EVALUATION OF MACHINE LEARNING AND DEEP LEARNING TECHNIQUES APPLIED IN SCI INJURY

A SYSTEMATIC REVIEW

Resumo

The application of technologies associated with the Fourth Industrial Revolution, particularly Machine Learning (ML) and Deep Learning (DL), has expanded into multiple fields of knowledge, promoting greater efficiency in addressing complex professional challenges. In the medical and physiotherapeutic domains, one of the most significant difficulties lies in the effective treatment of Spinal Cord Injury (SCI). In this context, the integration of Artificial Intelligence (AI) techniques into support systems emerges as a promising strategy to enhance therapeutic outcomes, while simultaneously addressing issues related to information security. This study presents a systematic review, following the PRISMA-P methodology, to identify and analyze AI-based techniques applied to SCI treatment. A total of 168 studies were initially identified, and after applying the eligibility criteria, 12 articles (7.14%) were selected for detailed analysis. The findings reveal a variety of therapeutic approaches involving Brain-Machine Interfaces (BMIs), which face challenges related to the heterogeneity of computational systems, sensors, and actuators used in healthcare applications.