Research Information System
2nd International Conference on E-Health and TeleMedicine (ICEHTM 2014), İstanbul, Turkey, 22 - 24 May 2014, pp.171-175, (Full Text)
The monitoring of the progression of Parkinson’s disease (PD) is held on frequently in the clinic and an inconvenient and time-consuming process since PD is generally observed mostly in elderly people whose physical visits to the clinic are troublesome, and the physical examinations must be performed by trained medical staff. Besides, from the machine learning perspective, building a generalizable non-invasive PD diagnosis and monitoring decision support system requires enough number of training samples collected from PD patients at regular intervals.Therefore, self-administered and non-invasive telemonitoring applications that enable the PD patients to collect data at home and transmit it over the internet to a dedicated server have recently become popular. In addition to these, the use of computer systems to collect data from subjects enables the researchers to develop various tests which cannot be easily performed with paper and pencil. In this study, considering that PD affects the handwriting motor abilities of patients, we collected handwriting samples of PD patients who have admitted to the Department of Neurology in Cerrahpaşa Faculty of Medicine, Istanbul University via a graphics tablet which was used in many biomedical studies before like cancer imaging, influence of stress on servical myelopathy, and brain imaging. In addition to the static spiral drawing test (SST) which can also be performed traditionally with pencil and paper, we propose a new test called dynamic spiral test (DST) which, unlike SST, can only be performed with the use of electronic equipments such as a tablet and a computer. We present the comparative results of SST and DST, and show that handwriting samples collected with a computerized system can be used to build generalizable PD telemonitoring systems.