A diagnostic model of nerve root compression localization in lower lumbar disc herniation based on random forest...

Area of Science:

• Biomedical Engineering
• Neurology
• Orthopedics

Background:

• Lumbar disc herniation (LDH) is a common condition causing nerve root compression.
• Accurate localization of the compressed nerve root is crucial for effective treatment.
• Current diagnostic methods may have limitations in precisely identifying the affected nerve root.

Purpose of the Study:

• To investigate muscle activation patterns in LDH patients during walking using surface electromyography (SEMG).
• To develop and validate a diagnostic model employing the random forest (RF) algorithm for precise nerve root localization in LDH.
• To assess the diagnostic performance of the established SEMG-based RF model.

Main Methods:

• Recruited 58 LDH patients and 30 healthy controls.
• Collected bilateral SEMG data from tibialis anterior (TA) and lateral gastrocnemius (LG) during walking.
• Analyzed SEMG parameters including RMS-peak, RMS-peak time, MPF, and MF.
• Developed an RF model using eight SEMG parameters and validated it through repeated experiments.
• Evaluated model performance using accuracy, precision, recall, F1-score, Kappa, and ROC curve analysis.

Main Results:

• Distinct SEMG patterns were observed: delayed TA activation and decreased LG activation in the L5 group; decreased and earlier LG activation in the S1 group.
• The RF model achieved an average accuracy of 84% and an area under the ROC curve of 0.93.
• The RMS peak time of the TA muscle was identified as the most significant SEMG parameter for diagnosis.

Conclusions:

• The developed RF model effectively assists in the localization diagnosis of compressed nerve roots in LDH patients.
• SEMG parameters offer valuable data for refining and optimizing future diagnostic models for LDH.
• This approach shows promise for improving the diagnostic accuracy and efficiency in managing LDH.