Table 3. Summary of methodologies used in the investigations.

AI: artificial intelligence, ML: machine learning

Methodology	Description	Percentage of studies
Imaging Data Analysis with AI/ML Models	Utilized advanced AI/ML models like convolutional neural networks and radiomics analysis to analyze imaging data from modalities such as MRI, PET, and CT for diagnosis, prognosis, and prediction.	24%
Animal Models and In Vitro Assays	Employed animal models and in vitro assays, including techniques like immunohistochemistry and electrophysiological recordings, to evaluate nerve injury, therapeutic approaches, and drug efficacy.	24%
Analysis of Electronic Health Records and Patient Data	Applied AI/ML techniques to examine electronic health records, demographic information, and clinical laboratory tests to construct diagnostic, prognostic, and predictive models.	12%
Research on Robotic Devices for Rehabilitation	Utilized robotic systems and devices to deliver specific stimulation patterns for rehabilitating nerve damage and restoring normal function.	12%
Computational Modeling and Simulations	Employed computational techniques like ML and mathematical modeling to explore pharmacological targets and neural signaling.	6%
Analysis of Human Subject Data	Conducted data collection and analysis on human subjects, including symptoms, neuropsychological assessments, and imaging data, to develop AI/ML models.	6%