Application of a novel nested ensemble algorithm in predicting motor function recovery in patients with traumatic cervical spinal cord injury

Traumatic cervical spinal cord injury (TCSCI) often leads to motor dysfunction, assessed using the ASIA Impairment Scale. Predicting motor function recovery is crucial for planning effective treatments and rehabilitation. This study introduces a novel nested ensemble algorithm that uses early motor scores to predict motor function recovery six months post-injury in TCSCI patients. The algorithm combines multiple machine learning models in two stages, enhancing prediction accuracy and reliability, which can help personalize care for TCSCI patients.

• 1
  The nested ensemble algorithm achieved an accuracy of 80.6% and an F1 score of 80.6% in predicting motor function recovery after TCSCI.
• 2
  The confusion matrix showed a low false-negative rate, indicating practical implications for prognostic prediction of TCSCI.
• 3
  The novel algorithm, based on early AMS, is a useful tool for predicting motor function recovery 6 months after TCSCI.