Establishment of real-time diagnostic system for needle electromyography by audio features
| Project/Area Number |
17K09800
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Neurology
|
|---|
| Research Institution | Kanazawa Medical University (2019-2020)
The University of Tokushima (2017-2018) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-04-01 – 2021-03-31
|
| Project Status |
Completed (Fiscal Year 2020)
|
| Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
|
|---|
| Keywords | 針筋電図 / 人工知能 / 音特徴量 / 自動判別 / AI / 機械学習 / ディープラーニング / 針筋電図検査 / 安静時放電 / 神経内科 / 神経筋疾患 |
|---|
| Outline of Final Research Achievements |
Resting potentials of needle electromyography (needle EMG) obtained from patients with neuromuscular diseases were databased. The classified waveforms were divided into 2-second audio files. Method 1) Audio characteristics were obtained from each audio file. Machine learning methods was used to classify the six resting potentials. The accuracy was 90.4%. Method 2) The same database as the method 1 was used. The audio information was transformed into melspectrogram as image files. The images were divided into training and test data. The training data were then trained with convolutional neural networks (CNNs). Image augmentation was useful in that the accuracy was 100%.
|
| Academic Significance and Societal Importance of the Research Achievements |
針筋電図の放電判別は専門医以外には容易ではなく、正確な診断を妨げている。本研究により機械学習やディープラーニングを用いることで高精度に判別が可能であることが分かった。今後は実用化に向けて対象となる波形を増やしていく。
|
Report
(5 results)
Research Products
(2 results)
