Predicting postoperative neurological outcomes of degenerative cervical myelopathy based on machine learning
We developed models based on machine learning to predict postoperative neurological outcomes. The LightGBM model presented the best predictive power regarding the surgical outcomes of DCM patients. Feature importance analysis revealed that variables, including age, preoperative JOA score, SF-36-PF, SF-36-GH, and SF-36-BP, were essential factors in the model. The feature-reduced LightGBM model, designed for ease of application, achieved nearly the same predictive power with fewer variables.
Authors
- Yanbin Zhao (ORCID: https://orcid.org/0000-0003-0153-9764)
- Yu Sun (ORCID: https://orcid.org/0000-0002-4656-7032)
- H Q Xi (ORCID: https://orcid.org/0000-0002-6918-1212)
- Yinze Diao (ORCID: https://orcid.org/0000-0003-3250-9445)
- Zexiang Liu (ORCID: https://orcid.org/0000-0001-8020-1619)
- Shuai Zhou (ORCID: https://orcid.org/0000-0002-0136-2770)
- Haoge Huang
- X. Chen (ORCID: https://orcid.org/0009-0001-1449-5628)
- Xiao Fan (ORCID: https://orcid.org/0000-0001-5147-6701)
- Hong Ji (ORCID: https://orcid.org/0000-0001-8814-580X)
- Feifei Zhou (ORCID: https://orcid.org/0000-0001-9114-9123)
Institutions
- Ministry of Education of the People's Republic of China (CN)
- Peking University (CN)
- Beijing Emergency Medical Center (CN)
- Peking University Third Hospital (CN)
Publication Details
- Journal
- Frontiers in Bioengineering and Biotechnology
- Published
- 2025-03-04
- DOI
- https://doi.org/10.3389/fbioe.2025.1529545
- Citations
- 2
- Primary Topic
- Cervical and Thoracic Myelopathy
- Type
- article
- Field-Weighted Citation Impact
- 4.01
Funders
- Peking University Third Hospital