基于深度学习的渐冻症患者实时监测系统
A Real-Time Monitoring System for ALS Based on Deep Learning

渐冻症(ALS)是一种严重影响运动神经元的神经退行性疾病，其早期症状通常表现为肌肉无力、抽搐和运动障碍。为了提高对渐冻症患者健康状况的实时监测能力，对病人进行24小时无接触看护。本研究采用深度卷积网络技术对图片进行跟踪运算，设计了实时监测系统，捕捉并甄别患者微表情、微动作来进行判断并报警，及时通知陪护人员前来处理危急情况，避免被看护人员发生危险。研究结果表明：该方法具有较高的检测精度和良好的实时性，可有效识别渐冻症患者的异常状态，显著提升健康监测的准确性与可靠性。本研究为渐冻症患者提供了一种便捷高效的监测手段，为个性化健康管理与早期干预提供了新的技术支持。
ALS (Amyotrophic Lateral Sclerosis) is a neurodegenerative disease that seriously affects motor neurons, and its early symptoms usually manifest as muscle weakness, twitches, and movement disorders. In order to improve the real-time monitoring ability of the health condition of ALS patients, patients are monitored without physical contact for 24 hours. This study uses deep convolutional neural network technology to perform tracking calculations on images and designs a real-time monitoring system that captures and identifies patients’ micro-expressions and micro-motions to make judgments and issue alarms. It notifies caregivers to handle emergencies within 3 seconds, thereby avoiding danger to the monitored person. The results show that the method has high detection accuracy and good real-time performance, can effectively identify abnormal states of ALS patients, and significantly improves the accuracy and reliability of health monitoring. This study provides a convenient and efficient monitoring method for ALS patients and provides new technical support for personalized health management and early intervention.