Rough Set improved Therapy-Based Metaverse Assisting System

Jin Cao, Yanhui Jiang, Chang Yu, Feiwei Qin, Zekun Jiang

Introduction

Chronic neck and shoulder pain (CNSP) represents a significant global health challenge, impacting individual quality of life and broader economic productivity. While traditional interventions—ranging from physiotherapy to pharmacological treatments—are widely available, they often suffer from high costs, low precision, and significant recurrence rates. A critical oversight in conventional treatment is the neglect of the psychological dimension of chronic pain. Traditional methods frequently fail to address the cognitive-behavioral cycles—where thoughts, feelings, and physical sensations reinforce one another—that exacerbate physical discomfort.

This paper proposes an interactive, metaverse-based assistive system designed to bridge this gap. By integrating physiological sensing with Cognitive Behavioral Therapy (CBT) principles, the system offers a personalized, home-based solution for pain management. The core technical innovation involves the application of Rough Set Theory to optimize noisy Electromyography (EMG) sensor data, ensuring high-precision feedback within a virtual therapeutic environment.

System Architecture

The proposed system is integrated into a "Metaverse" platform and consists of two primary modules designed to create a closed-loop therapeutic experience:

1. Pain Detection Module

The hardware foundation utilizes an Arduino-based setup equipped with Electromyography (EMG) sensors and Inertial Measurement Units (IMU). The EMG sensors measure muscle activation levels, while the IMUs track posture and movement patterns. This dual-sensor approach allows the system to capture a comprehensive snapshot of the user's physical state in real-time.

2. Cognitive Therapy Module

Drawing on CBT principles, this module provides physical interventions such as heat and massage therapy. These interventions are specifically designed to stimulate the hippocampus and increase serotonin levels, addressing both the physiological symptoms of pain and the accompanying mental distress. By situating these therapies within a metaverse environment, the system enhances treatment immersion, making home-based rehabilitation more engaging and easier to monitor remotely.

Technical Innovation: Rough Set Theory

Bio-signals, particularly EMG data, are notoriously "noisy" and characterized by uncertainty. Standard averaging techniques often fail to capture the nuances of muscle fatigue or tension accurately. To solve this, the researchers employ Rough Set Theory to process raw data. This mathematical approach allows the system to:

  • Establish Indiscernibility Relations: Grouping data points that are functionally identical to simplify the decision-making space.
  • Calculate Attribute Importance: Filtering out irrelevant or redundant data points to focus on the most significant indicators of pain.
  • Screen Meaningful Samples: Enhancing the precision of the pain detection algorithm by focusing on high-quality data subsets.

By applying weight parameters ($\alpha$ and $eta$) to balance data dependence and importance, the system normalizes raw signals into actionable insights for the therapy module.

Closing the Feedback Loop

Unlike traditional wearable devices that provide static interventions (e.g., a heating pad that stays at one temperature), this system creates a dynamic feedback loop. Physiological data from the EMG sensors and subjective feedback from the user (via button presses) directly influence the intensity and duration of the therapy.

This approach treats the patient's mental state as a critical component of the system. By incorporating CBT, the goal is to break the cycle where physical pain leads to anxiety, which in turn increases muscle tension. The metaverse serves as a clinical space where the environment can respond to the user's real-time physiological state, providing a more "acceptable and practical" alternative to frequent clinical visits.

Evaluation and Future Directions

As the paper primarily details the system design and the logic of the underlying algorithms, clinical outcomes are currently based on an experimental framework. The authors have outlined a plan to validate the system with 15–20 participants with a history of chronic pain.

Preliminary results focus on the mathematical robustness of the framework, demonstrating that the Rough Set algorithms successfully normalize raw EMG signals for real-time assessment. However, limitations remain, including the small initial sample size and the potential complexity of the hardware setup for non-technical users.

Future work will focus on refining these algorithms through advanced machine learning techniques and conducting broader clinical trials to validate the long-term efficacy of metaverse-based therapy in reducing the recurrence of chronic neck and shoulder pain.

Summary updated

BibTeX
@inproceedings{cao2024rough,
  author = {Jin Cao and Yanhui Jiang and Chang Yu and Feiwei Qin and Zekun Jiang},
  title = {{Rough Set improved Therapy-Based Metaverse Assisting System}},
  booktitle = {Metacom},
  year = {2024},
  doi = {10.1109/metacom62920.2024.00066},
}
Figure