Department of Electrical and Information Engineering, University of Cassino and Southern Lazio, 03043 Cassino, Italy.
Department of Medicine and Health Sciences "Vincenzo Tiberio", University of Molise, 86100 Campobasso, Italy.
Sensors (Basel). 2021 Jun 18;21(12):4196. doi: 10.3390/s21124196.
Reliable diagnosis of early-stage Parkinson's disease is an important task, since it permits the administration of a timely treatment, slowing the progression of the disease. Together with non-motor symptoms, other important signs of disease can be retrieved from the measurement of the movement trajectory and from tremor appearances. To measure these signs, the paper proposes a magnetic tracking system able to collect information about translational and vibrational movements in a spatial cubic domain, using a low-cost, low-power and highly accurate solution. These features allow the usage of the proposed technology to realize a portable monitoring system, that may be operated at home or in general practices, enabling telemedicine and preventing saturation of large neurological centers. Validation is based on three tests: movement trajectory tracking, a rest tremor test and a finger tapping test. These tests are considered in the Unified Parkinson's Disease Rating Scale and are provided as case studies to prove the system's capabilities to track and detect tremor frequencies. In the case of the tapping test, a preliminary classification scheme is also proposed to discriminate between healthy and ill patients. No human patients are involved in the tests, and most cases are emulated by means of a robotic arm, suitably driven to perform required tasks. Tapping test results show a classification accuracy of about 93% using a k-NN classification algorithm, while imposed tremor frequencies have been correctly detected by the system in the other two tests.
可靠的早期帕金森病诊断是一项重要任务，因为它可以及时进行治疗，减缓疾病的进展。除了非运动症状外，还可以从运动轨迹的测量和震颤表现中提取出其他重要的疾病迹象。为了测量这些迹象，本文提出了一种磁跟踪系统，该系统能够在空间立方域中收集关于平移和振动运动的信息，使用低成本、低功耗和高精度的解决方案。这些特性允许使用所提出的技术来实现便携式监测系统，该系统可以在家中或一般实践中操作，实现远程医疗并防止大型神经中心的饱和。验证基于三个测试：运动轨迹跟踪、静止震颤测试和手指敲击测试。这些测试被认为是在统一帕金森病评定量表中，并作为案例研究来证明系统跟踪和检测震颤频率的能力。在敲击测试的情况下，还提出了一种初步的分类方案，以区分健康患者和患病患者。测试中不涉及任何人类患者，大多数情况都是通过机器人手臂模拟的，机器人手臂被适当驱动以执行所需的任务。使用 k-NN 分类算法，敲击测试的分类准确率约为 93%，而在另外两个测试中，系统正确地检测到了施加的震颤频率。
