Academic Journal
Rapid prediction of cerebral edema on CT scan after traumatic brain injury.
| Τίτλος: | Rapid prediction of cerebral edema on CT scan after traumatic brain injury. |
|---|---|
| Συγγραφείς: | Felix RB; Fischell Department of Bioengineering, University of Maryland College Park, College Park, Maryland, USA.; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Podell J; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Urquizo K; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Alamdeen Y; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Beagan D; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Cheema M; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Parikh GY; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Teeter W; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Emergency Medicine, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Charles McC. Mathias, Jr., National Study Center for Trauma & Emergency Medical Systems, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Chen H; Epidemiology and Public Health, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Chen LK; Information Systems, University of Maryland Baltimore County, Baltimore, Maryland, USA., Vesselinov R; Charles McC. Mathias, Jr., National Study Center for Trauma & Emergency Medical Systems, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Epidemiology and Public Health, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Fisher JP; Fischell Department of Bioengineering, University of Maryland College Park, College Park, Maryland, USA., Yang S; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Department of Anesthesiology, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Institute of Health Computing, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Hu P; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Department of Anesthesiology, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA.; Institute of Health Computing, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA., Badjatia N; Program in Trauma, University of Maryland Baltimore School of Medicine, Baltimore, Maryland, USA. |
| Πηγή: | Trauma surgery & acute care open [Trauma Surg Acute Care Open] 2026 Apr 08; Vol. 11 (2), pp. e002021. Date of Electronic Publication: 2026 Apr 08 (Print Publication: 2026). |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: BMJ Country of Publication: England NLM ID: 101698646 Publication Model: eCollection Cited Medium: Internet ISSN: 2397-5776 (Electronic) Linking ISSN: 23975776 NLM ISO Abbreviation: Trauma Surg Acute Care Open Subsets: PubMed not MEDLINE |
| Imprint Name(s): | Original Publication: [London] : BMJ, [2016]- |
| Περίληψη: | Background: Traumatic brain injury (TBI) affects 1.7 million individuals annually in the USA, with cerebral edema (CE) as a critical determinant of outcomes and neurosurgical interventions. Delays in TBI diagnosis and triage in prehospital or resource-limited settings contribute to suboptimal care. This study evaluated the predictive potential of machine learning models using clinical and physiological data to detect CE on initial head CT scans, addressing this gap. Methods: We conducted a mixed retrospective and prospective study of 1222 suspected TBI patients. Data included clinical characteristics, radiographic scores (Marshall and Rotterdam), and physiological features derived from ECG and photoplethysmography signals collected within 1 hour of admission. Statistical models employed logistic regression and gradient boosting, using Shapley additive explanations for feature importance. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC). Results: The models demonstrated high predictive accuracy for CE-related radiographic features, including midline shift and cisternal abnormalities (AUROC: 0.79-0.85). Clinical features such as Glasgow Coma Scale components and intubation status, combined with physiological variables like heart rate variability, contributed significantly to predictions. In contrast, predictions for other findings (eg, epidural hematoma) showed lower discriminatory power. Prehospital applicability was highlighted by the reliance on readily available physiological data. Conclusions: Machine learning models effectively predict CE prior to CT scans, offering a rapid decision support tool for triage and neurosurgical prioritization in austere and resource-limited settings. Early identification of CE could improve patient outcomes by optimizing transport and treatment strategies. Future research should focus on multicenter validation and streamlined data collection to enhance generalizability and clinical utility. Level of Evidence: Level III, Prognostic. (Copyright © Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.) |
| Competing Interests: | None declared. |
| References: | Neurotherapeutics. 2011 Jan;8(1):39-53. (PMID: 21274684) Eur J Emerg Med. 2016 Feb;23(1):61-4. (PMID: 26375361) J Neuroimaging. 2016 Sep;26(5):489-93. (PMID: 27159772) World Neurosurg. 2021 Apr;148:80-89. (PMID: 33412317) BMJ. 2020 Mar 18;368:m441. (PMID: 32188600) J Bioinform Comput Biol. 2005 Apr;3(2):185-205. (PMID: 15852500) Sci Rep. 2024 Mar 31;14(1):7618. (PMID: 38556518) J Emerg Nurs. 2009 Apr;35(2):e5-40. (PMID: 19285163) Bioinformatics. 2001 Jun;17(6):520-5. (PMID: 11395428) Neurosurgery. 2012 Jul;71(1):80-5. (PMID: 22382208) J Trauma Acute Care Surg. 2015 Nov;79(5):838-42. (PMID: 26317818) Am Surg. 2018 Feb 1;84(2):201-207. (PMID: 29580346) Injury. 2022 Jun;53(6):2060-2068. (PMID: 35190184) Pediatr Crit Care Med. 2014 Jul;15(6):554-62. (PMID: 24751786) PLoS One. 2018 Mar 29;13(3):e0195087. (PMID: 29596477) Sci Rep. 2023 Jan 9;13(1):403. (PMID: 36624110) J Med Syst. 2017 Jan;41(1):3. (PMID: 27817131) MMWR Morb Mortal Wkly Rep. 2013 Jul 12;62(27):549-52. (PMID: 23842444) Arch Acad Emerg Med. 2024 Sep 13;13(1):e9. (PMID: 39465060) J Trauma Acute Care Surg. 2016 Jun;80(6):897-906. (PMID: 27027555) Stat Methods Med Res. 2019 Aug;28(8):2455-2474. (PMID: 29966490) J Trauma Acute Care Surg. 2021 Feb 1;90(2):268-273. (PMID: 33502145) Circulation. 1996 Mar 1;93(5):1043-65. (PMID: 8598068) Front Neurol. 2020 Oct 30;11:599268. (PMID: 33193067) Front Neurol. 2018 Sep 25;9:761. (PMID: 30319521) Ann Emerg Med. 1990 Nov;19(11):1280-7. (PMID: 2240725) J Neurotrauma. 1992 Mar;9 Suppl 1:S287-92. (PMID: 1588618) Acad Radiol. 2014 May;21(5):605-11. (PMID: 24703472) J Neurosurg. 2024 May 10;:1-10. (PMID: 39076152) Neurosurgery. 2005 Dec;57(6):1173-82; discussion 1173-82. (PMID: 16331165) J Neurotrauma. 2011 Sep;28(9):1727-38. (PMID: 21355816) J Trauma Acute Care Surg. 2012 Feb;72(2):487-90. (PMID: 22327988) Emerg Med J. 2007 Aug;24(8):588-91. (PMID: 17652691) |
| Contributed Indexing: | Keywords: Cerebral Edema; Machine learning; traumatic brain injury |
| Entry Date(s): | Date Created: 20260413 Date Completed: 20260413 Latest Revision: 20260813 |
| Update Code: | 20260813 |
| PubMed Central ID: | PMC13064127 |
| DOI: | 10.1136/tsaco-2025-002021 |
| PMID: | 41971763 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 2397-5776 |
|---|---|
| DOI: | 10.1136/tsaco-2025-002021 |