KR101703547B1 - Method and apparatus for estimating effective channel of functional near-infrared spectroscopy - Google Patents
Method and apparatus for estimating effective channel of functional near-infrared spectroscopy Download PDFInfo
- Publication number
- KR101703547B1 KR101703547B1 KR1020150079615A KR20150079615A KR101703547B1 KR 101703547 B1 KR101703547 B1 KR 101703547B1 KR 1020150079615 A KR1020150079615 A KR 1020150079615A KR 20150079615 A KR20150079615 A KR 20150079615A KR 101703547 B1 KR101703547 B1 KR 101703547B1
- Authority
- KR
- South Korea
- Prior art keywords
- channel
- channels
- blood flow
- effective
- classification
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000004497 NIR spectroscopy Methods 0.000 title claims abstract description 18
- 230000003727 cerebral blood flow Effects 0.000 claims abstract description 51
- 238000005259 measurement Methods 0.000 claims abstract description 40
- 230000007177 brain activity Effects 0.000 claims abstract description 33
- 238000004611 spectroscopical analysis Methods 0.000 claims abstract description 22
- 230000033001 locomotion Effects 0.000 claims abstract description 12
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 230000029058 respiratory gaseous exchange Effects 0.000 claims abstract description 9
- 230000017531 blood circulation Effects 0.000 claims abstract description 8
- 210000004761 scalp Anatomy 0.000 claims abstract description 7
- 230000006266 hibernation Effects 0.000 claims abstract description 4
- 210000004556 brain Anatomy 0.000 claims description 20
- 230000005059 dormancy Effects 0.000 claims description 18
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 230000003247 decreasing effect Effects 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 6
- 238000000059 patterning Methods 0.000 claims description 5
- 230000005021 gait Effects 0.000 description 28
- 230000006870 function Effects 0.000 description 15
- 238000012549 training Methods 0.000 description 15
- 238000011282 treatment Methods 0.000 description 13
- 102000001554 Hemoglobins Human genes 0.000 description 10
- 108010054147 Hemoglobins Proteins 0.000 description 10
- 210000003710 cerebral cortex Anatomy 0.000 description 10
- 208000006011 Stroke Diseases 0.000 description 9
- 230000004913 activation Effects 0.000 description 8
- 238000002560 therapeutic procedure Methods 0.000 description 7
- 238000002599 functional magnetic resonance imaging Methods 0.000 description 6
- 230000009467 reduction Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 5
- 210000000974 brodmann area Anatomy 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 230000001054 cortical effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000006185 dispersion Substances 0.000 description 3
- 210000003128 head Anatomy 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000002490 cerebral effect Effects 0.000 description 2
- 230000007012 clinical effect Effects 0.000 description 2
- 230000000004 hemodynamic effect Effects 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 230000007659 motor function Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000005316 response function Methods 0.000 description 2
- 206010017577 Gait disturbance Diseases 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000036992 cognitive tasks Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007585 cortical function Effects 0.000 description 1
- 238000002567 electromyography Methods 0.000 description 1
- 230000010482 emotional regulation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 210000003141 lower extremity Anatomy 0.000 description 1
- 210000000337 motor cortex Anatomy 0.000 description 1
- 230000004118 muscle contraction Effects 0.000 description 1
- 238000001584 occupational therapy Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000002600 positron emission tomography Methods 0.000 description 1
- 210000000976 primary motor cortex Anatomy 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000003238 somatosensory effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 230000004382 visual function Effects 0.000 description 1
- 230000004584 weight gain Effects 0.000 description 1
- 235000019786 weight gain Nutrition 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0082—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes
- A61B5/0084—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes for introduction into the body, e.g. by catheters
- A61B5/0086—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes for introduction into the body, e.g. by catheters using infrared radiation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0075—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/026—Measuring blood flow
- A61B5/0261—Measuring blood flow using optical means, e.g. infrared light
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1123—Discriminating type of movement, e.g. walking or running
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
Abstract
There is provided an effective positioning method according to the brain activity measurement of the functional near infrared ray spectroscopy. A method for effective positioning according to brain activity measurement of functional near-infrared spectroscopy includes obtaining cerebral blood flow signals from N measurement channels corresponding to a plurality of positions of scalp by functional near-infrared spectroscopy (fNIRS); Removing a noise component including respiration, blood circulation, and movement of the subject in the acquired cerebral blood flow signal; A pre-processing step of dividing the noise canceled signal into time series data for execution and stoppage, calculating a hibernation classification fitness and giving priority to the calculated classification fitness; And if the correlation of the NM number of inferior channels for each of the M honorable channels is greater than or equal to the first reference value, then the corresponding L number of inferior channels are classified into the superior channel Lt; RTI ID = 0.0 > M + L < / RTI > effective channels.
Description
Field of the Invention [0002] The present invention relates to an effective positioning method and apparatus for measuring brain activity of functional near-infrared spectroscopy (fNIRS), and more particularly, The present invention relates to a method and an apparatus for determining an effective measurement position closely related to a task being performed in acquiring brain activity from blood flow information of a human cerebral cortex measured using a functional NIR spectroscopy apparatus will be.
Classical rehabilitation treatments that have been used mainly for the gait rehabilitation of stroke patients include weight gain, weight shift, and balance training in a static state. However, this therapy improves muscle contraction and general condition, but abnormal gait continues to appear after treatment.
Recently, functional electrical stimulation, treadmill walking training, and robot - assisted walking therapy have been used as a new therapeutic approach for rehabilitation training to improve the walking pattern of stroke patients. It is known that when these treatments are performed concurrently with general treatment, significant improvement is shown in the results of gait characteristics.
In the present study, we used the Motoricity Index, the Fugl-Meyer Index, the Rivermead Motor Assessment, and the Functional Ambulatory Category (Holden et al., 1984, Collen et al., 1990, Demeurisse et al., 1980).
These evaluation items can evaluate clinical treatment effects on lower limb movements such as balance sense, joint range, and independent walking. However, as the motor function is restored from the viewpoint of brain plasticity, changes in cerebral cortical activity and reconstruction of motor neural network I could not evaluate it.
On the other hand, partial weighted treadmill rehabilitation training has been widely used as a method to improve gait after stroke, but the effect on it and the appropriate timing for rehabilitation therapy are not known.
In particular, according to previous studies (Cochrane Review 2012, Duncan et al., 2011, Morone et al., 2012), clinical treatment effects and optimal use of generalized floor walking rehabilitation, treadmill rehabilitation, Is not yet established.
Therefore, it is required to acquire the basic technology for the clinical evaluation of brain plasticity for the functional recovery of gait rehabilitation by examining the effect of clinical treatment on brain plasticity in treadmill gait rehabilitation .
In addition, fMRI (Functional MRI) and positron emission tomography (PET), which are used to measure brain activation for gait evaluation of gait rehabilitation, have been used for supine position (gait) Measurements are very limited, but gait under gravitational acceleration, an essential element of walking, can not be measured in real time.
In contrast, fNIRS and EEG (electroencephalogram) among non-invasive brain activation measurement methods, which are currently in the spotlight, can measure the gait training in real time. However, it is possible to measure the movement of the head, trunk, It is difficult to acquire a faithful brain activity signal including neurophysiological information related to gait.
Therefore, by measuring and analyzing not only the cerebral blood flow signals (fNIRS, EEG) but also the electromyography (EMG) and goniometer (IMU) signals, the physical characteristics of the cerebral blood flow signal And it is necessary to utilize it for more effective gait rehabilitation treatment method.
Another non-invasive method of measuring brain activation is the functional near-infrared spectroscopy, which measures the concentration of hemoglobin or oxidation of cerebral blood flow in the human cerebral cortex.
In general, functional near-infrared spectroscopy (FIR) is more cost-effective than functional magnetic resonance imaging (fMRI) and can be measured while the human being is moving. Therefore, various tasks related to clinical rehabilitation treatment or motion for patient's exercise rehabilitation training The brain is actively used for functional research, and the demand is increasing.
However, functional near-infrared (FIR) spectroscopy has a limitation in that it is difficult to accurately identify the active site due to its low spatial resolution compared with the functional magnetic resonance imaging (fMRI). Various methods have been tried to solve this problem, It depends on the viewpoint.
Therefore, it is desirable to develop a software technique that can accurately identify the active brain location regardless of the measurement method and the measurement device.
In order to solve the problems of the prior art as described above, one embodiment of the present invention uses a statistical technique and a learning technique for cerebral blood flow information measured by functional near-infrared spectroscopy, And a method and apparatus for effective positioning according to brain activity measurement of a functional near-infrared spectroscopy method which can be effectively determined.
According to an aspect of the present invention, there is provided an effective positioning method according to the measurement of brain activity of a functional near infrared ray spectroscopy. The method of determining the effective position according to the brain activity measurement of the functional near-infrared spectroscopy may include obtaining a cerebral blood flow signal from N measurement channels corresponding to a plurality of positions of the scalp by functional near-infrared spectroscopy (fNIRS); Removing noise components including respiration, blood circulation, and movement of the subject from the acquired cerebral blood flow signal; A pre-processing step of dividing the noise-removed signal into time-series data for performance and dormancy to calculate a performance-dormant classification fitness and giving priority to the calculated classification fitness; And classifying the M number of good channels and the number of NM poor channels according to the priority decreasing rate, and if the correlations of the NM number of poor channels with respect to each of the M number of good channels are equal to or greater than a first reference value, Lt; RTI ID = 0.0 > M + L < / RTI > effective channels.
In one embodiment, the pre-processing step comprises: sorting the noise canceled signal into time series data for performance and dormancy; Mapping the sorted data to a new classification space having independent parameters such as pause and hibernation using common space pattern (CSP); Generating a discrimination function for discriminating performance and dormancy for the channel on the classification space; And a step of computing an execution-pause classification fitness for each channel from the generated distinct function using an SVM (support vector machine) algorithm.
In one embodiment, the determining of the effective channel may include classifying the channel into the superior channel if the rate of decrease by priority is greater than the second reference value, and classifying the channel into the inferior channel if the rate is less than the second reference value. Calculating a one-to-one correlation of the inferior channel with respect to each of the superior channels; And reclassifying the poor channel having the calculated degree of correlation for all of the good channels to the good channel equal to or greater than the first reference value.
In one embodiment, the method of determining an effective position according to brain activity measurement of the functional near infrared ray spectroscopy may further include determining an associated light position for acquiring the cerebral blood flow signal according to the determined final effective channel.
According to an aspect of the present invention, there is provided an effective position determining apparatus for measuring brain activity of a functional near infrared ray spectroscopy. An effective position determining device according to the brain activity measurement of the functional near infrared ray spectroscopy includes a data acquiring part for acquiring a cerebral blood flow signal by functional near infrared ray spectroscopy (fNIRS); A data obtaining unit for obtaining a cerebral blood flow signal from N measurement channels corresponding to a plurality of positions of the scalp by functional near infrared ray spectroscopy (fNIRS); A noise eliminator for removing a noise component including respiration, blood circulation, and movement of the subject in the acquired cerebral blood flow signal; A pre-processing unit for dividing the noise-removed signal into time series data for performance and dormancy to calculate a performance-dormant classification fitness and giving priority to the calculated classification fitness according to the calculated classification fitness; And classifying the M number of good channels and the number of NM poor channels according to the priority decreasing rate, and if the correlations of the NM number of poor channels with respect to each of the M number of good channels are equal to or greater than a first reference value, And determines an M + L effective channel finally.
In one embodiment, the preprocessor may be configured to sort the noise canceled signal with time series data for performance and dormancy, and to perform the sorted data using common space patterning (CSP) And generates a discrimination function for distinguishing performance and pauses for the channel on the classification space, and calculates an execution-pause classification suitability for each channel from the generated discrimination function using the SVM algorithm .
In one embodiment, the valid channel determination unit classifies the effective channel into the good channel if the rate of decrease by priority is greater than the second reference value, classifies the channel into the poor channel if the rate is below the second reference value, To-one correlation of the inferior channel to reclassify the inferior channel having the calculated degree of correlation to the first reference value or higher with respect to all of the inferior channels.
In one embodiment, the effective channel determination unit may determine an associated light pole position for acquiring the cerebral blood flow signal according to the determined final effective channel.
The effective positioning method and apparatus according to the measurement of brain activity of the functional near infrared ray spectroscopy according to the embodiment of the present invention can provide information about the active position of the cerebral cortex which is clinically important in gait rehabilitation, , It is possible to select the measurement position according to the rehabilitation state of the patient.
In addition, the embodiment of the present invention can be utilized as a quasi-quantitative method for clinical evaluation of brain plasticity in gait rehabilitation therapy using physiological signals, and accordingly, Time of application, treatment period, etc.
In addition, the embodiment of the present invention can provide software that can accurately identify the brain activation position irrespective of the measuring instrument and the hardware.
1 is a flowchart of an effective positioning method according to a brain activity measurement of a functional near infrared ray spectroscopy according to an embodiment of the present invention.
FIG. 2 is a photograph of a state in which (a) a measurement apparatus is worn and (b) an electrode is installed, in order to apply the method of FIG.
FIG. 3 is a view showing a location of a cerebral blood flow signal measurement using a functional NIR spectroscope during a tremillal gait rehabilitation training of a stroke patient.
FIG. 4 is a graph showing a procedure for measuring cerebral blood flow signals during a tremile walking rehabilitation training in a stroke patient.
FIG. 5 is a graph showing the cerebral cortex brain activity image obtained from cerebral blood flow signal measurement during treadmill gait rehabilitation training in a stroke patient, (b) a graph showing changes in the concentration of oxidized hemoglobin in each channel, and (c) And the peak value of the concentration.
FIG. 6 is a graph showing a result of extracting effective channels using the method according to an embodiment of the present invention, (a) a brain activity image, (b) a graph of channel classification results, and (c) a change in the concentration of oxidized hemoglobin in each channel Graph, and (d) the peak value of the concentration of oxidized hemoglobin.
FIG. 7 is a schematic block diagram of an effective positioning apparatus according to an embodiment of the present invention for measuring brain activity of functional near-infrared spectroscopy.
Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art to which the present invention pertains. The present invention may be embodied in many different forms and is not limited to the embodiments described herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and the same or similar components are denoted by the same reference numerals throughout the specification.
The present invention relates to a method for precisely locating a brain active site, as opposed to observing only changes in the cerebral cortex during gait in gait rehabilitation therapy using a treadmill that has been actively studied.
In addition, the present invention is to be used for deriving an appropriate treatment method such as a clinical effect test of a robot-assisted walking therapy (Lokomat, Walkbot, etc.) using a treadmill in which clinical effects are being studied, Each cerebral blood flow signal measured from a plurality of independent measurement channels is divided into a time series of task-on period and a task-off period, ), And sorting and sorting these data, we can use statistical techniques to distinguish between performance and dormancy in individual channels. We can use this learning method to distinguish between individual channels The difference between performance and idle is numerically calculated, and the order of channels is determined according to the size of the discriminant value, and the number of good channels is cut off The correlation between the right channel and the inferior channel is calculated numerically by a statistical technique, and finally, the effective channel is selected including the inferior channels having a certain degree of correlation, so that the measurement position corresponding to the effective channel is performed To a brain active location that is directly associated with the task.
On the other hand, conventional studies for brain signal measurement and cerebral cortical activity analysis are mostly based on the Brodmann area (BA).
This Brodmann area (BA) is defined by dividing the cerebral cortex into 52 regions (Broadman region) by the cytoarchitectonic method. Here, the Broadman region includes an executive function, a motor function, a somatosensory, an attention, a visual function, a memory, an emotional regulation, And cortical functions such as sound.
On the other hand, the cerebral cortex area related to the gait function is 1, 2, 3, 4, 6, 8, and 9 regions for the Broadman region. The present invention suggests a method for deriving an appropriate treatment method by utilizing the cerebral blood flow signal for such a region.
Hereinafter, an effective positioning method according to the measurement of brain activity of functional NIR spectroscopy according to an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a flowchart of an effective positioning method according to the measurement of brain activity of functional NIR spectroscopy according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a state (a) FIG. 3 is a view showing a position of a cerebral blood flow signal measurement using a functional near-infrared ray spectroscopic apparatus during a tremillal gait rehabilitation training of a stroke patient, and FIG. 4 is a view showing a position A cerebral blood flow signal measurement procedure.
The
More specifically, as shown in FIG. 1, a cerebral blood flow signal can be obtained from N measurement channels corresponding to a plurality of positions of the scalp by functional near-infrared spectroscopy (fNIRS) (step S101).
Here, as a non-invasive brain activation measuring method, the cerebral blood flow signal can be obtained by using functional near infrared ray spectroscopy (fNIRS) or measuring the gait training state in real time using EEG. For example, as shown in FIG. 2, a cerebral blood flow signal can be acquired by mounting a functional NIR spectroscope during a stroke patient's treadmill walking rehabilitation training. As shown in FIG. 3, for example, 10 light receiving poles and 10 light emitting poles can be used for measuring cerebral blood flow signals. In this case, a total of 31 light absorbing channels can be formed. Thus, cerebral blood flow signals can be obtained from N measurement channels at a plurality of locations on the scalp.
On the other hand, the acquisition of the cerebral blood flow signal is performed in a predetermined temporal measurement order with respect to the stopping and performing. For example, as shown in Fig. 4, the walking operation is performed, 20 seconds, and 20 seconds as a single compartment, for a total of 140 seconds including a total of three time zones and a final dormant 20 second zone.
Next, a bio-noise component such as respiration, blood circulation, movement, etc. of the subject can be removed from the acquired cerebral blood flow signal (step S102). That is, noise can be removed to improve the signal quality of the measured cerebral blood flow signal.
At this time, the cerebral blood flow signal of the measured functional near infrared ray spectroscopy (fNIRS) is measured using a hemodynamic response function (hrf), a wavelet transform, or a finite impulse response filter (FIR) Noise components can be removed.
Thus, only the cerebral blood flow signal including neurophysiological information related to gait can be obtained by removing unwanted noise such as movement of the head, trunk, etc. and physiological noise such as breathing and heartbeat.
Next, the noise-canceled cerebral blood flow signal can be sorted into time series data for execution and rest (step S103). At this time, the cerebral blood flow signals can be sorted into time series data of execution and pause for each channel, for example, in a matrix form. For example, data obtained for 60 seconds and 80 seconds of dormancy for 31 channels can be expressed as a matrix of 31 X 600 (performance) and 31 X 800 (dormancy), respectively.
Next, the sorted data can be mapped to a new classification space having independent variables such as performance and dormancy (step S104). At this time, mapping to the classification space can use various mathematical and statistical techniques, for example, common space patterning (CSP) can be used.
Here, the common space patterning (CSP) may be a spatial filter that increases the dispersion difference for two classes of performance and dormancy. Such a common spatial pattern (CSP) can make the cerebral blood flow signal more distinct or reduce the dimension of the feature.
Next, it is possible to generate a discrimination function for discriminating performance and pause for the channel in the classification space (step S105). Here, since the data of the channels in the performance-idle classification space are concentrated or distributed in a form in which execution and discontinuity are discriminated, it is possible to mathematically define a performance-discontinuity discrimination function that can discriminate the performance-discontinuity function. This distinction function can be generated, for example, using a SVM algorithm as a support vector that maximizes the dispersion difference for execution and dormancy sequences.
Next, the performance-dormancy classification fitness for each channel can be calculated from the generated discrimination function (step S106). Here, the performance-idle classification fitness is a value indicating how clearly the performance and the idle state of each channel can be distinguished from each other, and may be a result of calculating scores for the performance-idle discrimination for each channel from the defined discrimination function . This performance-dormant classification fitness can be computed by an optimal statistical and mathematical technique and can be a mathematical expression for the maximum variance of performance and downtime.
The channels can be sorted according to classification fitness ranking according to the computed performance-hung classification fitness. Here, priority can be given to the lowest-order channel from the highest-order channel of the classification suitability in accordance with the order and stored.
Next, the superior channel and the poor channel can be classified according to the priority decreasing rate (step S107). Here, the reduction rate by priority means an inter-channel differential value, which means a relative decrease value of the classification fitness of the corresponding channel with respect to the classification fitness of the previous upper channel. That is, it is possible to classify up to the previous upper channel as a superior channel and classify the remaining channels into the poor channel based on the channel in which the decrease value between the channels according to the ranking drops sharply.
At this time, if the change of the decrease value is not large or constant, an arbitrary decrease value at a reasonable level can be designated as the second reference value (?). That is, if the rate of decrease by priority is larger than the second reference value?, It is classified as a superior channel, and if it is less than the second reference value?, It can be classified as a poor channel. Here, assuming that there are M number of good channel, the poor channel may be N-M number.
Alternatively, this rate of reduction may be performed by computing the learning error of the classification fitness. For example, if a channel with a significantly lower error rate is identified by repeatedly updating the classification fitness learning error while decreasing the channel from the channel with the lowest classification fitness, the channels with little influence on the learning error are classified into the superior channel A channel with a marked error rate can be classified as an inferior channel.
Next, one-to-one correlation of the inferior channel with respect to each of the superior channels can be calculated (step S108). That is, one-to-one correlation can be calculated for N-M inferior channels for each of the M honorable channels. Here, the channel correlation is calculated using known statistical and mathematical techniques, and can be calculated using, for example, signal coherence.
Next, according to the computed correlation, a part of the inferior channel may be reclassified as the superior channel to extract the final effective channel (step S109). At this time, when the calculated channel correlation degree is set to a maximum of 1, the channels having a first reference value Λ or more are defined as candidate channels that can be escaped from the poor channel, and the correlation with all the good channels among the candidate channels is set as a first reference value (Λ) or higher can be reclassified as a superior channel. That is, it is possible to reclassify a poor channel whose correlation degree is equal to or higher than the first reference value with respect to all good channels among the poor channels as a good channel.
Here, the first reference value? Can be assumed to be, for example, 0.7, but it can be set to a higher value depending on the situation. Assuming that the number of candidate channels escaped from the inferior channel is L, if all the correlation values of the NM inferior channels for each of the M honorable channels are equal to or greater than the first reference value, the corresponding L inferior channel is escaped from the inferior channel Lt; RTI ID = 0.0 > M + L < / RTI > effective channels. At this time, the remaining N- (M + L) channels remain in the inferior channel.
Next, an associated light position for acquiring the cerebral blood flow signal may be determined according to the determined final effective channel (step S110). When the final M + L effective channels are extracted as described above, it can be determined that the positions measured by the channels are directly related to the work (exercise task, cognitive task, etc.) being performed. At this time, it can be compared with the statistical brain activation image, and it can be verified that it is suitable as an effective channel based on the change of the oxidized hemoglobin concentration to the effective channels, and the light emitting pole and the light receiving pole involved in forming the effective channel can be determined.
According to this method, the present invention can provide information on the active position of the cerebral cortex having high reliability, which is clinically important in gait rehabilitation, occupational therapy, etc., so that it is possible to select the measurement position according to the rehabilitation state of the patient, It can be used as a semi-quantitative method of brain plasticity evaluation of gait rehabilitation therapy using signal. It can be applied to the timing and the treatment process of rehabilitation according to lesion size, position, It is possible to provide software that can accurately identify the brain active location regardless of hardware.
Hereinafter, an example of an effective positioning method according to the measurement of brain activity of the functional near infrared ray spectroscopy according to the embodiment of the present invention will be described with reference to FIGS. 5 and 6. FIG.
FIG. 5 is a graph showing the cerebral cortex brain activity image obtained from cerebral blood flow signal measurement during treadmill gait rehabilitation training in a stroke patient, (b) a graph showing changes in the concentration of oxidized hemoglobin in each channel, and (c) (B), a graph of channel classification results, (c) a graph showing the results of each channel, and FIG. 6 And (d) a table showing the peak value of the concentration of oxidized hemoglobin.
Referring to FIG. 5A, as an example of a brain activity image, it is a probability that the brain activity occurs in the cerebral cortex area is relatively displayed with the contrast of the numerical value and the color. This image is implemented by Statistical Parametric Mapping (SPM), which means that the larger the number of bright colors, the higher the probability of brain activation than the other regions in the corresponding cortical area.
For this reason, as shown in FIG. 5B and FIG. 5C, the subject has a high brain activity in the peripheral channels around the 12th channel, and thus the activity of the primary motor cortex is most active, Region and the entire motor cortex, indicating that brain activity is occurring.
Referring to FIG. 6A, the effective channel extracted from the brain activity image of FIG. 5A, which is a result of the same experiment, is shown by circled numbers. At this time, as shown in FIG. 6B, the superior channel and the poor channel can be distinguished according to the reduction rate of the classification fitness between the channels, and the superior channel can be extracted as the effective channel.
Here, the determined effective channel numbers are 2, 7, 12, 21, 10, and 11 channels according to priority. When the subject performs gait training, the brain activity information of the subject is detected more in the effective channels than in the other channels. As shown in FIG. 6C, the concentration of oxidized hemoglobin Changes can be clearly distinguished by high and low.
Also, as shown in FIG. 6D, it is shown that the average value of the peak-to-peak between performance and hibernation of the concentration of oxidized hemoglobin indicating the degree of brain activity is obtained as a large value in the effective channels. T4, T5, and T7, which are involved in the formation of the channel, and the light-receiving poles R1, R3, and R4, which are involved in the formation of the channel, R4, and R7 (see FIG. 3) can be determined as valid measurement positions.
Hereinafter, an effective channel extracting apparatus according to an embodiment of the present invention will be described with reference to FIG. FIG. 7 is a schematic block diagram of an effective channel extracting apparatus according to an embodiment of the present invention for measuring brain activity of functional NIR spectroscopy.
7, the
The
The valid
The
The
More specifically, the
The
The
The
The effective
More specifically, the effective
The effective
Also, the effective
The
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
700: effective position determining device 710: data obtaining unit
720: Valid position extracting unit 722: Noise removing unit
724: Preprocessing section 726: Effective channel determining section
730:
Claims (8)
Removing noise components including respiration, blood circulation, and movement of the subject from the acquired cerebral blood flow signal;
A pre-processing step of dividing the noise-removed signal into time-series data for performance and dormancy to calculate a performance-dormant classification fitness and giving priority to the calculated classification fitness; And
Classifying the M number of good channels and the number of NM poor channels according to the priority decreasing rate and if the correlations of the NM number of poor channels with respect to each of the M number of good channels are equal to or greater than a first reference value, And finally determining M + L effective channels by reclassifying the right channel into the right channel.
The pre-
Sorting the noise canceled signal into time series data for execution and pause;
Mapping the sorted data to a new classification space having independent parameters such as pause and hibernation using common space pattern (CSP);
Generating a discrimination function for discriminating performance and dormancy for the channel on the classification space; And
And computing a performance-pause classification fitness for each channel from the generated distinct function using a SVM (support vector machine) algorithm.
Wherein determining the valid channel comprises:
Classifying the channel into the good channel if the rate of decrease according to the priority is greater than the second reference value and classifying it into the poor channel if the rate is lower than the second reference value;
Calculating a one-to-one correlation of the inferior channel with respect to each of the superior channels; And
And recategorizing, for all of the poor channels, the poor channel whose calculated degree of correlation is equal to or greater than the first reference value, to the superior channel, based on brain activity measurement of the functional near-infrared spectroscopy.
And determining an associated light pole position for acquiring the cerebral blood flow signal according to the determined final effective channel. ≪ RTI ID = 0.0 > 18. < / RTI >
A data obtaining unit for obtaining a cerebral blood flow signal from N measurement channels corresponding to a plurality of positions of the scalp by functional near infrared ray spectroscopy (fNIRS);
A noise eliminator for removing a noise component including respiration, blood circulation, and movement of the subject in the acquired cerebral blood flow signal;
A pre-processing unit for dividing the noise-removed signal into time series data for performance and dormancy to calculate a performance-dormant classification fitness and giving priority to the calculated classification fitness according to the calculated classification fitness; And
Classifying the M number of good channels and the number of NM poor channels according to the priority decreasing rate and if the correlations of the NM number of poor channels with respect to each of the M number of good channels are equal to or greater than a first reference value, And an effective channel determining unit for determining the M + L effective channels by reclassifying the right channel into the good channel.
The pre-
Sorting the noise canceled signal into time series data for execution and rest,
Mapping the sorted data to a new classification space having independent parameters such as the execution and the rest using the common space patterning (CSP)
Generating a distinction function for distinguishing execution and pause for the channel on the classification space,
An effective position determining device according to a brain activity measurement of a functional near infrared ray spectroscopy method for calculating an execution-pause classification fitness for each channel from the generated discrimination function using an SVM algorithm.
The valid channel determination unit may determine,
Classifying the channel into the superior channel if the rate of decrease according to the priority is greater than the second reference value, classifying the channel into the poor channel if the rate is less than the second reference value,
One-to-one correlation of the inferior channel with respect to each of the superior channels, and reclassifying the inferior channel having the calculated degree of correlation to the superior channel to the superior channel for all of the inferior channels, Of the brain.
Wherein the effective channel determination unit determines an associated light pole position for acquiring the cerebral blood flow signal according to the determined final effective channel.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR20150050186 | 2015-04-09 | ||
| KR1020150050186 | 2015-04-09 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20160121348A KR20160121348A (en) | 2016-10-19 |
| KR101703547B1 true KR101703547B1 (en) | 2017-02-07 |
Family
ID=57250932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020150079615A KR101703547B1 (en) | 2015-04-09 | 2015-06-05 | Method and apparatus for estimating effective channel of functional near-infrared spectroscopy |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR101703547B1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190031124A (en) | 2017-09-15 | 2019-03-25 | 재단법인대구경북과학기술원 | Method and Device for removing noise from multi-channel fNIRS |
| KR20190140705A (en) * | 2018-06-12 | 2019-12-20 | 재단법인대구경북과학기술원 | Method and Device for removing noise from Functional near-infrared spectroscopy signal |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020209688A1 (en) | 2019-04-12 | 2020-10-15 | 한국과학기술원 | Method, system, and non-transitory computer-readable recording medium for estimating biometric information about head using machine learning |
| CN110680282B (en) * | 2019-10-09 | 2020-10-27 | 黑龙江洛唯智能科技有限公司 | Method, device and system for detecting temporary abnormal state of brain |
| KR102288267B1 (en) * | 2020-07-22 | 2021-08-11 | 액티브레인바이오(주) | AI(Artificial Intelligence) BASED METHOD OF PROVIDING BRAIN INFORMATION |
| KR102241759B1 (en) * | 2020-07-22 | 2021-04-20 | (주)엔브레인 | AI(Artificial Intelligence) BASED BRAIN INFORMATION PROVIDING DEVICES |
| KR102222335B1 (en) * | 2020-12-01 | 2021-03-04 | (주)엔브레인 | AI(Artificial Intelligence) BASED DEVICE THAT PROVIDES BRAIN INFORMATION BASED ON INFORMATION COLLECTED WHILE THE USER IS IN MOTION |
| KR102211049B1 (en) * | 2020-12-01 | 2021-02-04 | (주)엔브레인 | (AI(Artificial Intelligence) BASED BRAIN INFORMATION PROVIDING DEVICE USING BIOMETRIC INFORMATION |
| KR102211048B1 (en) * | 2020-12-01 | 2021-02-04 | (주)엔브레인 | AI(Artificial Intelligence) BASED BRAIN INFORMATION PROVIDING DEVICES USING AUDITORY INFORMATION |
| KR102211050B1 (en) * | 2020-12-01 | 2021-02-04 | (주)엔브레인 | AI(Artificial Intelligence) BASED BRAIN INFORMATION PROVIDING DEVICES USING MEDICAL IMAGING INFORMATION |
| KR102211030B1 (en) * | 2020-12-01 | 2021-02-04 | (주)엔브레인 | AI(Artificial Intelligence) BASED BRAIN INFORMATION PROVIDING DEVICES USING BRAIN STIMULATION |
| CN113633262B (en) * | 2021-08-13 | 2023-08-18 | 嘉兴知芯电子科技有限公司 | Near infrared spectrum brain function imaging system and imaging signal sampling method |
| KR20240002544A (en) | 2022-06-29 | 2024-01-05 | 원광대학교산학협력단 | Method for determining suicide risk group using fNIRS signal |
| KR102571116B1 (en) * | 2023-02-21 | 2023-08-24 | 고려대학교 세종산학협력단 | Method and device of brain-computer interface based on optimal channel selection algorithm using nirs |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140107494A1 (en) | 2011-05-31 | 2014-04-17 | Nagoya Institute Of Technology | Cognitive Impairment Determination Apparatus, Cognitive Impairment Determination System and Program |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101430823B1 (en) | 2012-10-23 | 2014-08-14 | 주식회사 싸이버메딕 | Walking analysis treadmill of unweight control |
-
2015
- 2015-06-05 KR KR1020150079615A patent/KR101703547B1/en active IP Right Grant
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140107494A1 (en) | 2011-05-31 | 2014-04-17 | Nagoya Institute Of Technology | Cognitive Impairment Determination Apparatus, Cognitive Impairment Determination System and Program |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190031124A (en) | 2017-09-15 | 2019-03-25 | 재단법인대구경북과학기술원 | Method and Device for removing noise from multi-channel fNIRS |
| KR20190140705A (en) * | 2018-06-12 | 2019-12-20 | 재단법인대구경북과학기술원 | Method and Device for removing noise from Functional near-infrared spectroscopy signal |
| KR102123024B1 (en) | 2018-06-12 | 2020-06-16 | 재단법인 대구경북과학기술원 | Method and Device for removing noise from Functional near-infrared spectroscopy signal |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20160121348A (en) | 2016-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101703547B1 (en) | Method and apparatus for estimating effective channel of functional near-infrared spectroscopy | |
| Melnik et al. | Systems, subjects, sessions: To what extent do these factors influence EEG data? | |
| Pinti et al. | A novel GLM-based method for the Automatic IDentification of functional Events (AIDE) in fNIRS data recorded in naturalistic environments | |
| Gwin et al. | Removal of movement artifact from high-density EEG recorded during walking and running | |
| US4736751A (en) | Brain wave source network location scanning method and system | |
| Jungnickel et al. | Mobile brain/body imaging (MoBI) of physical interaction with dynamically moving objects | |
| Wu et al. | Resting-state cortical connectivity predicts motor skill acquisition | |
| JP7221693B2 (en) | Method and magnetic imaging device for cataloging cortical function in the human brain | |
| Gow Jr et al. | Lexical influences on speech perception: a Granger causality analysis of MEG and EEG source estimates | |
| KR102143900B1 (en) | System for providing subject-independent brain-computer interface and method thereof | |
| US20160029965A1 (en) | Artifact as a feature in neuro diagnostics | |
| Islam et al. | EEG mobility artifact removal for ambulatory epileptic seizure prediction applications | |
| US20070179396A1 (en) | Method and System for Detecting and Classifying Facial Muscle Movements | |
| CA2657407A1 (en) | Analysis of brain patterns using temporal measures | |
| WO2016075324A1 (en) | Scoring method based on improved signals analysis | |
| Naghsh et al. | Spatial analysis of EEG signals for Parkinson’s disease stage detection | |
| Chen et al. | Scalp EEG-based pain detection using convolutional neural network | |
| Leutheuser et al. | Comparison of the AMICA and the InfoMax algorithm for the reduction of electromyogenic artifacts in EEG data | |
| Kohan et al. | Interview based connectivity analysis of EEG in order to detect deception | |
| Khalid et al. | Epileptic MEG spikes detection using amplitude thresholding and dynamic time warping | |
| Jiang et al. | Characterization of bimanual cyclical tasks from single-trial EEG-fNIRS measurements | |
| Wang et al. | On the channel density of EEG signals for reliable biometric recognition | |
| Kamal et al. | Decoding the correlation between heart activation and walking path by information-based analysis | |
| CN113614751A (en) | Electroencephalogram signal identification and extraction | |
| JP6834318B2 (en) | Stress evaluation device and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E701 | Decision to grant or registration of patent right | ||
| GRNT | Written decision to grant |