| 摘要 |
PURPOSE: To provide a device for estimating a living body ental state capable of reducing burden on an examinee and detecting the content of a category desired to inform by the examinee. CONSTITUTION: Learning data is generated from electromagnetic field distribution calculated by an electromagnetic field distribution calculation part 6 based on head model data and ental state category data to store the relation of an in-brain activity part and a living body internal state by a learning data generating part 7, and it is learned by a neural network in a neural network part 10 via a data normalizing part 3 and an input buffer 4. After learning is completed, a noise in the electromagnetic field distribution on the scalp of a living body measured by an electromagnetic field distribution measuring part 1 is eliminated by a wavelet conversion part 2, and an in-brain activating area is estimated by the neural network, and it is converted to expression representing an internal state by a unit output conversion part 8. |
