{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,17]],"date-time":"2026-05-17T06:43:22Z","timestamp":1779000202803,"version":"3.51.4"},"reference-count":29,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T00:00:00Z","timestamp":1659312000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T00:00:00Z","timestamp":1659312000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["1837985"],"award-info":[{"award-number":["1837985"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2014003"],"award-info":[{"award-number":["2014003"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["1837985"],"award-info":[{"award-number":["1837985"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Inform Decis Mak"],"published-print":{"date-parts":[[2022,12]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec>\n                <jats:title>Background<\/jats:title>\n                <jats:p>Traumatic Brain Injury (TBI) is a common condition with potentially severe long-term complications, the prediction of which remains challenging. Machine learning (ML) methods have been used previously to help physicians predict long-term outcomes of TBI so that appropriate treatment plans can be adopted. However, many ML techniques are \u201cblack box\u201d: it is difficult for humans to understand the decisions made by the model, with post-hoc explanations only identifying isolated relevant factors rather than combinations of factors. Moreover, such models often rely on many variables, some of which might not be available at the time of hospitalization.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Methods<\/jats:title>\n                <jats:p>In this study, we apply an interpretable neural network model based on tropical geometry to predict unfavorable outcomes at six months from hospitalization in TBI patients, based on information available at the time of admission.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Results<\/jats:title>\n                <jats:p>The proposed method is compared to established machine learning methods\u2014XGBoost, Random Forest, and SVM\u2014achieving comparable performance in terms of area under the receiver operating characteristic curve (AUC)\u20140.799 for the proposed method vs. 0.810 for the best black box model. Moreover, the proposed method allows for the extraction of simple, human-understandable rules that explain the model\u2019s predictions and can be used as general guidelines by clinicians to inform treatment decisions.<\/jats:p>\n              <\/jats:sec><jats:sec>\n                <jats:title>Conclusions<\/jats:title>\n                <jats:p>The classification results for the proposed model are comparable with those of traditional ML methods. However, our model is interpretable, and it allows the extraction of intelligible rules. These rules can be used to determine relevant factors in assessing TBI outcomes and can be used in situations when not all necessary factors are known to inform the full model\u2019s decision.<\/jats:p>\n              <\/jats:sec>","DOI":"10.1186\/s12911-022-01953-z","type":"journal-article","created":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T06:02:48Z","timestamp":1659333768000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["An interpretable neural network for outcome prediction in traumatic brain injury"],"prefix":"10.1186","volume":"22","author":[{"given":"Cristian","family":"Minoccheri","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Craig A.","family":"Williamson","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mark","family":"Hemmila","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kevin","family":"Ward","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Erica B.","family":"Stein","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jonathan","family":"Gryak","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kayvan","family":"Najarian","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2022,8,1]]},"reference":[{"key":"1953_CR1","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1186\/s13049-020-00738-5","volume":"28","author":"A Abujaber","year":"2020","unstructured":"Abujaber A, et al. Prediction of in-hospital mortality in patients with post traumatic brain injury using National Trauma Registry and Machine Learning Approach. Scand J Trauma Resusc Emerg Med. 2020;28:44.","journal-title":"Scand J Trauma Resusc Emerg Med"},{"key":"1953_CR2","doi-asserted-by":"publisher","first-page":"1366","DOI":"10.3389\/fneur.2019.01366","volume":"10","author":"RL Amorim","year":"2020","unstructured":"Amorim RL, et al. Prediction of early TBI mortality using a machine learning approach in a LMIC population. Front Neurol. 2020;10:1366.","journal-title":"Front Neurol"},{"key":"1953_CR3","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1186\/s12883-019-1246-1","volume":"19","author":"P Bangirana","year":"2019","unstructured":"Bangirana P, Giordani B, Kobusingye O, et al. Patterns of traumatic brain injury and six-month neuropsychological outcomes in Uganda. BMC Neurol. 2019;19:18.","journal-title":"BMC Neurol"},{"key":"1953_CR4","doi-asserted-by":"publisher","first-page":"766","DOI":"10.1212\/WNL.56.6.766","volume":"56","author":"K Becker","year":"2001","unstructured":"Becker K, et al. Withdrawal of support in intracerebral hemorrhage may lead to self-fulfilling prophecies. Neurology. 2001;56:766\u201372.","journal-title":"Neurology"},{"key":"1953_CR5","doi-asserted-by":"crossref","unstructured":"Farzaneh N, Williamson CA, Gryak J, Najarian K. A hierarchical expert-guided machine learning framework for clinical decision support systems: an application to traumatic brain injury prognostication. npj Digit. Med. 2021;4:78.","DOI":"10.1038\/s41746-021-00445-0"},{"key":"1953_CR6","doi-asserted-by":"crossref","unstructured":"Faul M, Xu L, Wald MM, Coronado VG. Traumatic brain injury in the united states: emergency department visits, hospitalizations and deaths 2002-2006. https:\/\/www.cdc.gov\/traumaticbraininjury\/pdf\/bluebook.pdf, 2010.","DOI":"10.15620\/cdc.5571"},{"issue":"13\u201314","key":"1953_CR7","doi-asserted-by":"publisher","first-page":"1606","DOI":"10.3109\/02699052.2013.831127","volume":"27","author":"TG Finnanger","year":"2013","unstructured":"Finnanger TG, et al. Differentiated patterns of cognitive impairment 12 months after severe and moderate traumatic brain injury. Brain Inj. 2013;27(13\u201314):1606\u201316.","journal-title":"Brain Inj."},{"key":"1953_CR8","doi-asserted-by":"publisher","first-page":"515","DOI":"10.1016\/S1474-4422(14)70030-4","volume":"13","author":"M Geurts","year":"2014","unstructured":"Geurts M, et al. End-of-life decisions in patients with severe acute brain injury. Lancet Neurol. 2014;13:515\u201324.","journal-title":"Lancet Neurol"},{"key":"1953_CR9","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1016\/j.jclinepi.2020.03.005","volume":"122","author":"BY Gravesteijn","year":"2020","unstructured":"Gravesteijn BY, et al. Machine learning algorithms performed no better than regression models for prognostication in traumatic brain injury. J Clin Epidemiol. 2020;122:95\u2013107.","journal-title":"J Clin Epidemiol"},{"issue":"11","key":"1953_CR10","doi-asserted-by":"publisher","first-page":"1144","DOI":"10.3390\/jpm11111144","volume":"11","author":"S-D Hsu","year":"2021","unstructured":"Hsu S-D, et al. Machine learning algorithms to predict in-hospital mortality in patients with traumatic brain injury. J Person Med. 2021;11(11):1144.","journal-title":"J Person Med"},{"key":"1953_CR11","doi-asserted-by":"publisher","first-page":"1025","DOI":"10.1089\/neu.2005.22.1025","volume":"22","author":"CW Hukkelhoven","year":"2005","unstructured":"Hukkelhoven CW, et al. Predicting outcome after traumatic brain injury: development and validation of a prognostic score based on admission characteristics. J Neurotrauma. 2005;22:1025\u201339.","journal-title":"J Neurotrauma"},{"issue":"3","key":"1953_CR12","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1109\/21.256541","volume":"23","author":"J-SR Jang","year":"1993","unstructured":"Jang J-SR. ANFIS: adaptive-network-based fuzzy inference system. IEEE Trans Syst Man Cybern. 1993;23(3):665\u201385.","journal-title":"IEEE Trans Syst Man Cybern"},{"key":"1953_CR13","doi-asserted-by":"publisher","first-page":"20","DOI":"10.4103\/0976-3147.193543","volume":"8","author":"M Majdan","year":"2017","unstructured":"Majdan M, Brazinova A, Rusnak M, Leitgeb J. Outcome prediction after traumatic brain injury: comparison of the performance of routinely used severity scores and multivariable prognostic models. J Neurosci Rural Pract. 2017;8:20.","journal-title":"J Neurosci Rural Pract"},{"key":"1953_CR14","doi-asserted-by":"publisher","first-page":"202","DOI":"10.1089\/neu.2018.6276","volume":"37","author":"K Matsuo","year":"2010","unstructured":"Matsuo K, et al. Machine learning to predict in-hospital morbidity and mortality after traumatic brain injury. J Neurotrauma. 2010;37:202\u201310.","journal-title":"J Neurotrauma"},{"key":"1953_CR15","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1038\/nrneurol.2016.89","volume":"12","author":"T McMillan","year":"2016","unstructured":"McMillan T, et al. The glasgow outcome scale-40 years of application and refinement. Nat Rev Neurol. 2016;12:477\u201385.","journal-title":"Nat Rev Neurol"},{"key":"1953_CR16","doi-asserted-by":"crossref","unstructured":"Moore N, Brennan P, Baillie J. Wide variation and systematic bias in expert clinicians\u2019 perceptions of prognosis following brain injury. Br J Neurosurg. 2013;27:340\u20133.","DOI":"10.3109\/02688697.2012.754402"},{"key":"1953_CR17","first-page":"11","volume":"13","author":"C-S Rau","year":"2018","unstructured":"Rau C-S, et al. Mortality prediction in patients with isolated moderate and severe traumatic brain injury using machine learning models. PLoS ONE. 2018;13:11.","journal-title":"PLoS ONE"},{"key":"1953_CR18","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1097\/SLA.0000000000000600","volume":"260","author":"P Rhee","year":"2014","unstructured":"Rhee P, et al. Increasing trauma deaths in the United States. Ann Surg. 2014;260:13\u201321.","journal-title":"Ann Surg"},{"key":"1953_CR19","doi-asserted-by":"publisher","first-page":"32","DOI":"10.1186\/s12873-016-0098-x","volume":"16","author":"S Rizoli","year":"2016","unstructured":"Rizoli S, et al. Early prediction of outcome after severe traumatic brain injury: a simple and practical model. BMC Emerg Med. 2016;16:32.","journal-title":"BMC Emerg Med"},{"key":"1953_CR20","doi-asserted-by":"crossref","unstructured":"Stulemeijer M, van der Werf S, Borm GF, Vos PE. Early prediction of favourable recovery 6 months after mild traumatic brain injury. J Neurol Neurosurg Psychiatry 2008;79(8)","DOI":"10.1136\/jnnp.2007.131250"},{"key":"1953_CR21","doi-asserted-by":"crossref","unstructured":"Takagi T, and Sugeno M. Fuzzy identification of systems and its applications to modeling and control. IEEE Trans Syst Man Cybern. 1985;15(1):116\u2013132 .","DOI":"10.1109\/TSMC.1985.6313399"},{"key":"1953_CR22","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1016\/S0140-6736(74)91639-0","volume":"304","author":"G Teasdale","year":"1974","unstructured":"Teasdale G, Jennett B. Assessment of coma and impaired consciousness: a practical scale. Lancet. 1974;304:81\u20134.","journal-title":"Lancet"},{"issue":"1","key":"1953_CR23","doi-asserted-by":"publisher","first-page":"234","DOI":"10.3171\/2017.3.JNS162720","volume":"129","author":"A Vedantam","year":"2018","unstructured":"Vedantam A, Robertson CS, Gopinath SP. Clinical characteristics and temporal profile of recovery in patients with favorable outcomes at 6 months after severe traumatic brain injury. J Neurosurg. 2018;129(1):234\u201340.","journal-title":"J Neurosurg"},{"key":"1953_CR24","doi-asserted-by":"crossref","unstructured":"Voormolen DC, et al. Outcomes after complicated and uncomplicated mild traumatic brain injury at three-and six-months post-injury: results from the CENTER-TBI Study. J Clin Med 2020;9(5).","DOI":"10.3390\/jcm9051525"},{"key":"1953_CR25","doi-asserted-by":"publisher","first-page":"2457","DOI":"10.1056\/NEJMoa1404304","volume":"371","author":"DW Wright","year":"2014","unstructured":"Wright DW, et al. Very early administration of progesterone for acute traumatic brain injury. N Engl J Med. 2014;371:2457\u201366.","journal-title":"N Engl J Med"},{"key":"1953_CR26","first-page":"995","volume":"2019","author":"H Yao","year":"2019","unstructured":"Yao H, et al. Using a fuzzy neural network in clinical decision support for patients with advanced heart failure. IEEE Int Conf Bioinf Biomed (BIBM). 2019;2019:995\u20139.","journal-title":"IEEE Int Conf Bioinf Biomed (BIBM)"},{"key":"1953_CR27","unstructured":"Yao H, et al. A novel tropical geometry-based interpretable machine learning method: application in prognosis of advanced heart failure. Preprint, 2021."},{"key":"1953_CR28","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1007\/BF00485052","volume":"30","author":"LA Zadeh","year":"1975","unstructured":"Zadeh LA. Fuzzy logic and approximate reasoning. Synthese. 1975;30:407\u201328.","journal-title":"Synthese"},{"key":"1953_CR29","unstructured":"Zhang L, Naitzat G, Lim L-H. Tropical geometry of deep neural networks. In: Proceedings of the 35th international conference on machine learning 2018;80:5824\u20135832."}],"container-title":["BMC Medical Informatics and Decision Making"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-022-01953-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12911-022-01953-z\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12911-022-01953-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,1]],"date-time":"2022-08-01T06:02:57Z","timestamp":1659333777000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedinformdecismak.biomedcentral.com\/articles\/10.1186\/s12911-022-01953-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,1]]},"references-count":29,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,12]]}},"alternative-id":["1953"],"URL":"https:\/\/doi.org\/10.1186\/s12911-022-01953-z","relation":{},"ISSN":["1472-6947"],"issn-type":[{"value":"1472-6947","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,8,1]]},"assertion":[{"value":"12 May 2022","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 July 2022","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 August 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study is approved by the University of Michigan Institutional Review Board (IRBMED) under application number HUM00098656. The written informed consent from patients is waived by the University of Michigan Institutional Review Board (IRBMED) because this study involves no more than minimal risk to the subjects. All methods were performed in accordance with the relevant guidelines and regulations.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"Jonathan Gryak and Kayvan Najarian are named inventors on a patent application related to the tropical geometry-based fuzzy neural network used in this work. All other authors declare no conflict of interest.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"203"}}