Journal of Clinical Medicine, Journal Year: 2024, Volume and Issue: 13(15), P. 4503 - 4503

Published: Aug. 1, 2024

Background: Prediction of neurorehabilitation outcomes after a Spinal Cord Injury (SCI) is crucial for healthcare resource management and improving prognosis rehabilitation strategies. Artificial neural networks (ANNs) have emerged as promising alternative to conventional statistical approaches identifying complex prognostic factors in SCI patients. Materials: database 1256 patients admitted was analyzed. Clinical demographic data characteristics were used predict functional using both ANN linear regression models. The former structured with input, hidden, output layers, while the identified significant variables affecting outcomes. Both aimed evaluate compare their accuracy measured by Independence Measure (SCIM) score. Results: models key predictors outcomes, such age, injury level, initial SCIM scores (correlation actual outcome: R = 0.75 0.73, respectively). When also alimented parameters recorded during hospitalization, highlighted importance these additional factors, like motor completeness complications showing an improvement its (R 0.87). Conclusions: seemed be not widely superior classical statistics general, but, taking into account non-linear relationships among variables, emphasized impact hospitalization on recovery, particularly respiratory issues, deep vein thrombosis, urological complications. These results suggested that recovery

Language: Английский