Vharoone Sukumar, Salman Syed, Joyce Jeyaratnam
supervised by Dr.Kassam & Dr.Umapathy
Parkinson's disease is the most common neurodegenerative among those 50 and older. It has an estimated worldwide prevalence of 3-12 million, with another 0.5-1.3 million novel cases every year. In Parkinson's disease, idiopathic neuronal death leads to difficulty in skeletal muscle motor control. The hallmark symptoms of the disease include muscle rigidity, bradykinesia, postural instability, and resting tremors, the last of which begins first in one of the arms and then in the contralateral leg. The Unified Parkinson's Disease Rating Scale piles Parkinson's patients into discrete categories based on symptom presentation; this is imprecise since it does not take into account individual variations, as well as subjective since it depends on the independent assessment of various clinicians who come from diverse background experiences. To date, Parkinson's does not have a cure and treatment is limited to controlling symptoms and preventing disease progression. In addition, as the disease advances, patients' symptoms become more pronounced and their responsiveness to treatment reduces, ultimately becoming fatal as patients are unable to control their breathing and swallowing. Thus, in order to ameliorate diagnosis and introduce impartiality, we will procure a personal tremor-data logger that could be worn as a wrist-watch band that can record, quantify, and enable the analysis of the frequency and range of Parkinson's tremors. This will provide an invaluable tool to collect patient data in order to objectively assess disease stage and progression, as well as observe patient responsiveness to treatments and drugs.
Patients' involuntary muscle motion will be measured as EMG signals by a single MEMSic accelerometer (translational motion in the x- and y-axes) which will be placed on the arm. This accelerometer will be configured as a wearable wristwatch band. Since live patient testing is not possible to be carried out during the course of the academic year, vibrations that are characteristic of patient frequencies will be simulated using a mechanical jig.
The data from the accelerometer will be processed by dsPIC33F microcontroller. The microcontroller will receive data directly from the accelerometer and will subsequently process this signal with Fast Fourier Transform (FFT). FFT can distinguish vibrations of different frequencies and therefore can be applied to distinguish the different disease. Additionally, processing at this stage also allows the rejection of data obtained from voluntary motion (i.e. artefacts). Additionally once the signal is processed the data will be transmitted via XBee device which will be connected to the microcontroller, thereafter this data will be transmitted to the coordinator device.
Once data is transferred to a PC, it will be further processed to filter noise and other spurious signals caused by ambient electrical interference. Subsequently, the pertinent contents of the EMG signals such as its amplitude and frequency can be quantified and analyzed by LABView.
Project targeted applications:The applications of this project include tremor analysis of limb tremors of Parkinson's disease patients thus allowing physicians to distinguish from other patients with different type of tremors.