{"doi":"10.3389/fped.2025.1522845","title":"Machine learning-based prediction of mortality in pediatric trauma patients","abstract":"<jats:sec><jats:title>Background</jats:title><jats:p>This study aimed to develop a predictive model for mortality outcomes among pediatric trauma patients using machine learning (ML) algorithms.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>We extracted data on a cohort of pediatric trauma patients (18 years and younger) from the National Trauma Data Bank (NTDB). The main aim was to identify clinical and physiologic variables that could serve as predictors for pediatric trauma mortality. Data was split into a development cohort (70%) to build four ML models and then tested in a validation cohort (30%). The area under the receiver operating characteristic curve (AUC) was used to assess each model's performance.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>In 510,381 children, the gross mortality rate was 1.6% (<jats:italic>n</jats:italic> = 8,250). Most subjects were male (67%, <jats:italic>n</jats:italic> = 342,571) and white (62%, <jats:italic>n</jats:italic> = 315,178). The AUCs of the four models ranged from 92.7 to 97.7 with XGBoost demonstrating the highest AUC. XGBoost demonstrated the highest accuracy of 97.7%.</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>Machine learning algorithms can be effectively utilized to build an accurate pediatric mortality prediction model that leverages variables easily obtained upon trauma admission.</jats:p></jats:sec>","journal":"Frontiers in Pediatrics","year":2025,"id":675862,"datarank":0.37087663960513656,"base_score":1.9459101490553132,"endowment":1.9459101490553132,"self_citation_contribution":0.29188652235829704,"citation_network_contribution":0.07899011724683952,"self_endowment_contribution":0.29188652235829704,"citer_contribution":0.07899011724683952,"corpus_percentile":null,"corpus_rank":null,"citation_count":6,"citer_count":6,"citers_with_citation_signal":3,"citers_with_endowment":3,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":null,"is_data_producer":false,"deposit_databanks":null,"is_oa":false,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":null,"fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1151756,"name":"Anish Murala","orcid":null,"position":1,"is_corresponding":false},{"id":1416712,"name":"Isabelle Decker","orcid":null,"position":2,"is_corresponding":false},{"id":295079,"name":"Karthik Rajasekaran","orcid":"0000-0003-2148-1643","position":3,"is_corresponding":false},{"id":417537,"name":"Alvaro Moreira","orcid":"0000-0003-3455-6608","position":4,"is_corresponding":false},{"id":1765943,"name":"Alex Deleon","orcid":null,"position":0,"is_corresponding":false}],"reference_count":0,"raw_metadata":{"has_enrichment":true,"resolved":true,"title":"Machine learning-based prediction of mortality in pediatric trauma patients","abstract":"<jats:sec><jats:title>Background</jats:title><jats:p>This study aimed to develop a predictive model for mortality outcomes among pediatric trauma patients using machine learning (ML) algorithms.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p>We extracted data on a cohort of pediatric trauma patients (18 years and younger) from the National Trauma Data Bank (NTDB). The main aim was to identify clinical and physiologic variables that could serve as predictors for pediatric trauma mortality. Data was split into a development cohort (70%) to build four ML models and then tested in a validation cohort (30%). The area under the receiver operating characteristic curve (AUC) was used to assess each model's performance.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>In 510,381 children, the gross mortality rate was 1.6% (<jats:italic>n</jats:italic> = 8,250). Most subjects were male (67%, <jats:italic>n</jats:italic> = 342,571) and white (62%, <jats:italic>n</jats:italic> = 315,178). The AUCs of the four models ranged from 92.7 to 97.7 with XGBoost demonstrating the highest AUC. XGBoost demonstrated the highest accuracy of 97.7%.</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>Machine learning algorithms can be effectively utilized to build an accurate pediatric mortality prediction model that leverages variables easily obtained upon trauma admission.</jats:p></jats:sec>","is_dataset_classified":null,"base_score":1.9459101490553132,"endowment":1.9459101490553132,"datacite_reuse_total":0,"file_count":0,"downloads":0,"views":0,"has_version_chain":false,"is_dataset":false,"is_oa":false,"pmid":"40083437","pmcid":"PMC11905922","openalex_id":"https://openalex.org/W4408020194","authors":[],"funders":[],"total_grants":0,"fwci":7.8567,"citation_percentile":0.97536224,"influential_citations":0,"citation_trend":[{"year":2025,"count":2},{"year":2026,"count":4}],"oa_status":"gold","license":"cc-by","oa_locations":[{"url":"https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2025.1522845/pdf","host_type":"journal"},{"url":"https://www.frontiersin.org/journals/pediatrics/articles/10.3389/fped.2025.1522845/pdf","host_type":"publisher"},{"url":"https://www.frontiersin.org/articles/10.3389/fped.2025.1522845/full","host_type":"publisher"},{"url":"https://doi.org/10.3389/fped.2025.1522845","host_type":"journal"},{"url":"https://pubmed.ncbi.nlm.nih.gov/40083437","host_type":"repository"},{"url":"https://www.ncbi.nlm.nih.gov/pmc/articles/11905922","host_type":"repository"},{"url":"https://doaj.org/article/fdb5223229b5444cb5ada3d7a6d097e1","host_type":"repository"},{"url":"https://europepmc.org/articles/PMC11905922","host_type":"Europe_PMC"},{"url":"https://europepmc.org/articles/PMC11905922?pdf=render","host_type":"Europe_PMC"}],"fields_of_study":["Trauma and Emergency Care Studies","Abdominal Trauma and Injuries","Trauma, Hemostasis, Coagulopathy, Resuscitation"],"mesh_terms":[],"keywords":["Medicine","Emergency medicine","Medical emergency","Pediatrics","Prediction","Mortality","Trauma","Machine Learning"],"sdg_mappings":[],"linked_datasets":[],"clinical_trials":[],"software_tools":[],"database_accessions":[],"source":"live","citation_network_status":"fetched"},"created_at":"2026-08-17T01:47:41.960374Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}