An innovative top-to-toe (TTT) screening system for early detecting sarcopenia

Sarcopenia, a condition characterized by the gradual decline of muscle mass and strength, predominantly affects elderly individuals and those leading sedentary lifestyles. Sarcopenia not only heightens the risk of falls, fractures, and hospitalizations among the elderly but also diminishes their quality of life while imposing a significant financial burden on healthcare systems. To address this challenge, our study aimed to develop an innovative integrated measuring system coupled with intelligent-based analysis to evaluate sarcopenia-related changes in neuromuscular functions and behaviors.We enrolled a total of 43 participants in this study, comprising 22 healthy controls, 7 individuals diagnosed with sarcopenia, and 14 with pre-sarcopenia. During the study, participants' electroencephalography (EEG), electromyography (EMG), and plantar pressure signals were simultaneously recorded while walking. After noise rejection preprocessing, event-related potentials extraction, and feature extraction to distinguish healthy controls from those with pre-sarcopenia or sarcopenia, support vector machine (SVM) algorithms were employed. The performance of the SVM model was evaluated using leave-one-out cross-validation.Our findings revealed that features extracted from EMG signals of the tibialis anterior (TA) muscle achieved an accuracy rate of approximately 75% in distinguishing sarcopenic individuals from healthy controls. Conversely, features indicative of distinguishing healthy controls from those with pre-sarcopenia were identified in EEG signals from channels C3 and C4, yielding an accuracy of around 70%. These results underscore the potential of features identified in EMG and EEG signals as crucial indicators for future medical diagnoses, offering insights into the assessment and management of sarcopenia.