Model for Predicting In-Hospital Mortality of Physical Trauma Patients Using Artificial Intelligence Techniques: Nationwide Population-Based Study in Korea

Abstract

Background Physical trauma–related mortality places a heavy burden on society. Estimating the mortality risk in physical trauma patients is crucial to enhance treatment efficiency and reduce this burden. The most popular and accurate model is the Injury Severity Score (ISS), which is based on the Abbreviated Injury Scale (AIS), an anatomical injury severity scoring system. However, the AIS requires specialists to code the injury scale by reviewing a patient's medical record; therefore, applying the model to every hospital is impossible. Objective We aimed to develop an artificial intelligence (AI) model to predict in-hospital mortality in physical trauma patients using the International Classification of Disease 10th Revision (ICD-10), triage scale, procedure codes, and other clinical features. Methods We used the Korean National Emergency Department Information System (NEDIS) data set (N=778,111) compiled from over 400 hospitals between 2016 and 2019. To predict in-hospital mortality, we used the following as input features: ICD-10, patient age, gender, intentionality, injury mechanism, and emergent symptom, Alert/Verbal/Painful/Unresponsive (AVPU) scale, Korean Triage and Acuity Scale (KTAS), and procedure codes. We proposed the ensemble of deep neural networks (EDNN) via 5-fold cross-validation and compared them with other state-of-the-art machine learning models, including traditional prediction models. We further investigated the effect of the features. Results Our proposed EDNN with all features provided the highest area under the receiver operating characteristic (AUROC) curve of 0.9507, outperforming other state-of-the-art models, including the following traditional prediction models: Adaptive Boosting (AdaBoost; AUROC of 0.9433), Extreme Gradient Boosting (XGBoost; AUROC of 0.9331), ICD-based ISS (AUROC of 0.8699 for an inclusive model and AUROC of 0.8224 for an exclusive model), and KTAS (AUROC of 0.1841). In addition, using all features yielded a higher AUROC than any other partial features, namely, EDNN with the features of ICD-10 only (AUROC of 0.8964) and EDNN with the features excluding ICD-10 (AUROC of 0.9383). Conclusions Our proposed EDNN with all features outperforms other state-of-the-art models, including the traditional diagnostic code-based prediction model and triage scale.

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

“…Our previous study [ 5 ] revealed that blood pressure, heart rate, body temperature, and other vital signs weakened the previous AI model’s performance, whereas incorporating vital signs strengthened our present AI model. This observation implies that ED and in-hospital mortality patients exhibit differing data distributions.…”

“…Thus, patients with multiple injuries were excluded from exclusive SRR calculations [ 20 ]. We used the survival probability determined from our previous study [ 5 ] because other studies [ 20 , 21 ] did not use ED mortality.…”

“…Our research team previously developed two AI models for predicting in-hospital trauma patient mortality [ 4 , 5 ]; one utilizes Abbreviated Injury Scale (AIS) codes and the other incorporates the International Classification of Diseases, 10th revision (ICD-10) codes and other variables from the Korean National Emergency Department Information System (NEDIS) data set. Deceased ED patients were excluded from these models, as we assumed they may have received an insufficient workup or indicate an inaccurate diagnosis in previous studies.…”

“…We concluded that ED mortality should be predicted using a different patient data set than in-hospital mortality, and that the new model should also have alternative weights and input variables. Therefore, in this study, we developed AI models for predicting ED mortality in trauma patients using the NEDIS data set that was not incorporated into the previous AI models [ 4 , 5 ].…”

Accuracy of a Prehospital Triage Protocol in Predicting In-Hospital Mortality and Severe Trauma Cases among Older Adults

“…Such a strategy may greatly help clinicians in facilitating patient risk assessment and triaging and optimize resource use. Elsewhere, machine-learning-based models predicted the need for urgent neurosurgery [44] or trauma mortality [45]. The potential clinical benefits of such models as decision-making and triage tools deserve further assessment in the prehospital environment.…”