An Artificial Intelligence Model for Predicting Trauma Mortality Among Emergency Department Patients in South Korea: Retrospective Cohort Study

被引:10
作者
Lee, Seungseok [1 ]
Kang, Wu Seong [2 ]
Kim, Do Wan [3 ]
Seo, Sang Hyun [4 ]
Kim, Joongsuck [2 ]
Jeong, Soon Tak [5 ]
Yon, Dong Keon [6 ,7 ]
Lee, Jinseok [1 ]
机构
[1] Kyung Hee Univ, Dept Biomed Engn, 1732 Deogyeong daero, Yongin 17104, South Korea
[2] Cheju Halla Gen Hosp, Jeju Reg Trauma Ctr, Dept Trauma Surg, Jeju, South Korea
[3] Chonnam Natl Univ, Chonnam Natl Univ Hosp, Dept Thorac & Cardiovasc Surg, Med Sch, Gwangju, South Korea
[4] Wonkwang Univ Hosp, Dept Radiol, Iksan, South Korea
[5] Ansanhyo Hosp, Dept Phys Med & Rehabil, Ansan, South Korea
[6] Kyung Hee Univ, Kyung Hee Univ Med Ctr, Dept Pediat, Coll Med, Seoul, South Korea
[7] Kyung Hee Univ, Med Res Inst, Ctr Digital Hlth, Med Ctr, Seoul, South Korea
关键词
artificial intelligence; trauma; mortality prediction; international classification of disease; emergency department; ICD; model; models; mortality; predict; prediction; predictive; emergency; death; traumatic; nationwide; national; cohort; retrospective; NEURAL-NETWORK; CONSCIOUSNESS; LEVEL;
D O I
10.2196/49283
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
Background: Within the trauma system, the emergency department (ED) is the hospital's first contact and is vital for allocating medical resources. However, there is generally limited information about patients that die in the ED. Objective: The aim of this study was to develop an artificial intelligence (AI) model to predict trauma mortality and analyze pertinent mortality factors for all patients visiting the ED. Methods: We used the Korean National Emergency Department Information System (NEDIS) data set (N=6,536,306), incorporating over 400 hospitals between 2016 and 2019. We included the International Classification of Disease 10th Revision (ICD-10) codes and chose the following input features to predict ED patient mortality: age, sex, intentionality, injury, emergent symptom, Alert/Verbal/Painful/Unresponsive (AVPU) scale, Korean Triage and Acuity Scale (KTAS), and vital signs. We compared three different feature set performances for AI input: all features (n=921), ICD-10 features (n=878), and features excluding ICD-10 codes (n=43). We devised various machine learning models with an ensemble approach via 5-fold cross-validation and compared the performance of each model with that of traditional prediction models. Lastly, we investigated explainable AI feature effects and deployed our final AI model on a public website, providing access to our mortality prediction results among patients visiting the ED. Results: Our proposed AI model with the all-feature set achieved the highest area under the receiver operating characteristic curve (AUROC) of 0.9974 (adaptive boosting [AdaBoost], AdaBoost + light gradient boosting machine [LightGBM]: Ensemble), outperforming other state-of-the-art machine learning and traditional prediction models, including extreme gradient boosting (AUROC=0.9972), LightGBM (AUROC=0.9973), ICD-based injury severity scores (AUC=0.9328 for the inclusive model and AUROC=0.9567 for the exclusive model), and KTAS (AUROC=0.9405). In addition, our proposed AI model outperformed a cutting-edge AI model designed for in-hospital mortality prediction (AUROC=0.7675) for all ED visitors. From the AI model, we also discovered that age and unresponsiveness (coma) were the top two mortality predictors among patients visiting the ED, followed by oxygen saturation, multiple rib fractures (ICD-10 code S224), painful response (stupor, semicoma), and lumbar vertebra fracture (ICD-10 code S320). Conclusions: Our proposed AI model exhibits remarkable accuracy in predicting ED mortality. Including the necessity for external validation, a large nationwide data set would provide a more accurate model and minimize overfitting. We anticipate that our AI-based risk calculator tool will substantially aid health care providers, particularly regarding triage and early diagnosis for trauma patients.
引用
收藏
页数:17
相关论文
共 50 条
[41]   Mortality and blood pressure in emergency patients with traumatic brain injury: a retrospective cohort study [J].
Juliana Pizza-Restrepo, Maria ;
Tatiana Buritica-Montoya, Iris ;
Hinestroza-Cordoba, Daniela ;
Guzman-Martinez, Santiago ;
Felipe Hurtado-Guerra, Luis ;
Mario-Ruiz, Rafael ;
Jaimes, Fabian .
IATREIA, 2016, 29 (04) :407-414
[42]   Predicting early mortality in hemodialysis patients: a deep learning approach using a nationwide prospective cohort in South Korea [J].
Noh, Junhyug ;
Park, Sun Young ;
Bae, Wonho ;
Kim, Kangil ;
Cho, Jang-Hee ;
Lee, Jong Soo ;
Kang, Shin-Wook ;
Kim, Yong-Lim ;
Kim, Yon Su ;
Lim, Chun Soo ;
Lee, Jung Pyo ;
Yoo, Kyung Don .
SCIENTIFIC REPORTS, 2024, 14 (01)
[43]   High mortality among tuberculosis patients on treatment in Nigeria: a retrospective cohort study [J].
Adamu, Aishatu L. ;
Gadanya, Muktar A. ;
Abubakar, Isa S. ;
Jibo, Abubakar M. ;
Bello, Musa M. ;
Gajida, Auwalu U. ;
Babashani, Musa M. ;
Abubakar, Ibrahim .
BMC INFECTIOUS DISEASES, 2017, 17
[44]   Artificial intelligence for predicting 30-day mortality after surgery for femoral shaft fractures: A retrospective study [J].
Gupta, Puneet ;
Shen, Hong-Jui ;
Patel, Kunj ;
Guo, Rui ;
Heinz, Eric R. ;
Manyam, Rameshbabu .
INDIAN JOURNAL OF ANAESTHESIA, 2025, 69 (06) :606-614
[45]   Mortality Among Patients With Polymyalgia Rheumatica: A Retrospective Cohort Study [J].
Partington, Richard ;
Muller, Sara ;
Mallen, Christian D. ;
Sultan, Alyshah Abdul ;
Helliwell, Toby .
ARTHRITIS CARE & RESEARCH, 2021, 73 (12) :1853-1857
[46]   The role of marijuana use disorder in predicting emergency department and inpatient encounters: A retrospective cohort study [J].
Campbell, Cynthia I. ;
Bahorik, Amber L. ;
Kline-Simon, Andrea H. ;
Satre, Derek D. .
DRUG AND ALCOHOL DEPENDENCE, 2017, 178 :170-175
[47]   A new simplified model for predicting 30-day mortality in older medical emergency department patients: The rise up score [J].
Zelis, Noortje ;
Buijs, Jacqueline ;
de Leeuw, Peter W. ;
van Kuijk, Sander M. J. ;
Stassen, Patricia M. .
EUROPEAN JOURNAL OF INTERNAL MEDICINE, 2020, 77 :36-43
[48]   Lower mortality from suicidal trauma among patients with a psychiatric diagnosis upon admission: Nationwide japanese retrospective cohort study [J].
Ishida, Takuto ;
Kuwahara, Yusuke ;
Shibahashi, Keita ;
Okura, Yoshihiro ;
Sugiyama, Kazuhiro ;
Hamabe, Yuichi ;
Mimura, Masaru ;
Suzuki, Takefumi ;
Uchida, Hiroyuki .
PSYCHIATRY RESEARCH, 2020, 293
[49]   A study of "left against medical advice" emergency department patients: an optimized explainable artificial intelligence framework [J].
Ahmed, Abdulaziz ;
Aram, Khalid Y. ;
Tutun, Salih ;
Delen, Dursun .
HEALTH CARE MANAGEMENT SCIENCE, 2024, 27 (04) :485-502
[50]   Lactate level, etiology, and mortality of adult patients in an emergency department: a cohort study [J].
Mathilde Pedersen ;
Vibeke Brandt ;
Jon G Holler ;
Annmarie T Lassen .
Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine, 23 (Suppl 1)