Embodied Neuromorphic Intelligence in Healthcare: Evaluating Pose-Matching Interaction Using fNIRS and Behavioral Data

In the era of Industry 5.0, the rapid development of the Internet of Things (IoT) is expected to extend its applications to broader human-computer interaction (HCI) and human-machine connectivity. With the increasing number of healthcare groups, there is an urgent need to develop embodied neuromorphic intelligent human-machine connectivity products based on these technologies. However, it remains a critical challenge to address the influencing factors of such product designs and how to quantify interaction efficacy. This study proposes a product framework combining IoT with embodied neuromorphic intelligence and conducts a user study. A cognitive rehabilitation product was developed using Leap Motion technology, with gesture recognition difficulty as a design variable, and product efficacy was quantified using a combination of brain-computer interaction and multi-source interactive feedback. Fifteen elderly and fifteen young participants engaged in puppet control tasks under resting, simple, and complex conditions. The study compared brain activation levels, brain network connectivity, and eight behavioral indicators. The results demonstrated that the difficulty level significantly affects interaction efficacy. This research reveals neurological changes in the rehabilitation process of healthcare groups and opens new directions for the design and efficacy evaluation of embodied neuromorphic intelligence in HCI rehabilitation products through IoT and big data analytics, thereby advancing the development of Healthcare Industry 5.0.