Top > Search of International Patents > Apparatus and method to determine a decision of a user based on brain wave analysis

Apparatus and method to determine a decision of a user based on brain wave analysis meetings

Foreign code F160008783
Posted date Aug 3, 2016
Country EPO
Application number 11821758
Gazette No. 2613222
Gazette No. 2613222
Date of filing Aug 30, 2011
Gazette Date Jul 10, 2013
Gazette Date May 1, 2019
International application number JP2011069524
International publication number WO2012029742
Date of international filing Aug 30, 2011
Date of international publication Mar 8, 2012
Priority data
  • P2010-195463 (Sep 1, 2010) JP
  • 2011JP69524 (Aug 30, 2011) WO
Title Apparatus and method to determine a decision of a user based on brain wave analysis meetings
Abstract According to a conventional BMI technology for controlling external equipment or transmitting an intention to another person by focusing on a biosignal such as brain activity, there has been the problem that a large-sized apparatus is required, the operation method is complex from the user's viewpoint, and noise is large. The present invention provides an apparatus and method such that an intention in the brain can be analyzed with high accuracy and at high speed and transmitted in real-time. A communication assist apparatus according to the present invention comprises an apparatus for presenting a visual stimulus on a display screen and the like, and a processing apparatus for processing brain wave data from an electroencephalograph that measures a brain wave after stimulus presentation by the presenting apparatus. The processing apparatus determines that a specific intention decision has been made in the brain when the product of an accumulated discrimination score according to a discriminant analysis function obtained by analyzing the brain wave data and a success rate exceeds a threshold value, and then outputs a determination result to a device.
Outline of related art and contending technology BACKGROUND
Currently, in the fields of information science, medical engineering applications, and welfare device control by brain activity, operation of a device is mainly performed through an input operation of various switches, a joystick, a mouse and the like by hand. Meanwhile, in the field of nursing care/welfare device development, a need is felt for a device that can be operated by other parts of the body for operators with physically disabled hands or legs.
Conventionally, speech assisting devices using a character board or picture cards are known. Technologies related to devices for assisting the conversation of users with disabilities in language function or hearing are discussed in Patent Documents 1 and 2, for example. Patent Document 1 describes a technology such that the letters of the 50 Japanese syllabary on a character board are sequentially lighted on a block by block basis and selected, and a technology involving the use of a message board for selecting a request for bodily care. Patent Document 2 describes a communication assist system for selecting a symbol display by using a portable information terminal. However, in the field of nursing care/welfare device development, a device that enables the direct transmission of an intention without requiring the conventional operations is desired for operators who cannot perform complex input operations, such as the elderly or the sick. There is also a need for a device for patients with speech impediment or the elderly that enables the transmission of an intention regarding basic personal nursing care or feelings, to the helper in a simpler manner.
The progress in brain science in recent years has led to various studies on the relationship between human thought and behavior and brain activity. Attention is being focused on Brain-Machine Interface (BMI) technology for controlling an external device or conveying an intention to others by noting a biosignal, such as brain activity.
For example, neuromarketing is the study for investigating the brain activity related to consumer behavior by using a functional MRI (fMRI) device and the like. A study is known in which, through a brain activity measurement experiment using a fMRI apparatus, brain regions associated with preferences or brand awareness for Coca cola and Pepsi cola were identified and brain activity differences were investigated (Non-patent Document 1).
The present inventors have shown, through a measurement technique involving extracellular recording of the action potential of a single neuron by electrodes disposed in an animal brain, and by a neuronal ensemble activity simulation, that the representation of relationships among a plurality of external stimuli (experimental conditions) in the brain can be estimated in terms of low-dimensional spatial information (see Non-patent Document 2). However, many aspects of brain activity are still unknown, and the measurement methods have restrictions.
The present inventors have also proposed a virtual decision function and shown a calculation method therefor (see Non-patent Document 3). Non-patent Document 3 describes a method for predicting an alternative behavior from a single neuron activity as an example of neural activity.
Patent Document 1: JP-A-2004-286768
Patent Document 2: JP-A-2003-255825

Non-patent Document 1: McClure SM, et al., "Neural Correlates of Behavioral Preference for Culturally Familiar Drinks" Neuron 44, p379-387, 2004
Non-patent Document 2: N. Matsumoto and R. Hasegawa, "Prediction of Multidimensional Decision-making Based on Single Trial of Frontal Association Area Neuronal Ensemble", Neuroscience Research, Vol. 58, Supplement 1, page S161 (P2-f34), 2007
Non-patent Document 3: R. Hasegawa et al., "Single trial-based prediction of a go/no-go decision in monkey superior colliculus", Neural Networks 19(2006) 1223-1232
IEEE Transactions on Biomedical Engineering Vol. 55, No. 6 of June 2008 p1754-1763 shows Asynchronous P300-Based Brain-Computer Interfaces: A Computational Approach With Statistical Models by H Zhang, C Guan and C Wang. US Patent Application, Publication number US 2010/0145214 A1 shows a system and method for processing brain signals in a BCI system. US Patent Application, Publication number US 2009/0062679 A1 shows categorizing perceptual stimuli by detecting subconscious responses.
Scope of claims [claim1]
1. A communication assist apparatus comprising:
a stimulus-presenting apparatus (2); and
a processing apparatus (6) that processes brain wave data from an electroencephalograph that measures a brain wave of a user after stimulus presentation to the user by the stimulus-presenting apparatus (2),
characterized in that the processing apparatus (6) determines that a specific intention decision has been made in the brain on the basis of a discriminant function obtained by analyzing the brain wave data and a success rate, the success rate for a certain number of times of stimulus presentation being obtained by discriminant analysis of data obtained in a training session,
wherein the processing apparatus (6) determines that the specific intention decision has been made in the brain when the product of an accumulated discrimination score according to the discriminant function obtained by analyzing the brain wave data and the success rate exceeds a threshold value, and then outputs a determination result to a device.

[claim2]
2. The communication assist apparatus according to claim 1, wherein
the function obtained by analyzing the data measured by the electroencephalograph is a multivariate analysis function.

[claim3]
3. The communication assist apparatus according to claim 1, wherein
weighting of variables may be set for each brain wave channel and at intervals of time elapsed after stimulus presentation.

[claim4]
4. A communication assist method for execution on an electronic device comprising
receiving brain wave data from an electroencephalograph that measures a brain wave of a user after stimulus presentation to the user by a stimulus-presenting apparatus, the method being characterised in that it further comprises
determining that a specific intention decision has been made in the brain on the basis of a discriminant function obtained by analyzing the brain wave data and a success rate, the success rate for a certain number of times of stimulus presentation being obtained by discriminant analysis of data obtained in a training session, and
determining that the specific intention decision has been made in the brain when the product of an accumulated discrimination score according to the discriminant function obtained by analyzing the data measured by the electroencephalograph and the success rate exceeds a threshold value, and then outputting a determination result to a device.

[claim5]
5. The communication assist method according to claim 4, wherein
the function obtained by analyzing the data measured by the electroencephalograph is a multivariate analysis function.

[claim6]
6. The communication assist method according to claim 4, wherein
weighting of variables may be set for each brain wave channel and at intervals of time elapsed after stimulus presentation.
  • Applicant
  • NATIONAL INSTITUTE OF ADVANCED INDUSTRIAL SCIENCE AND TECHNOLOGY
  • Inventor
  • HASEGAWA Ryohei
  • HASEGAWA Yukako
  • TAKAI Hideaki
IPC(International Patent Classification)
Specified countries Contracting States: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

PAGE TOP

close
close
close
close
close
close