学术急诊医学档案

卷 14 编号 1 (2026)

Machine Learning Models for Predicting In-Hospital Mortality and Pres-sure Ulcer Development following Traumatic Spinal Cord Injury; A Cross-sectional Study

学术急诊医学档案, 卷 14 编号 1 (2026), 1 十月 2025 , 第 e49 页
https://doi.org/10.22037/aaem.v14i1.2989

摘要

Introduction: Traumatic spinal cord injury (TSCI) remains a significant cause of long-term disability, with mortality and pressure ulcer (PU) development being key determinants of patient outcomes. This study aimed to develop and interpret machine learning (ML) models for predicting in-hospital mortality and PU formation in TSCI patients using data from the National Spinal Cord Injury Registry of Iran (NSCIR-IR). Methods: Patients with TSCI admitted between 2015 and 2023 were included in the analysis. Data preprocessing included iterative imputation, feature selection with cross-validation, and class balancing using SMOTE. Nine ML algorithms were trained and validated on an 80:20 split dataset. Model performance was assessed using the area under the receiver operating characteristic curve (AUC) and other standard metrics. Feature importance and interpretability were evaluated using permutation importance and Shapley Additive Explanations (SHAP) analysis. Results: A total of 499 patients were included. The LightGBM classifier achieved the highest performance in mortality prediction (AUC = 0.85; 95% CI: 0.73-0.94), followed by KNN and Naive Bayes. For PU prediction, the Random Forest model performed best (AUC = 0.83; 95% CI: 0.72-0.93), outperforming XGBoost and LightGBM. Ventilator use, ASIA grade, and sensory and motor scores were the strongest predictors of mortality, while first aid given, ventilator use, and the number of injured vertebrae were the strongest predictors of PU risk. SHAP analysis confirmed these findings. Conclusion: ML algorithms can accurately identify TSCI patients at high risk of mortality and PU development using routinely collected clinical data. Integrating such interpretable ML tools into early triage systems could enable timely preventive interventions and improve patient outcomes.

关键词:
  • Traumatic spinal cord injury
  • Machine Learning
  • Mortality
  • Pressure Ulcer

参考

1. Ahuja CS, Wilson JR, Nori S, Kotter M, Druschel C, Curt A, et al. Traumatic spinal cord injury. Nat Rev Dis Primers. 2017;3(1):17018.

2. Singh A, Tetreault L, Kalsi-Ryan S, Nouri A, Fehlings MG. Global prevalence and incidence of traumatic spinal cord in-jury. Clin Epidemiol. 2014;6:309-31.

3. Maleki MS, Khedri B, Roodposhti ME, Majdabadi HA, Seyedrezaei SO, Amanat N, et al. Epidemiology of traumatic spinal cord injuries in Iran; a systematic review and meta-analysis. Arch Acad Emerg Med. 2022;10(1):e80.

4. Wilczweski P, Grimm D, Gianakis A, Gill B, Sarver W, McNett M. Risk factors associated with pressure ulcer devel-opment in critically ill traumatic spinal cord injury patients. J Trauma Nurs. 2012;19(1):5-10.

5. Ahuja CS, Badhiwala JH, Fehlings MG. “Time is spine”: the importance of early intervention for traumatic spinal cord in-jury. Spinal Cord. 2020;58(9):1037-9.

6. Wilson JR, Cadotte DW, Fehlings MG. Clinical predictors of neurological outcome, functional status, and survival after traumatic spinal cord injury: a systematic review. J Neuro-surg Spine. 2012;17(1 Suppl):11-26.

7. Joseph C, Nilsson Wikmar L. Prevalence of secondary med-ical complications and risk factors for pressure ulcers after traumatic spinal cord injury during acute care in South Afri-ca. Spinal Cord. 2016;54(7):535-9.

8. Dietz N, Vaitheesh J, Alkin V, Mettille J, Boakye M, Drazin D. Machine learning in clinical diagnosis, prognostication, and management of acute traumatic spinal cord injury (SCI): A systematic review. J Clin Orthop Trauma. 2022;35:102046.

9. Aly S, Chen Y, Ahmed A, Wen H, Mehta T. Utilization of machine learning algorithm in the prediction of rehospitaliza-tion during one-year post traumatic spinal cord injury. Spinal Cord. 2025;63(4):214-21.

10. Fallah N, Noonan VK, Waheed Z, Rivers CS, Plashkes T, Bedi M, et al. Development of a machine learning algorithm for predicting in-hospital and 1-year mortality after traumat-ic spinal cord injury. Spine J. 2022;22(2):329-36.

11. Zajac KK, Schubauer K, Simman R. The unavoidable pressure injury/ulcer: a review of skin failure in critically ill patients. J Wound Care. 2024;33(Sup9): S18-S22.

12. Kottner J, Beeckman D. Incontinence-associated dermati-tis and pressure ulcers in geriatric patients. G Ital Dermatol Venereol. 2015;150(6):717-29.

13. Shimizu T, Inomata K, Suda K, Matsumoto Harmon S, Komatsu M, Ota M, et al. A multimodal machine learning model integrating clinical and MRI data for predicting neuro-logical outcomes following surgical treatment for cervical spi-nal cord injury. Eur Spine J. 2025;34(9):3747-55.

14. Wang Y, Luo X, Wang J, Li W, Cui J, Li Y. Development and Validation of Machine Learning Models for Predicting 7-Day Mortality in Critically Ill Patients with Traumatic Spinal Cord Injury: A Multicenter Retrospective Study. Neurocrit Care. 2026;44(1):176-90.

15. Mabray MC, Talbott JF, Whetstone WD, Dhall SS, Phillips DB, Pan JZ, et al. Multidimensional Analysis of Magnetic Res-onance Imaging Predicts Early Impairment in Thoracic and Thoracolumbar Spinal Cord Injury. J Neurotrauma. 2016;33(10):954-62.

16. Talbott JF, Whetstone WD, Readdy WJ, Ferguson AR, Bresnahan JC, Saigal R, et al. The Brain and Spinal Injury Center score: a novel, simple, and reproducible method for assessing the severity of acute cervical spinal cord injury with axial T2-weighted MRI findings. J Neurosurg Spine. 2015;23(4):495-504.

17. Sizheng Z, Boxuan H, Feng X, Dianying Z. A functional outcome prediction model of acute traumatic spinal cord in-jury based on extreme gradient boost. J Orthop Surg Res. 2022;17(1):451.

18. Kitagawa K, Maki S, Furuya T, Shiratani Y, Nagashima Y, Maruyama J, et al. Development of a machine learning model and a web application for predicting neurological outcome at hospital discharge in spinal cord injury patients. Spine J. 2025;25(7):1483-93.

  • 摘要 ##plugins.themes.ojsPlusA.frontend.article.viewed##: 1 ##plugins.themes.ojsPlusA.frontend.article.times##
  • pdf (English) ##plugins.themes.ojsPlusA.frontend.article.downloaded##: 0 ##plugins.themes.ojsPlusA.frontend.article.times##

##plugins.themes.ojsPlusA.frontend.article.downloadstatastics##

##plugins.block.makeSubmission.linkLabel##

##plugins.block.makeSubmission.linkLabel##

SJR

SCImago Journal & Country Rank

COPE

最新一期

消息