Optimal methods of stimulus presentation and frequency analysis in P300-based brain-computer interfaces for patients with severe motor impairment.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to Ryuji Neshige.
Explore the source record for details and available documents.
For the purpose of determining the optimal dose, pergolide was administered to patients with Parkinson's disease using a dose-escalation protocol up to the maximum dosage permissible in Japan. The study was completed in 16 subjects with Parkinson's disease. Prior to the start of the dose-escalation study, the daily dose of pergolide was 500 microg in 6 patients and 750 microg in 10 patients. The daily dose was raised by 250 microg at 4-week intervals up to the approved maximum dose of 1,250 microg/day, which was used as the final level. Just before the start of study and after administration for 4 weeks at each dose level, UPDRS-III and PDQ-39 assessments were performed. The mean UPDRS-III score was 28.6 +/- 7.9 before the start of the study and 17.3 +/- 8.3 after the end, indicating an improvement (improvement rate, 39.3%; p < 0.01). The mean PDQ-39 score was 45.2 +/- 21.8 before the start and 37.3 +/- 17.8 after the end, indicating an improvement (improvement rate, 17.5%; p < 0.05). Clinical conditions including their QOL improved significantly after dose-escalation up to the 1,250 microg.
We had developed a communication system using a theory of P300 through a computer. 17 healthy subjects participated in all 3 experiments, while 4 ALS and 1 OPCA patients took part in the first 2 experiments only. In the 1st experiment, 4 short sentences of 'yes very much', 'yes' 'no' and 'can't answer' were shown on the display. The subjects were asked to select one of them as target. During the experiment 1, color of background was randomly changed for each sentence as the stimulation for the period of 300 msec and the interstimulus interval was 1,500 msec. In the 2nd experiment, 46 letters of Japanese character and several symbols were shown on the display (5 columns x 10 rows). In the 3rd experiment, we used 16 pictures together with X (4 columns x 5 rows). During recordings, the color of background was randomly changed for each row and then for each letter/picture of column containing target in experiment 2 as well as in experiment 3. The subjects were asked to start counting and when the target was stimulated, they were instructed to stop the counting process and repeat counting again until they encountered the stimulation of target again. This was repeated for the entire recordings. Computer defined a target by a biggest positive voltage between 300 and 600 msec and/or by the most powerful spectrum in 2-5 Hz after Fast Fourier transform in an averaged P300 (averaging number = 5-7). We then calculated the percentage of correct target by comparing the result and actual target. Results indicate that this communication aids can be used conveniently in patients with severe motor disabilities.