WO2012169010A1

WO2012169010A1 - Biometric feature measurement device and biometric feature measurement program

Info

Publication number: WO2012169010A1
Application number: PCT/JP2011/063039
Authority: WO
Inventors: 木戸　邦彦; 山本　剛; 邦昭小澤
Original assignee: 株式会社日立製作所
Priority date: 2011-06-07
Filing date: 2011-06-07
Publication date: 2012-12-13

Abstract

The invention makes it possible to conduct simple cerebral function tests even on testees for whom computer operation is difficult and slow such as physically challenged people and elderly people. The invention comprises: an input unit that inputs characters by selecting characters with a cursor that sequentially scans characters or character sets on a character board in a biometric feature measurement device that uses language; a character input smoothness-evaluating unit that evaluates the smoothness of character input based on the number of scans needed for the character input; and a cerebral function-evaluating unit that evaluates the level of cerebral function from the smoothness of character input.

Description

Biological feature measuring apparatus and biological feature measuring program

The present invention relates to a biological feature measuring apparatus and a biological feature measuring program.

There is Patent Document 1 as background art in this technical field. This document describes a system that tests Kanji identification tests and broad tests on computers for the purpose of early detection of early symptoms of senile dementia, and tests the degree of dementia from answers within the response time limit. Yes.

Patent Document 2 discloses an apparatus for supporting character communication of a disabled person or an elderly person by sequentially lighting characters on the dial for each block and selecting a block with a touch sensor or the like to input characters. There is a description about the device.

JP 2005-46289 JP JP 2004-286768 A

As a simple brain frontal lobe function test, a test using a word generation task is often performed. In this exercise, you will select a single hiragana character such as “A” and have it answer as many words as possible with “A” per minute. However, some people with severe physical disabilities, such as amyotrophic lateral sclerosis (ALS), cannot wear a ventilator and cannot speak, and their motor functions are limited to facial muscles, eyeballs, and limbs. It only moves slightly, making the application of the test extremely difficult. For the same reason, it is difficult to carry out the above-mentioned kanji identification test and the broad test. On the other hand, it is common for such patients to communicate with each other by means of the above-described dial and switch communication device. However, character input by the device takes several tens of seconds per character. For this reason, there is a limit in performing the type of tests for enumerating answers in a short time, such as a test using a word generation task, a kanji identification test, and a kanahiroi test.
In addition, elderly people who are not used to computer operations have similar problems.

The present invention provides a biological feature measuring apparatus and a biological feature measuring program capable of performing simple brain function measurement even for a slow subject who is difficult to operate a computer, such as a physically handicapped person or an elderly person. Objective.

The present invention examines brain function as a biometric feature by evaluating the smoothness of the ability to recall words based on the smoothness of character input using the dial.

If an example of the biometric feature measuring apparatus of the present invention is given, in a biometric feature measuring apparatus using a language, character input is performed by selecting a character from a cursor that sequentially scans a character or a set of characters on a dial. An input unit to perform, a character input smoothness evaluation unit that evaluates the smoothness of character input based on the number of scans required for the character input, and a brain function evaluation unit that evaluates a brain function level from the smoothness of the character input; It is characterized by having.

In another example of the biometric feature measuring apparatus of the present invention, in a biometric feature measuring apparatus using a language, a character is selected from a cursor that sequentially scans a character or a set of characters on a dial. Repeated the same task as a brain function test using a language, an input unit for inputting characters with a character, a character input smoothness evaluating unit for evaluating the smoothness of character input based on the number of scans required for the character input, and An activation degree calculation unit that calculates the activation degree of the brain from the brain function measurement data, and a brain function level determination unit that determines the brain function level based on the smoothness of the character input and the activation degree of the brain It is characterized by having.

An example of the biometric feature measurement program of the present invention is a biometric feature measurement program that causes a computer to execute the following procedure so as to measure a biometric feature using a language. Displaying a dial on the screen, scanning a character or a set of characters on the dial sequentially with a cursor, and moving the cursor on the dial via the input means of the computer Receiving a character input; storing the number of scans required for the character input in a storage unit; evaluating a smoothness of character input based on the stored number of scans; and the character input And a step of evaluating a brain function level from the smoothness of the brain.

Another example of the biometric feature measurement program of the present invention is a biometric feature measurement program that causes a computer to execute the following procedure so as to measure a biometric feature using a language. A step of displaying a dial on a display screen of the computer, a step of sequentially scanning a character or a set of characters on the dial with a cursor, and an input of the computer by moving the cursor on the dial Receiving character input via the means, storing the number of scans required for the character input in a storage means, and evaluating the smoothness of character input based on the stored number of scans; From the brain function measurement data when the same task as the brain function test using language is repeated, Calculating a degree, on the basis of smoothness of the character input and the activation of the brain and is characterized by comprising a step of determining the brain function level.

According to the present invention, a simple brain function test can be performed even for a slow subject who is difficult to operate a computer, such as a physically handicapped person or an elderly person.

It is a block block diagram of the frontal lobe function test | inspection apparatus of Example 1 of this invention. It is a figure which shows the hardware constitutions of the frontal lobe function test | inspection apparatus of Example 1 of this invention. It is an example of the screen of a frontal lobe function test | inspection apparatus. It is an example of the dial of a frontal lobe function inspection apparatus. It is a flowchart explaining the process of a dial face input / output control part and a dial face / scan method conversion part. It is an example of the format regarding an input character string & scan history. It is a flowchart explaining the process regarding the difference calculation part and character input smooth vector generation part of a scanning history and the shortest scanning frequency | count. It is an example of a character input smooth vector. It is a flowchart explaining the process regarding the average number of scans per word. It is an example of the tritree used in the input character prediction part. It is an example of replacement of dial rows. It is a flowchart explaining the process of an input character estimation part and a dial face / scan method conversion part. It is a block block diagram of the frontal lobe function level determination part of Example 2 of this invention. It is an example of the task in brain function measurement. It is an example of the measurement position in brain function measurement. It is a flowchart explaining the process of Example 2 of this invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

In this embodiment, an example of a device that performs a brain function test will be described. Here, an example of the frontal lobe test is shown as an example of the biometric feature test and the brain function test. However, the present invention is not limited to this, and it is needless to say that the entire brain function can be tested. Hereinafter, it will be referred to as the frontal lobe inspection device 100 or the brain function inspection device 100.

FIG. 1 is a block diagram of a frontal lobe function inspection apparatus 100 which is an example of a biological feature measurement apparatus according to the first embodiment. The frontal lobe function testing device 100 includes a dial control unit 101, a character input smoothness evaluation unit 102, and a frontal lobe evaluation unit 103. The dial control unit 101 includes a switch input detection unit 104, a dial input / output control unit 105, a dial / scan method conversion unit 106, and an input character prediction unit 107. The character input smoothness evaluation unit 102 includes an input character string & scan history management unit 108, a difference calculation unit 109 between a scan history and the minimum number of scans, and a character input smoothness vector generation unit 110. The frontal lobe evaluation unit 103 includes a character input smoothness vector management unit 111, an average scan count calculation unit 112 per word, and a frontal lobe function level output unit 113.

Next, the hardware configuration of the frontal lobe function testing apparatus 100 will be described with reference to FIG. The frontal lobe function testing device 100 includes a CPU 201, a memory 202, an external storage device 203 such as a hard disk, a communication device 204 for communicating with other devices via a network 210, and input devices such as a switch, a keyboard, and a mouse. 205, an output device 206 such as a monitor or a printer, a reading device 207 that reads data from a storage medium 209 such as a CD-ROM or FD, and an interface 208 that transmits and receives data between these components. It is a general computer.

The frontal lobe function testing apparatus 100 can be realized by executing a predetermined program loaded on the memory 202 by the CPU 201. The predetermined program is input from the storage medium 209 in which the program is stored via the reading device 207 or input from the network 210 via the communication device 204 and directly loaded on the memory 202 or once. Then, the data may be stored in the external storage device 203 and then loaded onto the memory 202.

The invention of the program according to the present invention is a program that is incorporated in a computer and operates as a biological feature measuring device. By incorporating the program of the present invention into a computer, the biological feature measuring apparatus shown in the block diagram of FIG. 1 is configured.

Next, the screen configuration of the frontal lobe function testing apparatus 100 will be described with reference to FIG. The screen 300 includes an input character display unit 301 that displays input characters, a dial 302 in which characters are arranged in a grid, and a cursor 303. In the cursor 303 of FIG. 3, the cursor is focused on the block “A row to a row”. Here, an example of a two-block scan with a cursor will be described with reference to FIG.

First, scanning for column selection will be described.
In the first scan of column selection, the “A row to row” block 403 shown on the dial 401 and the “Ha row to W row” block 404 shown on the dial 402 are alternately focused. If the block 403 “A row to row” is selected, in the second scan of the column selection, the block 407 “A row to row” shown on the dial 405 and the block 406 are shown. Blocks 406 of “Sa line to Na line” are alternately focused.
If the block 408 “Sa line to Na line” is selected, the “Sa line to Ta line” block 411 shown on the dial 409 and the face 410 are shown as the third scan of column selection. The “na row” blocks 412 are alternately focused.
When the column “ka row” is selected, the column is selected, and the process moves to scanning for character selection. As a first-stage scan of character selection, a block 415 of “ka, ki, koku” indicated by the dial 413 and a block 416 of “ke, ko” indicated by the dial 414 are alternately focused. Here, when “Keko” is selected in block 416, the character “ke” and the character “ko” are alternately scanned as the second scan of the character selection, and finally one of the characters is scanned. Will be selected.

The processing of the dial input / output control unit 105 and the dial / scanning method conversion unit 106 of the dial control unit 101 will be described with reference to FIG.

First, at step 501, the cursor is set to the initial position. In this embodiment, the initial position is always the state in which the left block is in focus. For example, in the first scan, the initial position is the state where the block 403 shown in the dial 401 in FIG. 4 is focused. In step 502, a cursor scan is started. Here, the scan is performed according to a predetermined cycle. For example, if the period is 1 second, the

blocks

403 and 404 are alternately focused every second in the first scan of the column.

During scanning, in step 503, information on the cursor position in the middle is recorded in the input character string & scan history management unit 108 in FIG. 1 in the format of FIG. FIG. 6 is an example of a scan history when inputting characters “ka” and “ri” for the character “Akari” in response to a character generation task related to the first character “A”. As an example, the case of inputting the character “ka” will be described. The section 601 in FIG. 6 describes [column: dial 1: 1-1-1]. This is the first step in the dial arrangement designated by the name “clock 1” for the scan related to “column”. This means that the block 403 in FIG. 4 has been scanned as the eye scan. The first 1 of “1-1” represents the first-stage scan, and the last 1 represents the block 403. If it is “1-2”, it means that the block 404 in FIG. 4 has been scanned as the first scan. Similarly, section 602 [line: dial 1, 1-or -1] in FIG. 6 is the first stage in the dial arrangement designated by the name “dial 1” for the scan related to “line”. This means that the block 415 in FIG. 4 has been scanned as an eye scan. The first 1 of "1- or -1" represents the first scan, and the next "ka-1" is the first block of "ka line". That is, it means the block 415 in FIG. If “ka-2”, the second block of “ka row”. That is, it means the block 416 in FIG. In the present embodiment, “Dial 1” represents a dial having the same arrangement as the standard 50-note table as shown in FIG.

In step 504, the switch input is monitored from the input device 205 in FIG. If the switch is ON, the cursor scan is stopped at step 505. If the column selection is completed at step 506, the process proceeds to step 507. If the column selection is in progress, the block is changed at step 516. As shown in FIG. 4, when the first stage scan performed in block 403 and block 404 is completed, the block change is performed in the second stage, such as reducing the block to block 407 and block 408. This is a process for changing the size to be used. After the block change is performed in step 516, the process returns to step 501.

In step 507, the initial position of the cursor is set for line scanning. In this embodiment, the initial position is always the state where the upper block is focused. For example, in the first-stage scan relating to “ka line”, a state where the block 415 shown in the dial 413 in FIG. In step 508, a cursor scan is started. Here, the scan is performed according to a predetermined cycle. For example, if the period is 1 second, the

blocks

415 and 416 are alternately focused every second in the first scan of the column. During scanning, in step 509, the information on the cursor position in the middle is recorded in the input character string & scan history management unit 108 in FIG. 1 in the format of FIG. In step 510, switch input is monitored from the input device 205 of FIG. If the switch is ON, the cursor scan is stopped in step 511. If the character selection is completed in step 512, the process proceeds to step 513. If the character selection is in progress, the block is changed in step 517. After changing the block in step 517, the process returns to step 507.

In step 513, the selected character is recorded. Here, if a double click is detected in step 514, a line feed is made. If no double click is detected, the process returns to step 501.

With reference to FIG. 7, the processing of the difference calculation unit 109 and the character input smoothness vector generation unit 110 of the scan history and the shortest scan count of the character input smoothness evaluation unit 102 will be described.
In step 701, the scan history is acquired from the input character string & scan history management unit 108. In step 702, one character is acquired from the input character. Here, the character “Akari” will be described as an example for the character generation task in FIG. 6 where the first character is “A”. In this case, the first character is “ka”. In step 703, dial face information is acquired. In the case of FIG. 6, “Dial 1” is set as described in the section 601. As described above, “Dial 1” is a dial having the same arrangement as the standard 50-note table as shown in FIG. In step 705, the minimum number of scans in each stage is calculated. For example, in the case of “KA” of “AKARI”, the first-stage scan of column selection starts from the left side as shown in block 403 in FIG. Therefore, since “ka row” is included in the block 403, the minimum number of scans is one. Similarly, in the second-stage scan of column selection, “ka row” is included in the block 407, so the minimum number of scans is 1. On the other hand, in step 707, the actual number of scans is obtained from the scan history. In FIG. 6, the first scan of column selection is one time. In step 708, the difference between the minimum number of scans obtained in step 705 and the actual number of scans obtained in step 706 is calculated. In step 706, if there is no input character, the average number of scans per stage is calculated from the difference of each stage calculated in step 708. For example, in the case of FIG. 6, when “ka” is input, “ka” is input with the shortest number of scans, and thus the difference in each stage is zero. That is, the difference is 0 in all three stages of column selection and two stages of character selection. Therefore, the average number of scans per stage is zero.

Next, in the case of “RI” input, the minimum number of scans in the first row of the column selection is 2 times, the second row is 2 times, and the third row is 1 time. The minimum number of scans in the first stage of character selection is 1 and the second stage is 2 times. In the scan history of FIG. 6, the second stage of character selection is missed, and scanning is performed five times. Therefore, the difference in the second stage of character selection is 3.

On the other hand, the number of stages is 5 in total, 3 for column selection and 2 for character selection. Therefore, the average number of scans per stage is 3/5 = 0.6.

Finally, in step 710, a character input smoothness vector is generated. In the above example, the average number of scans per stage number of “ka” is 0, “ri” is 0.6, and the character input smoothness vector is (0, 0.6) as shown in FIG. This character input smoothness vector is recorded in the character input smoothness vector management unit 111 of FIG.

Next, the processing of the average scan count calculation unit 112 per word in the frontal lobe evaluation unit 103 will be described with reference to FIG. In step 901, a character input smoothness vector is acquired from the character input smoothness vector management unit 111 for each word successfully input in the word generation task. In step 902, the number of characters constituting each word is calculated. For example, in the case of “AKARI”, the number of constituent characters is two except for the leading “A”. Next, in step 903, the average number of scans for each word is calculated. For example, in the case of “light”, in the case of the character input smoothness vector of FIG. 8, 0.6 / 2 = 0.3. Finally, in step 904, the average number of scans per word is calculated. That is, for all the words that have been successfully input in the word generation task, the sum of the average number of scans obtained in step 903 is taken and divided by the number of words of the words that have been successfully input. Finally, the frontal lobe function level output unit 113 in FIG. 1 outputs the value calculated in step 904.

Here, in the block conversion of step 516 in FIG. 5, the arrangement of the dials may be changed according to the input characters. This process will be described with reference to FIG. In step 1201 related to the input character prediction unit 107 in FIG. 1, input characters are predicted from the tri-tree. In this embodiment, description will be made based on FIG. For example, when “Aka” is input, the

trees

1001, 1002, 1003, and 1004 are traced, and “RI”, “SI”, “SU”, and “NE” are predicted as the next input characters. If “Akasu” is input, “ri” is predicted as the next input character for the tree 1003 by the tree 1005 at the lower part.

Based on this input character prediction, in step 1202, according to the number of characters of the predicted character, priority is given to “A” to “W”, and the columns are rearranged. For example, when “Aka” is entered, “Sa line” including the input predicted characters “shi” and “su” has the highest priority, followed by “Ra line” including “ri”, “Na row” including “Ne” has the same priority.

In step 1203, the current cursor position is obtained, and in step 1204, the top three columns having a large number including the next input character are replaced. For example, immediately after “ka” is input, “sa line”, “ta line”, and “na line” on the right side of “ka line” are changed to “sa line” and “ra line” as indicated by 1101 in FIG. Replace it with the column “Na row”. When the dial arrangement is changed in this way, the name of the dial is changed. If the name is dial 2, when recording the scan history, in the case of section 601 in FIG. 6, [column: dial 1: 1- 1] becomes [column: dial 2: 1-1].

As described above, when a column that contains a character that is likely to be input next is replaced with a column that is close to the current cursor position, if the subject does not clearly recall the next input character, the cursor should be stopped as soon as possible. The possibility of not being able to be increased. That is, the operation of switching the columns contributes to the evaluation of the smoothness of the word recall ability.

In this embodiment, an example of an apparatus that also utilizes data obtained by brain function measurement will be described. In brain function measurement, it is conceivable to use a brain function measuring device such as optical topography or fMRI (functional MRI), but the use of these devices is not limited. In the present embodiment, optical topography is assumed for brain function measurement, and m-channel OxyHb signal S _k (t) (k = 1 to m) is considered as measurement data.

In Example 2, the block shown in FIG. 13 is added to the frontal lobe evaluation unit 103 of the apparatus 100. The blocks to be added are a brain function evaluation vector generation unit 1301, a composite vector generation unit 1302, a frontal function determination processing unit 1303, and a determination data management unit 1305.

In this embodiment, data is measured by the experimental method shown in FIG. 14 as brain function measurement. The rest period and the period for performing the task of activating the frontal lobe shall be repeated n times. In the present embodiment, it is assumed that the word generation task for 30 seconds is repeated after the rest for 30 seconds. FIG. 15 shows a measurement position. In this embodiment, the entire forehead is measured by 12 light sources and 12 light receiving units. A black circle 1501 is an example of a light source, and a white circle 1502 is a light receiving unit. It is an example. In the case of FIG. 15, there are 22 channels, and m = 22.

Next, processing of the brain function evaluation vector generation unit 1301, the combined vector generation unit 1302, and the frontal function determination processing unit 1303 will be described based on the flow of FIG.

In step 1601, block averaging is performed for each channel k of S _k (t) (k = 1 to 22) based on Equation (1).

In step 1601, after removing the trend component for each block, block averaging processing based on Equation 1 may be performed.

In step 1602, for each channel k, the activation degree U (k) of the brain is calculated from the absolute value of the difference between the average value of the rest obtained by block averaging and the average value of the task based on Equation 2.

In step 1603, a composite vector is generated from the brain activation level obtained for each channel k and the frontal lobe function level based on the smoothness of character input as follows.

In step 1604, the normality / abnormality is determined from the combined vector according to Equation 3.

That is, assuming that X ₁ = U (1), X ₂ = U (2),..., X _{m + 1} = L, w is calculated according to equation (4). To do. It is assumed that a _i is determined in advance based on linear discriminant analysis based on the three vectors collected by a plurality of subjects.

According to the present embodiment, the accuracy of the frontal lobe function test can be improved by combining with the brain function measurement data.

Even for subjects with physical disabilities and the elderly who have difficulty in computer operation and can be slow, a simple brain frontal lobe function test can be performed, which can be used for early detection of dementia and senile dementia. .

DESCRIPTION OF SYMBOLS 100 Frontal lobe function test apparatus 101 Dial control part 102 Character input smoothness evaluation part 103 Frontal lobe evaluation part 104 Switch input detection part 105 Dial input / output control part 106 Dial / scan method conversion part 107 Input character prediction part 108 Input character string & Scan history management unit 109 Difference calculation unit between scan history and minimum number of scans 110 Character input smoothness vector generation unit 111 Character input smoothness vector management unit 112 Average scan number per word calculation unit 113 Frontal lobe function level output unit 201 CPU
202 Memory 203 External Recording Device 204 Communication Device 205 Input Device 206 Output Device 207 Reading Device 208 Interface 209 Storage Medium 210 Network 300 Screen 301 Input Character Display Unit 302 Dial 303 Cursor 1301 Brain Function Evaluation Vector Generation Unit 1302 Synthetic Vector Generation Unit 1303 Frontal lobe function determination processing unit 1304 Frontal lobe function level output unit 1305 Determination data management unit 1501 Light source 1502 Light receiving unit

Claims

In a biological feature measurement device using language,
An input unit for inputting characters by selecting a character from a cursor that sequentially scans a character or a set of characters on the dial, and
A character input smoothness evaluation unit that evaluates the smoothness of character input based on the number of scans required for the character input;
And a brain function evaluation unit that evaluates a brain function level from the smoothness of the character input.
In the biological feature measuring device according to claim 1,
The character input smoothness evaluation unit, for each input character, the shortest scan number calculation unit for calculating the shortest number of scans necessary to input the character, the number of scans actually required for character input of the character, A biological feature measurement apparatus comprising a difference calculation unit for obtaining a difference from the shortest number of scans.
The biological feature measuring apparatus according to claim 2,
The biometric feature measurement apparatus, wherein the character input smoothness evaluation unit includes a scan history management unit that records information on a cursor position during a scan, and obtains the number of scans based on the recorded scan history.
In the biological feature measuring device according to claim 1,
The biometric feature measuring apparatus, wherein the input unit divides the dial into two parts, left and right or up and down, and the cursor alternately scans the two divided areas.
The biological feature measuring apparatus according to claim 1, further comprising:
An input character prediction unit that predicts a character to be input next from input characters, and a column or row of a dial plate that includes a large number of predicted characters, and the selected column or row is selected from the dial A biometric feature measuring apparatus comprising a dial conversion unit that replaces the remaining columns or rows.
In a biological feature measurement device using language,
An input unit for inputting characters by selecting a character from a cursor that sequentially scans a character or a set of characters on the dial, and
A character input smoothness evaluation unit that evaluates the smoothness of character input based on the number of scans required for the character input;
From the brain function measurement data when the same task as the brain function test using language is repeated, an activation degree calculation unit that calculates the activation degree of the brain,
A biological feature measuring apparatus comprising: a brain function level determination unit that determines a brain function level based on the smoothness of the character input and the activation degree of the brain.
A biometric feature measurement program that executes the following procedure in a computing unit of a computer so as to measure biometric features using a language,
Displaying a dial on a display screen of the computer;
Scanning a character or a set of characters on the dial sequentially with a cursor;
Receiving a character input via the input means of the computer by moving the cursor on the dial;
Storing the number of scans required for the character input in a storage means;
Evaluating the smoothness of character input based on the stored number of scans;
And a step of evaluating a brain function level from the smoothness of the character input.
The biological feature measurement program according to claim 7,
The step of evaluating the smoothness of character input includes, for each input character, calculating the shortest number of scans necessary to input the character, the number of scans actually required for character input of the character, A frontal lobe function inspection method comprising a step of obtaining a difference from the shortest scan count.
The biological feature measurement program according to claim 8,
The step of evaluating the smoothness of character input records information on a cursor position during a scan, and obtains the number of scans based on the recorded scan history.
In the biological feature measurement program according to claim 7,
The step of inputting characters is a biological feature measurement program characterized in that the dial is divided into left and right or up and down, and the cursor alternately scans the two divided regions.
The biological feature measurement program according to claim 7, further comprising:
Predicting the next input character from the input characters;
A biological feature measurement program comprising a step of selecting a column or row of a dial plate containing a large number of predicted characters and replacing the selected column or row with a remaining column or row of the dial.
A biometric feature measurement program that executes the following procedure in a computing unit of a computer so as to measure biometric features using a language,
Displaying a dial on a display screen of the computer;
Scanning a character or a set of characters on the dial sequentially with a cursor;
Receiving a character input via the input means of the computer by moving the cursor on the dial;
Storing the number of scans required for the character input in a storage means;
Evaluating the smoothness of character input based on the stored number of scans;
Calculating brain activation from brain function measurement data when the same task as the brain function test using language is repeated,
And a step of determining a brain function level based on the smoothness of the character input and the activation degree of the brain.