著者

宮崎 信次

出版者

バイオメカニズム学会

雑誌

バイオメカニズム (ISSN:13487116)

巻号頁・発行日

no.17, pp.227-234, 2004-08-25

被引用文献数

2 3

---

Several mouse emulator devices are commercially available for patients with high-level cervical cord injury (CCI), muscular dystrophy, and rheumatoid arthritis. These include head mounting pointers which linearly convert the 2-dimensional deviation of head motion to the 2-axis movements of the cursor. Limitations of inherent accuracy of the neck movement controllability restrict the use of these pointers to people with a good range of motion of the trunk and neck. Other mouse emulators utilize a miniature 2-dimensional position sensor which can be activated by mouth, then convert the information to the cursor movements. When the transformation is from 2-dimensional deviations to the position of the cursor, a problem similar to the head mounting pointers arises. When the 2-dimensional deviation is converted to the velocity of the cursor, the problem is solved. However, the currently available devices of this last type have other problems, e.g., overshoot and drift of the cursor during no input. The purpose of this study was to develop an inexpensive mouse emulator device utilizing a mini joystick and analog pressure sensor, and to test its clinical utility in shortterm experiments and middle-term field tests conducted by volunteer monitors. The study also aimed to combine this mouse emulator with commercially available automatic Japanese voice recognition software to enable those who cannot use their fingers for striking keyboards. The new mouse emulator, named "Joystick," converts the 2-dimensional angular deviation nonlinearly, i.e., stepwise parabolically, to the velocity of the cursor, and also provides a dead movement zone and automatic tracking of mechanically neutral position to prevent drift of the cursor during a resting state. A thorough investigation was made to find the automatic voice recognition software best suited for the present purpose. Japan IBM's Via Voice with ATOK15 was selected. Four CCI patients and four healthy subjects participated in the short-term evaluation of the device. The first task was to control the movement of the cursor around the monitor, and to drag and drop certain portions of the text. The second task was to activate Word, define text format, input predetermined text, revise it, and store it as a file with a name. The performance of each subject using the new device was compared with the performance of one C4 CCI subject using a conventional mouth stick and a track ball. The results were promising. Twelve volunteer monitors participated in the mid-term field evaluation of the device, and 6 out of 9 monitors wanted to buy the device after the 2- to 4-week monitor period, which proved that the device is useful for a large part of the subject population at which the present study was aimed.