for Computational Linguistics, 2019)."},{"key":"875_CR22","doi-asserted-by":"publisher","first-page":"1584","DOI":"10.1093\/jamia\/ocac086","volume":"29","author":"Z Zhang","year":"2022","unstructured":"Zhang, Z., Yan, C., Zhang, X., Nyemba, S. L. & Malin, B. A. Forecasting the future clinical events of a patient through contrastive learning. J. Am. Med. Inf. Assoc. 29, 1584\u20131592 (2022).","journal-title":"J. Am. Med. Inf. Assoc."},{"key":"875_CR23","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1038\/s41746-018-0029-1","volume":"1","author":"A Rajkomar","year":"2018","unstructured":"Rajkomar, A. et al. Scalable and accurate deep learning with electronic health records. NPJ Digit. Med. 1, 18 (2018).","journal-title":"NPJ Digit. Med."},{"key":"875_CR24","doi-asserted-by":"publisher","DOI":"10.1038\/s41398-020-01100-0","volume":"10","author":"C Su","year":"2020","unstructured":"Su, C. et al. Machine learning for suicide risk prediction in children and adolescents with electronic health records. Transl. Psychiatry 10, 413 (2020).","journal-title":"Transl. Psychiatry"},{"key":"875_CR25","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1007\/BF03086144","volume":"16","author":"AJ Six","year":"2008","unstructured":"Six, A. J., Backus, B. E. & Kelder, J. C. Chest pain in the emergency room: value of the HEART score. Neth. Heart J. 16, 191\u2013196 (2008).","journal-title":"Neth. Heart J."},{"key":"875_CR26","doi-asserted-by":"publisher","first-page":"98","DOI":"10.7326\/0003-4819-135-2-200107170-00010","volume":"135","author":"PS Wells","year":"2001","unstructured":"Wells, P. S. et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and d-dimer. Ann. Intern. Med. 135, 98\u2013107 (2001).","journal-title":"Ann. Intern. Med."},{"key":"875_CR27","doi-asserted-by":"publisher","first-page":"762","DOI":"10.1097\/AOG.0000000000004567","volume":"138","author":"M Wallace","year":"2021","unstructured":"Wallace, M., Gillispie-Bell, V., Cruz, K., Davis, K. & Vilda, D. Homicide during pregnancy and the postpartum period in the United States, 2018\u20132019. Obstet. Gynecol. 138, 762\u2013769 (2021).","journal-title":"Obstet. Gynecol."},{"key":"875_CR28","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1001\/jamainternmed.2016.8180","volume":"177","author":"BC Hohl","year":"2017","unstructured":"Hohl, B. C. et al. Association of drug and alcohol use with adolescent firearm homicide at individual, family, and neighborhood levels. JAMA Intern. Med. 177, 317\u2013324 (2017).","journal-title":"JAMA Intern. Med."},{"key":"875_CR29","unstructured":"Huang, K., Altosaar, J. & Ranganath, R. ClinicalBERT: modeling clinical notes and predicting hospital readmission. Preprint at: https:\/\/ui.adsabs.harvard.edu\/abs\/2019arXiv190405342H (2019)."},{"key":"875_CR30","unstructured":"Zhang, S. et al. Large-scale domain-specific pretraining for biomedical vision-language processing. Preprint at: https:\/\/arxiv.org\/pdf\/2303.00915.pdf (2023)."},{"key":"875_CR31","doi-asserted-by":"publisher","first-page":"e2122260","DOI":"10.1001\/jamanetworkopen.2021.22260","volume":"4","author":"L Baidoo","year":"2021","unstructured":"Baidoo, L., Zakrison, T. L., Feldmeth, G., Lindau, S. T. & Tung, E. L. Domestic violence police reporting and resources during the 2020 COVID-19 stay-at-home order in Chicago, Illinois. JAMA Netw. Open 4, e2122260 (2021).","journal-title":"JAMA Netw. Open"},{"key":"875_CR32","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1038\/s41746-020-0304-9","volume":"3","author":"JK Paulus","year":"2020","unstructured":"Paulus, J. K. & Kent, D. M. Predictably unequal: understanding and addressing concerns that algorithmic clinical prediction may increase health disparities. NPJ Digit. Med. 3, 99 (2020).","journal-title":"NPJ Digit. Med."},{"key":"875_CR33","doi-asserted-by":"publisher","first-page":"2393","DOI":"10.1093\/jamia\/ocab148","volume":"28","author":"HM Thompson","year":"2021","unstructured":"Thompson, H. M. et al. Bias and fairness assessment of a natural language processing opioid misuse classifier: detection and mitigation of electronic health record data disadvantages across racial subgroups. J. Am. Med. Inform. Assoc. 28, 2393\u20132403 (2021).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"875_CR34","unstructured":"Clinical Classifications Software Refined (CCSR) for ICD-10-CM Diagnoses. Healthcare Cost and Utilization Project (HCUP). December 2022. Agency for Healthcare Research and Quality. www.hcup-us.ahrq.gov\/toolssoftware\/ccsr\/dxccsr.jsp."},{"key":"875_CR35","unstructured":"HCUP CCS-Services and Procedures. Healthcare Cost and Utilization Project (HCUP). May 2021. Agency for Healthcare Research and Quality. www.hcup-us.ahrq.gov\/toolssoftware\/ccs_svcsproc\/ccssvcproc.jsp."},{"key":"875_CR36","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1093\/jamia\/ocz228","volume":"27","author":"Y Huang","year":"2020","unstructured":"Huang, Y., Li, W., Macheret, F., Gabriel, R. A. & Ohno-Machado, L. A tutorial on calibration measurements and calibration models for clinical prediction models. J. Am. Med Inf. Assoc. 27, 621\u2013633 (2020).","journal-title":"J. Am. Med Inf. Assoc."},{"key":"875_CR37","doi-asserted-by":"publisher","first-page":"837","DOI":"10.2307\/2531595","volume":"44","author":"ER DeLong","year":"1988","unstructured":"DeLong, E. R., DeLong, D. M. & Clarke-Pearson, D. L. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44, 837\u2013845 (1988).","journal-title":"Biometrics"},{"key":"875_CR38","doi-asserted-by":"publisher","first-page":"404","DOI":"10.1093\/biomet\/26.4.404","volume":"26","author":"CJ Clopper","year":"1934","unstructured":"Clopper, C. J. & Pearson, E. S. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika 26, 404\u2013413 (1934).","journal-title":"Biometrika"},{"key":"875_CR39","doi-asserted-by":"publisher","first-page":"1823","DOI":"10.1016\/j.ajem.2016.06.051","volume":"34","author":"E Kaufman","year":"2016","unstructured":"Kaufman, E. et al. Recurrent violent injury: magnitude, risk factors, and opportunities for intervention from a statewide analysis. Am. J. Emerg. Med. 34, 1823\u20131830 (2016).","journal-title":"Am. J. Emerg. Med."},{"key":"875_CR40","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1001\/jamasurg.2017.3116","volume":"153","author":"J Parreco","year":"2018","unstructured":"Parreco, J. & Rattan, R. Machine learning models for prediction of reinjury after penetrating trauma. JAMA Surg. 153, 184\u2013186 (2018).","journal-title":"JAMA Surg."},{"key":"875_CR41","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1097\/XCS.0000000000000421","volume":"236","author":"PT Albini","year":"2023","unstructured":"Albini, P. T., Zakhary, B., Edwards, S. B., Coimbra, R. & Brenner, M. L. Intimate partner violence and pregnancy: nationwide analysis of injury patterns and risk factors. J. Am. Coll. Surg. 236, 198\u2013207 (2023).","journal-title":"J. Am. Coll. Surg."},{"key":"875_CR42","doi-asserted-by":"crossref","unstructured":"Chen, T. & Guestrin, C. XGBoost: a scalable tree boosting system. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785\u2013794 (Association for Computing Machinery, 2016).","DOI":"10.1145\/2939672.2939785"},{"key":"875_CR43","unstructured":"Sundararajan, M., Taly, A. & Yan, Q. Axiomatic attribution for deep networks. In Proc. 34th International Conference on Machine Learning - Volume 70, 3319\u20133328 (JMLR.org, 2017)."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00875-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00875-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00875-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,7,19]],"date-time":"2023-07-19T19:19:56Z","timestamp":1689794396000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00875-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,7,19]]},"references-count":43,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["875"],"URL":"https:\/\/doi.org\/10.1038\/s41746-023-00875-y","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,7,19]]},"assertion":[{"value":"3 November 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 July 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 July 2023","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"The study was approved by the Institutional Review Board of Stanford University, with a waiver of consent for retrospective research on anonymized data.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"131"}}