XGBoost, a Machine Learning Method, Predicts Neurological Recovery in Patients with Cervical Spinal Cord Injury.
cervical spinal cord injury
extreme gradient boosting
machine learning
receiver operating curve
Journal
Neurotrauma reports
ISSN: 2689-288X
Titre abrégé: Neurotrauma Rep
Pays: United States
ID NLM: 101773091
Informations de publication
Date de publication:
2020
2020
Historique:
entrez:
5
7
2021
pubmed:
6
7
2021
medline:
6
7
2021
Statut:
epublish
Résumé
The accurate prediction of neurological outcomes in patients with cervical spinal cord injury (SCI) is difficult because of heterogeneity in patient characteristics, treatment strategies, and radiographic findings. Although machine learning algorithms may increase the accuracy of outcome predictions in various fields, limited information is available on their efficacy in the management of SCI. We analyzed data from 165 patients with cervical SCI, and extracted important factors for predicting prognoses. Extreme gradient boosting (XGBoost) as a machine learning model was applied to assess the reliability of a machine learning algorithm to predict neurological outcomes compared with that of conventional methodology, such as a logistic regression or decision tree. We used regularly obtainable data as predictors, such as demographics, magnetic resonance variables, and treatment strategies. Predictive tools, including XGBoost, a logistic regression, and a decision tree, were applied to predict neurological improvements in the functional motor status (ASIA [American Spinal Injury Association] Impairment Scale [AIS] D and E) 6 months after injury. We evaluated predictive performance, including accuracy and the area under the receiver operating characteristic curve (AUC). Regarding predictions of neurological improvements in patients with cervical SCI, XGBoost had the highest accuracy (81.1%), followed by the logistic regression (80.6%) and the decision tree (78.8%). Regarding AUC, the logistic regression showed 0.877, followed by XGBoost (0.867) and the decision tree (0.753). XGBoost reliably predicted neurological alterations in patients with cervical SCI. The utilization of predictive machine learning algorithms may enhance personalized management choices through pre-treatment categorization of patients.
Identifiants
pubmed: 34223526
doi: 10.1089/neur.2020.0009
pii: 10.1089/neur.2020.0009
pmc: PMC8240917
doi:
Types de publication
Journal Article
Langues
eng
Pagination
8-16Informations de copyright
© Tomoo Inoue et al., 2020; Published by Mary Ann Liebert, Inc.
Déclaration de conflit d'intérêts
No competing financial interests exist.
Références
AMIA Annu Symp Proc. 2007 Oct 11;:955
pubmed: 18694055
J Clin Epidemiol. 1992 Jun;45(6):613-9
pubmed: 1607900
Circulation. 2015 Nov 17;132(20):1920-30
pubmed: 26572668
Arch Phys Med Rehabil. 2017 Dec;98(12):2385-2392
pubmed: 28647550
Spinal Cord. 2020 Jun;58(6):682-688
pubmed: 31992857
Spine (Phila Pa 1976). 1999 Mar 15;24(6):605-13
pubmed: 10101829
J Neurotrauma. 2011 Aug;28(8):1371-99
pubmed: 20001726
World Neurosurg. 2016 Mar;87:124-31
pubmed: 26724625
J Clin Epidemiol. 2020 Jun;122:56-69
pubmed: 32169597
Neurospine. 2019 Dec;16(4):678-685
pubmed: 31905456
J Neurotrauma. 2015 Sep 15;32(18):1385-92
pubmed: 25658291
IEEE J Biomed Health Inform. 2019 Jul 26;:
pubmed: 31369388
Front Pharmacol. 2019 Oct 07;10:1155
pubmed: 31649533
Spine J. 2020 Feb;20(2):213-224
pubmed: 31525468
J Neurotrauma. 2014 Feb 1;31(3):284-91
pubmed: 24020382
World Neurosurg. 2015 May;83(5):867-78
pubmed: 23524031
Spine J. 2019 Apr;19(4):703-710
pubmed: 30179672
Global Spine J. 2020 Jan;10(1 Suppl):84S-91S
pubmed: 31934526
Spine (Phila Pa 1976). 2002 Aug 1;27(15):E348-55
pubmed: 12163735
Spinal Cord. 2009 Aug;47(8):582-91
pubmed: 19381157
J Neurotrauma. 2020 Jan 1;37(1):202-210
pubmed: 31359814
J Chronic Dis. 1987;40(5):373-83
pubmed: 3558716
AJNR Am J Neuroradiol. 2019 Apr;40(4):737-744
pubmed: 30923086
JMIR Med Inform. 2017 Nov 22;5(4):e45
pubmed: 29167089
Spine (Phila Pa 1976). 1990 Mar;15(3):161-8
pubmed: 2353251
Spine J. 2014 Aug 1;14(8):1601-10
pubmed: 24411833
Acad Emerg Med. 2016 Mar;23(3):269-78
pubmed: 26679719
Am J Emerg Med. 2018 Sep;36(9):1650-1654
pubmed: 29970272
J Neurotrauma. 2011 Aug;28(8):1401-11
pubmed: 20388006
AJNR Am J Neuroradiol. 2017 Mar;38(3):648-655
pubmed: 28007771
Clin Orthop Relat Res. 2017 May;475(5):1499-1504
pubmed: 27815685
Neuroepidemiology. 2012;38(4):219-26
pubmed: 22555590
Comput Math Methods Med. 2014;2014:276589
pubmed: 24575150
Asian Spine J. 2015 Oct;9(5):748-56
pubmed: 26435794
IEEE/ACM Trans Comput Biol Bioinform. 2020 Nov-Dec;17(6):2131-2140
pubmed: 30998478
JAMA Netw Open. 2019 Jan 4;2(1):e186937
pubmed: 30646206
J Clin Epidemiol. 2001 Oct;54(10):979-85
pubmed: 11576808
Neurosurgery. 2017 Apr 1;80(4):610-620
pubmed: 28362913
J Neurosurg Spine. 2015 Oct;23(4):495-504
pubmed: 26161519
Brain Inform. 2017 Sep;4(3):159-169
pubmed: 28434153
Global Spine J. 2011 Dec;1(1):1-8
pubmed: 24353930
Spinal Cord. 2015 Apr;53(4):265-77
pubmed: 25665542
Circ Cardiovasc Qual Outcomes. 2016 Nov;9(6):629-640
pubmed: 28263938
Neurospine. 2019 Dec;16(4):643-653
pubmed: 31905452
BMC Syst Biol. 2018 Nov 22;12(Suppl 6):105
pubmed: 30463545