KR20240031137A - Apparatus and method for classificating types of primary progressive aphasia based on task performance results - Google Patents

Abstract

This disclosure relates to a device and a classification method that can classify types of primary progressive aphasia based on tasks performed by the user. The primary progressive aphasia type classification device based on task performance results according to the present disclosure includes a memory configured to store task-related content, a microphone configured to collect the user's voice information, a display unit configured to display screen information, and a device configured to guide the user's speech. A primary progressive aphasia type classification device including a processor that outputs the task-related content on the display unit and classifies the user's primary progressive aphasia type based on the user's speech information collected through the microphone. can be provided.

Description

Primary progressive aphasia type classification device and classification method based on task performance results {APPARATUS AND METHOD FOR CLASSIFICATING TYPES OF PRIMARY PROGRESSIVE APHASIA BASED ON TASK PERFORMANCE RESULTS}

This disclosure relates to a device and a classification method that can classify types of primary progressive aphasia based on tasks performed by the user.

Primary progressive aphasia (PPA) is a neurological degenerative disease in which language and speech abilities are gradually impaired, and non-verbal cognitive and behavioral dysfunction is relatively preserved for at least 2 years after onset. It is a disability. PPA is divided into at least three clinical types depending on the causative pathophysiological disease: logopenic variant of PPA (lvPPA), semantic variant of PPA (svPPA), and nonfluent type (logopenic variant of PPA, lvPPA). Includes nonfluent/agrammatic variant of PPA, nfvPPA).

Looking at the characteristics of each type, first, lvPPA is generally fluent, but due to lexical retrieval defects, it is difficult to initiate speech or the flow of speech is interrupted due to hesitation, including filler, but this is due to speech motor disorders and must be distinguished. To compensate for these lexical retrieval defects, pauses within and between utterances are often used. In addition, errors such as word omission are a prominent feature in certain language tasks, including 'repeat'.

The most important characteristic of svPPA is semantic deficits, with lexical semantic impairment being particularly prominent. Due to this damage to meaning, they cannot understand simple words or have difficulty using words in speech, so they use demonstrative pronouns such as 'this, that' or alternative words, and have a fluent form characterized by circumlocution instead of target vocabulary. Shows utterance. However, like lvPPA, speech motor impairment is not observed.

nfvPPA is a non-fluent type of PPA. Due to a defect in the speech motor plan or program, the overall speaking speed is slow and the prosody is monotonous. Additionally, the length of the utterance is short and the utterance contains agrammatism. In terms of motor speech, it may be accompanied by apraxia of speech (AOS), and features such as self-correction and stuttering are particularly evident in polysyllabic utterances or utterances requiring complex articulation.

In this way, although each type of PPA has characteristic speech and can be distinguished from each other, when the disease is in its early stages, the speech characteristics of each type may not be prominent, so it may not be easy to distinguish between the types. Since classification is an essential process to detect the cause of PPA in advance and provide appropriate drug treatment, language evaluation can be helpful in the diagnosis. For this reason, although previous studies conducted in English-speaking countries have tried to classify by type, speech research has focused on English, and there has been no research that reflects the linguistic characteristics of Korean. For example, Korean speech has a prominent language-specific characteristic of clear inter-syllable spacing, so it is difficult to apply it to PPA subtype classification by directly reflecting the results of previous studies in English-speaking countries.

Republic of Korea Patent Registration No. 10-2444012

The purpose of the present disclosure is to provide a device and method for allowing a user to perform a plurality of tasks and classifying PPA subtypes based on the task performance results.

The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood by those skilled in the art from the description below.

The primary progressive aphasia type classification device based on the task performance results to achieve the above-mentioned technical task includes a memory configured to store task-related content, a microphone configured to collect the user's voice information, a display unit configured to display screen information, and a user's input device. Primary progressive aphasia comprising a processor that outputs the task-related content on the display unit to induce speech and classifies the user's primary progressive aphasia type based on the user's speech information collected through the microphone. A type classification device may be provided.

In one embodiment, the content may include at least one of text consisting of content, images, sentences, or paragraphs that induce conversation between the user and the evaluator.

In one embodiment, the processor performs a task performance result corresponding to the content that induces the conversation consisting of the sentence, a task performance result corresponding to the image, a task performance result corresponding to the text consisting of the paragraph, and the sentence. It is possible to classify the type of primary progressive aphasia by assigning high weight to the task performance results corresponding to the text.

In one embodiment, the processor extracts the total number of utterances from the utterance information, the total number of syllables, the number of syllables per utterance, the number of words per utterance, the overall pause frequency, the pause frequency between utterances, and the pause frequency within utterances (between words). + intra-word pause), inter-word pause frequency, intra-word pause frequency, total pause duration, inter-utterance pause time, intra-utterance pause time (inter-word + intra-word pause), inter-word pause time, intra-word pause time, speech speed and articulation Classification of the type of primary progressive aphasia can be performed based on at least one of the speeds.

In one embodiment, the processor is a nonfluent/agrammatic variant of PPA (nfvPPA), a semantic variant of PPA (svPPA), a logopenic variant of PPA (lvPPA), and a normal PPA. The type of primary progressive aphasia of the user can be classified as one of the following.

In one embodiment, the processor may determine whether the user's primary progressive aphasia type is normal, based on a task performance result corresponding to the image.

In one embodiment, the processor may determine whether the user's primary progressive aphasia type is nfvPPA, based on a task performance result corresponding to the text consisting of the paragraph.

In one embodiment, the processor sets a silent section of 100 ms (0.1 second) or more within an utterance and between utterances as a pause in the utterance information, and sets at least the total pause frequency and the pause time within the utterance extracted from the utterance information. Based on this, the user's primary progressive aphasia type can be classified as either svPPA or lvPPA.

In addition, the primary progressive aphasia type classification method based on task performance results to achieve the above-mentioned technical task includes the steps of outputting task-related content on a display unit to induce user speech, and collecting user speech information through a microphone. It may include a primary progressive aphasia type classification method including the step of classifying the user's primary progressive aphasia type based on the step and speech information of the user.

In addition, the present disclosure can be combined with a computer to provide a program stored in a computer-readable recording medium to execute the method for classifying the primary progressive aphasia type.

According to the present disclosure, differences in performance of speech rate (speech rate, articulation rate) and pauses (pause frequency, pause time, pause ratio) of patients with primary progressive aphasia (PPA) are analyzed to determine the type of primary progressive aphasia. and severity.

1 is an overall system diagram of the present disclosure.
Figure 2 is a block diagram of a server included in the classification device of the present disclosure.
Figure 3 is a block diagram of a terminal included in the classification device of the present disclosure.
Figure 4 is a flowchart showing a classification method according to the present disclosure.
Figure 5 is a conceptual diagram showing an example of a task displayed in the form of an image.
Figures 6 and 7 show the results of sensing the user's speech.

Like reference numerals refer to like elements throughout this disclosure. The present disclosure does not describe all elements of the embodiments, and general content or overlapping content between the embodiments in the technical field to which the present disclosure pertains is omitted. The term 'unit, module, member, block' used in the specification may be implemented as software or hardware, and depending on the embodiment, a plurality of 'unit, module, member, block' may be implemented as a single component, or It is also possible for one 'part, module, member, or block' to include multiple components.

Throughout the specification, when a part is said to be “connected” to another part, this includes not only direct connection but also indirect connection, and indirect connection includes connection through a wireless communication network. do.

Additionally, when a part "includes" a certain component, this means that it may further include other components rather than excluding other components, unless specifically stated to the contrary.

Throughout the specification, when a member is said to be located “on” another member, this includes not only cases where a member is in contact with another member, but also cases where another member exists between the two members.

Terms such as first and second are used to distinguish one component from another component, and the components are not limited by the above-mentioned terms.

Singular expressions include plural expressions unless the context clearly makes an exception.

The identification code for each step is used for convenience of explanation. The identification code does not explain the order of each step, and each step may be performed differently from the specified order unless a specific order is clearly stated in the context. there is.

In this specification, 'task' refers to content that induces the user to speak. Specifically, the content may be displayed in the form of at least one of text, image, and sound, and the user can visually check or listen to the content and then perform the corresponding task. Specific examples of the above tasks will be described later.

Hereinafter, the operating principle and embodiments of the present disclosure will be described with reference to the attached drawings.

In this specification, 'device according to the present disclosure' includes all various devices that can perform computational processing and provide results to the user. For example, the device according to the present disclosure may include all of a computer, a server device, and a portable terminal, or may take the form of any one.

Here, the computer may include, for example, a laptop, desktop, laptop, tablet PC, slate PC, etc. equipped with a web browser.

The server device is a server that processes information by communicating with external devices, and may include an application server, computing server, database server, file server, game server, mail server, proxy server, and web server.

The portable terminal is, for example, a wireless communication device that guarantees portability and mobility, such as PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), and PDA. (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, smart phone ), all types of handheld wireless communication devices, and wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted-device (HMD). may include.

The classification device according to the present disclosure may be implemented by at least one of a server and a terminal. Specifically, the device according to the present disclosure may be implemented by either a server or a terminal, or may be implemented as a system through data transmission and reception between the server and the terminal.

Below, a classification device according to the present disclosure will be described.

Referring to FIG. 1, the classification device according to the present disclosure may include at least one of a server 10 and a terminal 20.

The server 10 is connected to the terminal 20 through a network, and can receive data required for classification from the terminal 20 and then transmit the classification result to the terminal 20.

Meanwhile, it is obvious to those skilled in the art that the terminal 20 is not limited to the above-mentioned portable terminal, and may include a laptop, desktop, laptop, tablet PC, slate PC, etc. equipped with a processor. do.

As described above, the classification device according to the present disclosure can be implemented through data transmission and reception between the server 10 and the terminal 20. Additionally, the classification device according to the present disclosure includes a microphone 30 configured to receive voice information from the user 1. The microphone 30 may be included in the terminal 20 or may be connected as a separate device to enable data transmission and reception with the terminal 20.

Hereinafter, each of the server 10 and the terminal 20 for implementing the classification device according to the present disclosure will be described.

Figure 2 is a block diagram of a server included in the classification device of the present disclosure.

The server 100 according to the present disclosure may include at least one of a communication unit 110, a storage unit 120, and a processor 130.

The communication unit 110 may communicate with at least one of a terminal, an external storage (eg, a database 140), an external server, and a cloud server.

Meanwhile, in an external server or cloud server, it may be configured to perform at least a portion of the role of the processor 130. In other words, data processing or data computation can be performed on an external server or cloud server, and the present invention does not place any special restrictions on this method.

Meanwhile, the communication unit 110 may support various communication methods depending on the communication standard of the communicating object (eg, electronic device, external server, device, etc.).

For example, the communication unit 110 supports wireless LAN (WLAN), wireless-fidelity (Wi-Fi), wireless fidelity (Wi-Fi) Direct, digital living network alliance (DLNA), wireless broadband (WiBro), and WiMAX ( World Interoperability for Microwave Access), HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), 5G (5th Generation Mobile Telecommunication) , Bluetooth™, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi Direct, Wireless USB (Wireless Universal) Serial Bus) technology may be used to communicate with a communication target.

Next, the storage unit 120 can be configured to store various information related to the present invention. In the present invention, the storage unit 120 may be provided in the device itself according to the present invention. Alternatively, at least a portion of the storage unit 120 may refer to at least one of a database (DB) 140 and cloud storage (or cloud server). In other words, it can be understood that the storage unit 120 is sufficient as a space to store information necessary for the device and method according to the present invention, and there are no restrictions on physical space. Accordingly, hereinafter, the storage unit 120, the database 140, external storage, and cloud storage (or cloud server) will not be separately distinguished, and all will be referred to as the storage unit 120.

Next, the processor 130 may be configured to control the overall operation of the device related to the present invention. The processor 130 can process signals, data, information, etc. input or output through the components discussed above, or provide or process appropriate information or functions to the user.

The processor 130 may include at least one CPU (Central Processing Unit) and perform the function according to the present invention.

At least one component may be added or deleted in response to the performance of the components shown in FIG. 2. Additionally, it will be easily understood by those skilled in the art that the mutual positions of the components may be changed in response to the performance or structure of the device.

Hereinafter, the terminal included in the classification device of the present disclosure will be described in detail.

Figure 3 is a block diagram of a terminal included in the classification device of the present disclosure.

Referring to FIG. 3, the terminal 200 according to the present disclosure may include a communication unit 210, an input unit 220, a display unit 230, and a processor 240. The components shown in FIG. 3 are not essential for implementing the classification device according to the present disclosure, so the terminal described herein may have more or fewer components than the components listed above.

Among the components, the communication unit 210 may include one or more components that enable communication with an external device, for example, a broadcast reception module, a wired communication module, a wireless communication module, a short-range communication module, and location information. It may contain at least one of the modules.

Wired communication modules include various wired communication modules such as Local Area Network (LAN) modules, Wide Area Network (WAN) modules, or Value Added Network (VAN) modules, as well as USB (Universal Serial Bus) modules. ), HDMI (High Definition Multimedia Interface), DVI (Digital Visual Interface), RS-1302 (recommended standard 1302), power line communication, or POTS (plain old telephone service).

In addition to Wi-Fi modules and WiBro (Wireless broadband) modules, wireless communication modules include GSM (global System for Mobile Communication), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), and UMTS (universal mobile telecommunications system). ), TDMA (Time Division Multiple Access), LTE (Long Term Evolution), 4G, 5G, 6G, etc. may include a wireless communication module that supports various wireless communication methods.

The input unit 220 is for inputting image information (or signal), audio information (or signal), data, or information input from a user, and includes at least one of at least one camera, at least one microphone, and a user input unit. can do. Voice data or image data collected from the input unit can be analyzed and processed as a user's control command.

The display unit 230 is intended to generate output related to vision, hearing, or tactile sensation, and may include at least one of a display unit, an audio output unit, a haptip module, and an optical output unit. A touch screen can be implemented by forming a layered structure with the touch sensor or being integrated with the display unit. This touch screen functions as a user input unit that provides an input interface between the device and the user, and can simultaneously provide an output interface between the device and the user.

The display unit displays (outputs) information processed in the device. For example, the display unit may display execution screen information of an application program (eg, an application) running on the device, or UI (User Interface) and GUI (Graphic User Interface) information according to the execution screen information.

A task that allows the user 1 to perform a specific task may be displayed on the display unit. However, the task is not limited to this, and the task may be output through an audio output unit included in the display unit 230.

In addition to the above-described components, the above-described terminal may further include an interface unit and memory.

The interface unit serves as a passageway for various types of external devices connected to this device. These interface units include a wired/wireless headset port, an external charger port, a wired/wireless data port, a memory card port, and a port for connecting a device equipped with an identification module (SIM) ( port), an audio input/output (I/O) port, a video input/output (I/O) port, and an earphone port. In this device, appropriate control related to external devices connected to the interface unit can be performed.

The memory can store data that supports various functions of the device and programs for processor operation, and can store input/output data (e.g., music files, still images, videos, etc.), and can store data that supports various functions of the device. A number of application programs (application programs or applications) running in the device, data for operation of the device, and commands can be stored. At least some of these applications may be downloaded from an external server via wireless communication.

These memories include flash memory type, hard disk type, SSD type (Solid State Disk type), SDD type (Silicon Disk Drive type), and multimedia card micro type. , card-type memory (e.g., SD or It may include at least one type of storage medium among (only memory), PROM (programmable read-only memory), magnetic memory, magnetic disk, and optical disk. Additionally, the memory may be a database that is separate from the device, but is connected wired or wirelessly.

Meanwhile, the above-described terminal includes a processor 240. The processor includes a memory that stores data for an algorithm for controlling the operation of components within the device or a program that reproduces the algorithm, and at least one processor (not shown) that performs the above-described operations using the data stored in the memory. It can be implemented as: At this time, the memory and processor may each be implemented as separate chips. Alternatively, the memory and processor may be implemented as a single chip.

In addition, the processor may control any one or a combination of the above-described components in order to implement various embodiments according to the present disclosure described in the drawings below on the present device.

Meanwhile, at least one component may be added or deleted in accordance with the performance of the components shown in FIGS. 1 to 3. Additionally, it will be easily understood by those skilled in the art that the mutual positions of the components may be changed in response to the performance or structure of the device.

Below, a classification method using the above-described components will be described in detail. The classification method described later is implemented through data transmission and reception between the server and the terminal described above, and some steps of the method may be performed in either the server or the terminal. However, it is obvious to those skilled in the art that the method described below is not limited to this and can be performed independently in either a server or a terminal.

Hereinafter, a classification method using the components described in FIGS. 1 to 3 will be described.

Figure 4 is a flowchart of a classification method according to the present disclosure, Figure 5 is a conceptual diagram showing an example of a task displayed in the form of an image, and Figures 6 and 7 are the results of sensing the user's utterance.

First, the step of outputting content to the display unit proceeds (S110).

The content corresponds to the task and may be at least one of content that induces a conversation with a specific person (eg, an evaluator), an image, a paragraph, or a text consisting of a sentence.

In one embodiment, the display unit may display text guiding free conversation with a specific person or output a guidance message in the form of sound.

In one embodiment, the display unit may display an image so that the user can visually check the image and make a speech explaining the image. For example, as shown in FIG. 5, an image that induces the user to speak may be displayed on the display unit.

In one embodiment, the display unit may display a paragraph or sentence consisting of a plurality of sentences as text so that the user can visually check the text and read the text as it is.

In one embodiment, the display unit may output a paragraph or sentence consisting of a plurality of sentences in sound form so that the user listens to the paragraph or sentence output in sound form and then utters the same text.

Here, the content induces a corresponding utterance from the user. At this time, it is easy to predict the results in the order of sentences, paragraphs, images, and conversations. In other words, the simpler the task assigned to the user, the easier it is to predict the results.

Next, the user is guided to utter an utterance corresponding to the content, and the input unit collects the user's utterance information (S120).

After content is output through the display unit or audio output unit, a guidance comment guiding the user's speech may be output from the display unit or audio output unit.

The input unit records when the user speaks. The speech information refers to voice information generated by recording when a user performs a speech corresponding to a task.

When the content output on the display unit is content corresponding to a conversation, the user can freely conduct a conversation with a specific person, and the input unit can record the conversation and generate speech information. At this time, recording can be performed for a preset time.

Meanwhile, when the content output to the display unit is an image, the user may perform a speech explaining the displayed image for a preset time, and the input unit may record the speech to generate speech information.

Meanwhile, when the content output to the display unit is a paragraph or sentence, the user can perform a speech reading the displayed text, and the input unit may record the speech to generate speech information.

As an example, when collecting speech information, the collection may be conducted indoors in a quiet environment, and the person may be seated comfortably in a chair. In addition, it can be performed by placing the microphone within a preset distance (e.g., 15) from the user's mouth, maintaining the user's mouth and the microphone at 90 degrees, and then recording the user's speech.

In tasks that output content in the form of images (hereinafter referred to as picture description tasks), the user is given sufficient time to explore the image as a whole, and then is asked to explain the situations occurring in the image in as much detail as possible in sentences for one minute. This can be done in a directive manner (e.g., “Now explain in detail what is happening in this picture”). If you are unable to explain in sentences, you may re-instruct once, “Please explain in detail in sentences.”

Next, the step of predicting the user's primary progressive aphasia type based on the speech information proceeds (S130).

Here, the type of primary progressive aphasia can be predicted based on at least one of total speech time, total number of syllables, speech rate, articulation rate, pause frequency, pause time, and pause ratio extracted from speech information.

The total speech time refers to the time from producing the first syllable to completing the last syllable.

The total number of syllables refers to the total number of syllables calculated during the total speech time.

The speech rate is the total number of syllables divided by the total speech time, which means the total number of syllables calculated during the total speech time including pause time. At this time, the utterance time includes rest time. As an example, the time unit may be seconds, and syllables per second (SPS), which is an utterance unit analyzed as syllables, may be used.

The articulation speed is the total number of syllables ÷ articulation time, and in this case, the articulation time is the total speech time - pause time. In other words, the articulation time means the total number of syllables calculated during the articulation time excluding the pause time from the total utterance time.

Meanwhile, the pause is a section longer than the minimum time (for example, 0.1 seconds) among the parts where speech is not continued. If there is more than one pause, it can be calculated by adding them all together. Meanwhile, interprojections (e.g., rr...), other vocalizations (e.g., sighing, laughing, coughing, etc.), and noises can all be included in the rest period by manually adjusting the acoustic waveform.

A pause is a silent section without speech and is considered a physical pause in which no acoustic signal appears. Physical pause (silent pause) can be divided into articulatory pause and audible pause.

A physical pause is a silent section that appears in a speech and is a part without voice. These pauses do not produce an acoustic signal. An articulatory pause refers to a pause that inevitably occurs for the articulation of a specific phoneme, such as a stop or affricate. A perceptible pause is an audible pause within an utterance. These pauses can be used to measure the proportion of pauses within an utterance.

The pause frequency (total number of pauses) refers to the total number of pauses that appear during the total speech time. Here, the pause frequency can be defined in various ways depending on the standard. Specifically, the inter-utterance pause frequency is the frequency of pauses that appear between utterances. The frequency of pauses within an utterance is the frequency of pauses that appear within an utterance (between words + within words). Inter-word pause frequency is the frequency of pauses that appear between words. The frequency of pauses within a word is the frequency of pauses that appear between syllables in the middle of a word.

The pause time refers to the time when the pause appears during the total speech time. Pause can be measured in the portion of the waveform where a silent section of 0.1 to 2 seconds appears.

The pause ratio is (total pause time ÷ total ignition time), and means the ratio of the total pause time to the total ignition time.

Figures 6 and 7 are diagrams showing examples of Praat waveform analysis.

Referring to Figure 6, it can be seen that an articulation section (②) and a rest section (③) appear during the total speech section (①).

Referring to Figure 7, it can be seen that when uttering the sentence “Mom, there is a ball here,” pause (P) and disfluency (DF) occur.

Meanwhile, primary progressive aphasia type prediction may be to classify the user as one of nfvPPA, svPPA, lvPPA, and normal.

The above classification includes the total number of utterances extracted from the utterance information, the total number of syllables, the number of syllables per utterance, the number of words per utterance, the overall pause frequency, the pause frequency between utterances, the pause frequency within utterances (between words + within words), and the pause between words. Performed based on at least one of the following: frequency, intra-word pause frequency, overall pause period, inter-utterance pause time, intra-utterance pause time (inter-word + intra-word pause), inter-word pause time, intra-word pause time, speech rate, and articulation rate. It can be.

On the other hand, lvPPA has fewer speech movement disorders and true grammatical errors than nfvPPA, and fluency is relatively preserved. Additionally, in the picture description task, lvPPA patients had a faster speech rate, higher overall pause frequency, and longer pause times within speech than the svPPA patient group.

Meanwhile, in nfvPPA, the speed of articulator movements is reduced compared to lvPPA or svPPA in the picture description task, so speech and articulation speeds are relatively slow, and pause frequency and pause ratio also increase. Additionally, nfvPPA causes many grammatical errors when speaking.

Meanwhile, based on the results of the picture description task, it can be determined whether the user's primary progressive aphasia type is normal.

In one embodiment, the pause within a sentence is defined as 1 second or more, and the pause between sentences is defined as 2 seconds or more, and when the pause frequency in the picture description task is lower than the standard value, it is classified as normal among lvPPA, svPPA, nfvPPA, and normal. Can be classified.

Meanwhile, based on the results of the picture description task, it can be determined whether the user's primary progressive aphasia type is nfvPPA. In one embodiment, if the user does not meet the normal conditions, a pause in the sentence is defined as 1 second or more, and the pause between sentences is defined as 2nd or more, and in the picture description task, at least one of the following conditions is met: speech rate is below the standard value, articulation rate is below the standard value, pause frequency is above the standard value, and pause ratio is above the standard value. If satisfies, it can be classified as nfvPPA among nfvPPA, svPPA, and lvPPA.

On the other hand, if the user does not satisfy the conditions of normal and nfvPPA, the classification of svPPA and lvPPA will be performed by defining the minimum silence interval that defines the pause to be shorter than when determining normal and nfvPPA, and then utilizing the overall pause frequency and pause time within the utterance. You can.

In one embodiment, if the user does not satisfy the conditions of normal and nfvPPA, and the pause is defined as 0.1 seconds, if at least one condition of the overall pause frequency is greater than or equal to the reference value and the pause time within the utterance is greater than or equal to the standard value is met, It can be classified as lvPPA among svPPA and lvPPA.

If the user does not meet the conditions of normal, nfvPPA, and lvPPA, it can be classified as svPPA.

Finally, the step of outputting the prediction result proceeds (S140).

The prediction result may be output through the display unit and may be a classification result that classifies the user as one of nfvPPA, svPPA, lvPPA, and normal.

Hereinafter, an embodiment in which user speech information is collected and analyzed will be described in detail.

Example 1. Distinction between normal and patient groups

A total of two studies were conducted. Study 1 included a normal group and a PPA patient group (three types). The PPA group in Study 1 was a group of 36 patients diagnosed by a neurologist, including 22 lvPPA (69.5 ± 12.98 years), 8 svPPA (70.5 ± 8.21 years), and 6 nfvPPA (69.7 ± 7.03 years) by type. did. The normal control group consisted of 36 people (68.1 ± 2.85 years old). The picture description task described above was administered to each of these 72 people. The overall speech rate and pause characteristics of these utterances were analyzed. At this time, the IBM SPSS (Statistical Package for the Social Science, version 25.0) for Window program was used for statistical processing for analysis. Non-parametric covariance analysis was conducted to examine whether there were significant differences in speech rate, articulation rate, pause frequency, and pause ratio in picture description between the normal group and the overall PPA group, as well as between the normal group and the three subtypes of PPA patient groups, that is, the four groups. (Quade Nonparametric ANCOVA) was performed.

Study 2 targeted the normal group and PPA patient group (two types). The PPA group included 21 patients (74.9 ± 6.00) with lvPPA and 9 patients (73.1 ± 8.75 years old) with svPPA diagnosed by a neurologist. The normal control group consisted of 30 people (70.4 ± 7.42 years old). To examine whether there were differences in speech variables between the normal group and the two types of PPA, speech rate, articulation rate, pause frequency, and pause ratio were analyzed in picture description.

Example 2. Distinguish between nfvPPA and the other two PPA subtypes (lvPPA, svPPA)

In Study 1, we examined whether there was a difference between nfvPPA and two other PPA subtypes, namely the lvPPA and svPPA groups. In addition, a post hoc test was conducted to examine whether the measurement factors that showed significant differences differed between groups. For the reliability analysis, 30% of the total speech data was randomly selected and data from 24 people were analyzed, and intra-rater reliability and inter-rater reliability were calculated respectively.

Example 3. Distinction between lvPPA group and svPPA group

In both Study 1 and Study 2, we examined the differences between the lvPPA group and the svPPA group in the analysis factors. The difference between these two studies was the rest time applied during analysis. That is, in Study 2, unlike Study 1, the pause was set as short as 100 msec for analysis.

Analysis result 1. Comparison of performance between the overall PPA group and the normal group

In Study 1, the results of nonparametric analysis of covariance (Quade Nonparametric ANCOVA) on the difference in performance of the picture description task between the overall PPA group and the normal group are as follows. The frequency of pausing in the overall PPA group (5.97±2.9 times) was statistically significantly higher than that in the normal group (0.61±0.8 times) (F=14.092, p<.001). On the other hand, total speech time, total number of syllables, speech rate, and articulation rate were not significantly different between the normal group and the overall PPA group. In detail, there was no significant difference in total speech time between the overall PPA group (46.58±21.8 seconds) and the normal group (49.34±21.2 seconds) (F=.004, p>.05). There was also no significant difference in the total number of syllables between the total PPA group (104.10±52.8) and the normal group (145.50±58.8) (F=.386, p>.05). Additionally, there was no significant difference in speech rate between the overall PPA group (2.31±0.8 seconds) and the normal group (3.12±0.7 seconds) (F=1.825, p>.05). There was also no significant difference in articulation speed between the overall PPA group (3.34 ± 0.9 seconds) and the normal group (3.20 ± 0.6 seconds) (F=.071, p>.05).

When examining the differences between the normal and patient groups in Study 1, the performance of total number of syllables in the nfvPPA group (38.83±16.9) was significantly lower than that in the normal group (145.50±58.8) (p<.01). However, the differences in total number of syllables between the svPPA group (129.50 ± 51.8) and the normal group (145.50 ± 58.8), and between the lvPPA group (115.18 ± 43.9) and the normal group were not significant (p>.05).

Regarding speech rate performance, the nfvPPA group (0.97±0.3 seconds) had significantly slower firing rate than the normal group (3.31±0.7 seconds) (p<.01). However, the difference between the svPPA group (2.69 ± 0.5 seconds) and the lvPPA group (2.60 ± 0.6 seconds) compared to the normal group (3.31 ± 0.7 seconds) was not significant (p>.05).

Performance of articulation speed was also significantly slower in the nfvPPA group (2.02±0.4 seconds) than the normal group (3.20±0.6 seconds) (p<.001), and the other two types showed no significant difference from the normal group.

The performance of rest frequency was all in the nfvPPA group (8.17±2.3 times) (p<.01), svPPA group (5.33±3.5 times) (p<.05), and lvPPA group (5.35±2.6 times) (p<.05). It was significantly higher than the normal group (0.61±0.8 times).

Regarding rest time performance, the rest time in the nfvPPA group (21.82 ± 9.3 seconds), svPPA group (17.28 ± 13.3 seconds), and lvPPA group (13.58 ± 9.1 seconds) was significantly longer than that of the normal group (1.19 ± 1.5 seconds) (p <.05).

Rest ratio performance in the nfvPPA group (0.54±0.3%) was significantly higher than that in the normal group (0.03±0.0%) (p<.001). Additionally, the svPPA group (0.31±0.1%) and lvPPA group (0.29±0.1%) were significantly higher than the normal group (0.03±0.0%) (p<.05).

In Study 2, the overall rest frequency performance of the lvPPA group (22.29±10.71 times) (p<.05) was significantly higher than that of the normal group (16.07±7.09 times) (p<.05). The svPPA group (15.22±5.67 times) was no different from the normal group.

The performance of pause frequency between utterances was also significantly higher in the lvPPA group (7.81±3.79 times) (p<.05) than in the normal group (4.93±1.72 times) (p<.05). The svPPA group (5.56±1.94 times) had no difference from the normal group.

Analysis result 2. Performance comparison between nfvPPA and other two PPA subtypes (lvPPA, svPPA)

The results of nonparametric analysis of covariance (Quade Nonparametric ANCOVA) on the difference in performance of the picture description task between the three subtypes of the PPA group, the nfvPPA group, the svPPA group, and the lvPPA group, are as follows.

There was no significant difference in performance of total speech time between the three groups among the nfvPPA group (40.60±15.4 seconds), lvPPA group (46.30±20.8 seconds), and svPPA group (51.87±28.4 seconds) (F=.352, p>.05). ).

The performance of the total number of syllables among the three groups was higher in the nfvPPA group (38.83 ± 16.9) compared to the svPPA group (129.50 ± 51.8) (p < .01) and the lvPPA group (115.18 ± 43.9) (p < .01). The number was significantly small.

Regarding the performance of firing rate among the three groups, the nfvPPA group (0.97±0.3 seconds) had a higher firing rate than the svPPA group (2.69±0.5 seconds) (p<.05) or the lvPPA group (2.60±0.6 seconds) (p<.01). It was significantly slower. However, the difference between the svPPA group and the lvPPA group was not significant.

Regarding articulation speed performance among the three groups, the nfvPPA group (2.02±0.4 seconds) had significantly slower articulation speed than the svPPA group (3.73±0.6 seconds) or lvPP group (3.62±0.7 seconds) (p<.001). However, the difference between the svPPA group and the lvPPA group was not significant.

There was no significant difference in rest frequency performance among the three groups among the nfvPPA group, svPPA group, and lvPPA group, that is, the PPA subtype group (p>.05).

There was no significant difference in rest time performance among the three groups among the nfvPPA group, svPPA group, and lvPPA group, that is, the PPA subtype group (p>.05).

The rest ratio performance among the three groups was significantly higher in the nfvPPA group (0.54±0.3%) than the lvPPA group (0.29±0.1%) (p<.05). However, the differences between the nfvPPA group (0.54±0.3%) and svPPA group (0.31±0.1%) and between the svPPA group (0.31±0.1%) and lvPPA group (0.29±0.1%) were not significant (p>.05).

Analysis result 3. Comparison of performance between lvPPA group and svPPA group

In Study 1, there was no significant difference in performance in total number of syllables between the two groups between the lvPPA group (115.18 ± 43.9) and the svPPA group (129.50 ± 51.8). In Study 2, the performance of the total number of syllables between the two groups was not significant between the lvPPA group (184.08 ± 14.36) and the svPPA group (168.93 ± 27.03).

In Study 1, there was no significant difference in speech rate performance between the two groups between the lvPPA group (2.60 ± 0.6 seconds) and the svPPA group (2.69 ± 0.5 seconds). In Study 2, there was no significant difference in speech rate performance between the lvPPA group (2.45 ± 0.20 seconds) and the svPPA group (3.03 ± 0.38 seconds).

In Study 1, there was no significant difference in articulation speed performance between the lvPPA group (3.62 ± 0.7 seconds) and the svPPA group (3.73 ± 0.6 seconds) between the two groups (p>.05). In Study 2, there was no significant difference in articulation speed performance between the lvPPA group (6.27 ± 1.24 seconds) and the svPPA group (6.73 ± 0.93 seconds).

In Study 1, there was no significant difference in rest frequency performance between the lvPPA group and the svPPA group between the two groups (p>.05). However, in Study 2, there was a significant difference in rest frequency performance between the lvPPA group (21.99 ± 1.90) and the svPPA group (19.19 ± 3.57) between the two groups.

In Study 1, there was no significant difference in rest time performance between the lvPPA group and the svPPA group between the two groups (p>.05). Meanwhile, in Study 2, the performance of pauses within utterances between the two groups was significantly higher in the lvPPA group (12.25±1.45) than in the svPPA group (9.41±2.75) (p<.05).

In Study 1, there was no significant difference in rest ratio performance between the lvPPA group and the svPPA group between the two groups (p>.05).

For statistical analysis, a test was conducted using the SPSS (Statistical Package for the Social Science, version 26.0) program. To examine gender differences between the PPA group (lvPPA, svPPA) and the normal group, a chi-square test was used. test) was conducted, and one-way analysis of variance (ANOVA) was conducted to examine the differences in age, years of education, K-MMSE, and AQ between the PPA group (lvPPA, svPPA) and the normal group. In addition, gender , a generalized linear model was examined to analyze whether there were differences in major variables depending on PPA subtype when controlling for the influence of age, years of education, K-MMSE, and AQ. In this case, the major variable was a non-negative integer variable. Therefore, it was assumed to follow a Poisson distribution, and to emphasize the difference in research results between the normal group and the PPA group (lvPPA, svPPA), the normal group was set as the reference category and compared with other groups. Afterwards, the PPA group (lvPPA, svPPA) The estimated average values were compared.

To determine intra-examiner reliability and inter-examiner reliability, 6 people, representing 10% of the total data, were randomly selected. The first examiner was a researcher, and the second examiner was a graduate school graduate majoring in speech-language pathology and a speech-language rehabilitator with a level 2 certificate. Prior to full-scale analysis, the second examiner was made familiar with the analysis guidelines and analyzed after going through a practice process. As a result of the intra-examiner reliability analysis, the Cronbach's alpha coefficient was .977 for the total number of utterances, 1 for the total number of syllables, 1 for the total number of words, .985 for the number of syllables per utterance, .990 for the number of words per utterance, .991 for the total pause frequency, and .991 for the pause frequency between utterances. .848, intra-utterance pause frequency .942, inter-word pause frequency .921, intra-word pause frequency 1, total pause time .998, inter-utterance pause time .996, intra-utterance pause time .746, inter-word pause time .798, Intraword pause time was 1, speech rate was 1, and articulation rate was .884.

As a result of the inter-examiner reliability analysis, Cronbach's alpha coefficients were .971 for the total number of utterances, .993 for the total number of syllables, .996 for the total number of words, .934 for the number of syllables per utterance, .952 for the number of words per utterance, .971 for the total pause frequency, and .971 for the total number of pauses between utterances. Pause frequency 1, intra-utterance pause frequency .989, inter-word pause frequency .985, intra-word pause frequency 1, total pause time .999, inter-utterance pause time .991, intra-utterance pause time .882, inter-word pause time .845 , the pause time within a word was 1, the speech rate was .996, and the articulation rate was .966.

1. Comparison of utterance length

1) Comparison of total number of utterances

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with the total number of utterances as the dependent variable. As a result, the average total number of utterances in the lvPPA group, svPPA group, and normal group was 9.38±3.90, 7.11±1.69, and 6.50±1.85, respectively, and the difference between groups was statistically significant (p<.01).

The results of examining the generalized linear model with the total number of utterances as the dependent variable are as follows. The total number of utterances in the lvPPA group was estimated to be 3.19 (0.95) higher than the normal group (p=0.002), but the total number of utterances in the svPPA group was estimated to be no significant difference compared to the normal group (p=0.169). Additionally, age and K-MMSE were estimated to affect the total number of utterances (p<.05).

The results of the total number of utterances by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The estimated total number of utterances was 9.32±0.63 for the lvPPA group, 8.26±1.18 for the svPPA group, and 6.13±0.61 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (9.32±0.63) was significantly higher than the normal group (6.13±0.61) (p<.01).

2) Compare the total number of syllables

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with the total number of syllables as the dependent variable. As a result, the average total number of syllables for the lvPPA group, svPPA group, and normal group were 187.86±70.73, 138.78±59.52, and 143.93±61.32, respectively, and the difference between groups was statistically significant (p<.05).

The results of examining the generalized linear model with the total number of syllables as the dependent variable are as follows. The total number of syllables in the lvPPA group was estimated to be 48.07 (21.83) higher than the normal group (p=0.032), but the total number of syllables in the svPPA group was estimated to be no significant difference compared to the normal group (p=0.349). . Additionally, the regression coefficient for age was 3.25 (1.26), which was estimated to significantly affect the total number of syllables (p<.05).

The results of the total number of syllables by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The total number of syllables was estimated at 184.08±14.36 for the lvPPA group, 168.93±27.03 for the svPPA group, and 136.01±13.85 for the normal group, with no significant difference between the three groups.

3) Comparison of number of syllables per utterance

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with the number of syllables per utterance as the dependent variable. As a result, the average number of syllables per utterance for the lvPPA group, svPPA group, and normal group was 22.63±12.98, 19.64±8.51, and 23.22±9.25, respectively, and the differences between groups were not significant.

The results of examining the generalized linear model with the number of syllables per utterance as the dependent variable are as follows. As a result, no significant regression coefficient was estimated. This means that the effects of age, gender, years of education, K-MMSE, and AQ are not significant. Additionally, this means that there was no significant difference in the number of syllables per utterance in the lvPPA or svPPA group compared to the normal group.

The results of the number of syllables per utterance by PPA subtype estimated through a generalized linear model by controlling for each group's age, gender, years of education, K-MMSE, and AQ are as follows. The estimated number of syllables per utterance was estimated at 21.55±2.59 for the lvPPA group, 18.64±4.88 for the svPPA group, and 24.16±2.50 for the normal group, with no significant difference between the three groups.

4) Comparison of number of words per utterance

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with the number of words per utterance as the dependent variable. As a result, the average number of words per utterance for the lvPPA group, svPPA group, and normal group was 9.47±3.31, 7.26±2.70, and 7.55±2.55, respectively, and the difference between groups was statistically significant (p<.05).

The results of examining the generalized linear model with the number of words per utterance as the dependent variable are as follows. The number of words per utterance in the lvPPA group was estimated to be 2.15 (1.04) higher than that in the normal group (p=0.043), but the number of words per utterance in the svPPA group was estimated to be no significant difference compared to the normal group (p=0.600). . No other significant regression coefficients were estimated, meaning that the effects of age, gender, years of education, K-MMSE, and AQ were not significant.

The results of the number of words per utterance by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The estimated number of words per utterance was estimated at 9.44±0.68 for the lvPPA group, 8.17±1.28 for the svPPA group, and 7.29±0.66 for the normal group, with no significant difference between the three groups.

2. Comparison of pause frequency

1) Comparison of overall pause frequency

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with total pause frequency as the dependent variable. As a result, the average overall pause frequency of the lvPPA group, svPPA group, and normal group was 22.29±10.71, 15.22±5.67, and 16.07±7.09, and the difference between groups was statistically significant (p<.05).

The results of examining the generalized linear model with the overall rest frequency as the dependent variable are as follows. The total resting frequency of the lvPPA group was estimated to be 7.18 (2.88) higher than that of the normal group (p=0.016), but the total resting frequency of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.346). No other significant regression coefficients were estimated, meaning that the effects of age, gender, years of education, K-MMSE, and AQ were not significant.

The overall pause frequency results by PPA subtype, estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group, are as follows. The total rest frequency estimate was 21.99±1.90 for the lvPPA group, 19.19±3.57 for the svPPA group, and 14.81±1.83 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (21.99±1.90) was significantly higher than the svPPA group (19.19±3.57) and the normal group (14.81±1.83) (p<.05).

2) Comparison of pause frequency between utterances

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a univariate analysis of variance was conducted with the frequency of pauses between utterances as the dependent variable. As a result, the average pause frequency between utterances in the lvPPA group, svPPA group, and normal group was 7.81±3.79, 5.56±1.94, and 4.93±1.72, respectively, and the difference between groups was statistically significant (p<.001).

The results of examining a generalized linear model with the frequency of pauses between utterances as the dependent variable are as follows. The frequency of pauses between utterances in the lvPPA group was estimated to be 3.49 (0.91) higher than that of the normal group (p<0.001), but the frequency of pauses between utterances in the svPPA group was estimated to be no significant difference compared to the normal group (p=0.091). . Additionally, age (p<.05) and K-MMSE (p<.01) were estimated to affect the total number of utterances.

The results of the pause frequency between utterances by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The estimated pause frequency between utterances was 7.90±0.60 for the lvPPA group, 6.92±1.13 for the svPPA group, and 4.41±0.58 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (7.90±0.60) was significantly higher than the normal group (4.41±0.58) (p<.001).

3) Comparison of pause frequency within utterances (between words + within words)

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a univariate analysis of variance was conducted with the frequency of pauses in speech as the dependent variable. As a result, the average pause frequencies within speech for the lvPPA group, svPPA group, and normal group were 14.48±8.49, 9.67±4.87, and 12.20±7.46, respectively, and the differences between groups were not significant.

The results of examining a generalized linear model with pause frequency within utterance as the dependent variable are as follows. As a result, no significant regression coefficients were estimated. This means that the effects of age, gender, years of education, K-MMSE, and AQ are not significant. Additionally, this means that there was no significant difference in the frequency of pauses within speech in the lvPPA or svPPA groups compared to the normal group.

The results of the pause frequency in speech by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The estimated pause frequency within speech was estimated at 14.13±1.73 for the lvPPA group, 11.88±3.26 for the svPPA group, and 11.63±1.67 for the normal group, with no significant difference between the three groups.

4) Comparison of pause frequency between words

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with interword pause frequency as the dependent variable. As a result, the average inter-word pause frequency of the lvPPA group, svPPA group, and normal group was 13.90±8.48, 8.89±4.20, and 8.87±6.31, respectively, and the difference between groups was statistically significant (p<.05).

The results of examining a generalized linear model with inter-word pause frequency as the dependent variable are as follows. The inter-word pause frequency of the lvPPA group was estimated to be 4.99 (2.36) higher than that of the normal group (p=0.039), but the inter-word pause frequency of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.418). . Additionally, the regression coefficient for men was -3.93 (1.78), which was estimated to significantly affect the total number of syllables (p<.05).

The results of the interword pause frequency by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. The estimated inter-word pause frequency was estimated at 13.29±1.55 for the lvPPA group, 11.37±2.92 for the svPPA group, and 8.29±1.50 for the normal group, with no significant difference between the three groups.

5) Comparison of pause frequency within words

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with the frequency of pauses within words as the dependent variable. As a result, the average intra-word pause frequencies of the lvPPA group, svPPA group, and normal group were 1.19±1.33, 0.78±1.64, and 0.87±1.14, respectively, and the differences between groups were not significant.

The results of examining a generalized linear model with the frequency of pauses within words as the dependent variable are as follows. The regression coefficient for men was estimated to be -0.74±0.35 (p=0.040), and no other significant regression coefficients were estimated. This means that the effects of age, years of education, K-MMSE, and AQ were not significant, and that there was no significant difference in the frequency of pauses within words in the lvPPA or svPPA groups compared to the normal group.

The results of the intra-word pause frequency by PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. In terms of the estimated frequency of pauses within words, the lvPPA group was estimated to be 1.08±0.31, the svPPA group was estimated to be 1.05±0.58, and the normal group was estimated to be 0.81±0.30, with no significant differences between the three groups.

3. Rest time comparison

1) Comparison of total rest time

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with total rest time as the dependent variable. As a result, the average overall rest time of the lvPPA group, svPPA group, and normal group was 35.20±21.70, 22.08±8.81, and 14.77±8.31, respectively, and the difference between groups was statistically significant (p<.001).

The results of examining the generalized linear model with the total rest time as the dependent variable are as follows. The total rest time of the lvPPA group was estimated to be 22.15 (5.14) higher than that of the normal group (p<0.001), but the total rest time of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.147). In addition, the regression coefficient of K-MMSE was 2.10 (0.85), which was estimated to significantly affect the overall rest time (p<.05).

The overall rest time results for each PPA subtype, estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group, are as follows. The overall resting time estimate was 35.53±3.38 for the lvPPA group, 25.45±6.37 for the svPPA group, and 13.38±3.26 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (35.53±3.38) was significantly higher than the normal group (13.38±3.26) (p<.001).

2) Comparison of pause times between utterances

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with pause time between utterances as the dependent variable. As a result, the average pause times between utterances for the lvPPA group, svPPA group, and normal group were 23.06±19.62, 14.56±9.12, and 7.19±5.85, respectively, and the difference between groups was statistically significant (p<.001).

The results of examining a generalized linear model with pause time between utterances as the dependent variable are as follows. The pause time between utterances in the lvPPA group was estimated to be 16.83 (4.55) higher than that in the normal group (p=0.001), but the pause time between utterances in the svPPA group was estimated to be no significant difference compared to the normal group (p=0.192). . Additionally, years of education and K-MMSE were estimated to affect pause time between utterances (p<.05).

The results of the pause time between utterances by PPA subtype, estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group, are as follows. The estimated pause time between utterances was 23.31±2.99 for the lvPPA group, 16.07±5.63 for the svPPA group, and 6.48±2.89 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (23.31±2.99) was significantly higher than the normal group (6.48±2.89) (p<.01).

3) Comparison of rest time between feet and within words (between words + within words)

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with pause time within speech as the dependent variable. As a result, the average pause times within speech for the lvPPA group, svPPA group, and normal group were 12.14±7.65, 7.52±4.74, and 7.58±5.25, respectively, and the difference between groups was statistically significant (p<.05).

The results of examining the generalized linear model with pause time within utterance as the dependent variable are as follows. The pause time within utterance of the lvPPA group was estimated to be 5.32 (2.22) seconds higher than that of the normal group (p=0.020), but the pause time within utterance of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.487). It has been done. No other significant regression coefficients were estimated, meaning that the effects of age, gender, years of education, K-MMSE, and AQ were not significant.

The results of pause times within speech for each PPA subtype estimated through a generalized linear model, controlling for each group's age, gender, years of education, K-MMSE, and AQ, are as follows. The estimated pause time within speech was 12.22±1.46 for the lvPPA group, 9.38±2.75 for the svPPA group, and 6.90±1.41 for the normal group. In addition, as a result of confirming the difference between groups through a post hoc test, the lvPPA group (12.22±1.46) was significantly higher than the svPPA group (9.38±2.75) and the normal group (6.90±1.41) (p<.05).

4) Comparison of pause times between words

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with pause time between words as the dependent variable. As a result, the average pause times between words for the lvPPA group, svPPA group, and normal group were 11.80±7.57, 7.31±4.82, and 7.41±5.25, respectively, and the difference between groups was statistically significant (p<.05).

The results of examining the generalized linear model with pause time between words as the dependent variable are as follows. The inter-word pause time of the lvPPA group was estimated to be 5.24 (2.21) seconds higher than that of the normal group (p=0.021), but the inter-word pause time of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.474). It has been done. No other significant regression coefficients were estimated, meaning that the effects of age, gender, years of education, K-MMSE, and AQ were not significant.

The results of inter-word pause times for each PPA subtype estimated through a generalized linear model, controlling for each group's age, gender, years of education, K-MMSE, and AQ, are as follows. The estimated pause time between words was estimated at 11.93±1.45 for the lvPPA group, 9.24±2.74 for the svPPA group, and 6.69±1.40 for the normal group, with no significant difference between the three groups.

5) Comparison of rest time within phrases

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with pause time within words as the dependent variable. As a result, the average intraword pause times of the lvPPA group, svPPA group, and normal group were 0.34±0.79, 0.21±0.40, and 0.17±0.22, respectively, and the differences between groups were not significant.

The results of examining the generalized linear model with intra-word pause time as the dependent variable are as follows. As a result, no significant regression coefficients were estimated. This means that the effects of age, gender, years of education, K-MMSE, and AQ are not significant. Additionally, this means that there was no significant difference in intraword pause time in the lvPPA or svPPA group compared to the normal group.

The results of intra-word pause times for each PPA subtype, estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group, are as follows. The estimated intra-word pause time was estimated to be 0.29±0.12 for the lvPPA group, 0.14±0.23 for the svPPA group, and 0.21±0.12 for the normal group, with no significant difference between the three groups.

4. Comparison of speaking speed

1) Comparison of ignition speeds

Without controlling the influence of age, gender, years of education, K-MMSE, and AQ, a one-way analysis of variance was conducted with speech rate as the dependent variable. As a result, the average firing rates of the lvPPA group, svPPA group, and normal group were 2.57±0.76, 3.09±1.46, and 3.49±0.74, respectively, and the difference between groups was statistically significant (p<.01).

The results of examining the generalized linear model with speech rate as the dependent variable are as follows. The firing rate of the lvPPA group was estimated to be 1.13 (0.31) lower than that of the normal group (p=0.001), but the firing rate of the svPPA group was estimated to be no significant difference compared to the normal group (p=0.264). Additionally, the regression coefficient for years of education was 0.07 (0.03), which was estimated to significantly affect the total number of syllables (p<.05).

The firing rate results for each PPA subtype estimated through a generalized linear model by controlling the age, gender, years of education, K-MMSE, and AQ of each group are as follows. For the estimated firing rate, the lvPPA group was 2.45±0.20, the svPPA group was 3.03±0.38, and the normal group was 3.58±0.20. As a result of confirming the difference between groups through a post hoc test, the lvPPA group (2.45±0.20) was significantly lower than the normal group (3.58±0.20) (p<.001).

2) Comparison of articulation speed

One-way analysis of variance was conducted with articulation speed as the dependent variable without controlling for the influence of age, gender, years of education, K-MMSE, and AQ. As a result, the average articulation speed of the lvPPA group, svPPA group, and normal group was 6.27±1.24, 6.73±0.93, and 6.93±1.81, respectively, and the differences between groups were not significant.

The results of examining the generalized linear model with articulation speed as the dependent variable are as follows. As a result, no significant regression coefficients were estimated. This means that the effects of age, gender, years of education, K-MMSE, and AQ are not significant. Additionally, this means that there was no significant difference in articulation speed in the lvPPA or svPPA group compared to the normal group.

The results of articulation speed by PPA subtype estimated through a generalized linear model by controlling for each group's age, gender, years of education, K-MMSE, and AQ are as follows. The estimated articulation speed was estimated at 6.44±0.37 for the lvPPA group, 6.48±0.69 for the svPPA group, and 6.93±0.35 for the normal group, with no significant difference between the three groups.

The two most notable observations from the results of this experiment are as follows.

First, the lvPPA group showed a higher overall pause frequency compared to the svPPA group and the normal group.

Second, the lvPPA group was found to have longer pauses within speech than the svPPA group.

As described above, according to the present disclosure, differences in performance of speech rate (speech rate, articulation rate) and pauses (pause frequency, pause ratio) of patients with primary progressive aphasia (PPA) are analyzed to identify neurolinguistic disorders. Allows classification by type and severity.

Meanwhile, the disclosed embodiments may be implemented in the form of a recording medium that stores instructions executable by a computer. Instructions may be stored in the form of program code, and when executed by a processor, may create program modules to perform operations of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

Computer-readable recording media include all types of recording media storing instructions that can be decoded by a computer. For example, there may be Read Only Memory (ROM), Random Access Memory (RAM), magnetic tape, magnetic disk, flash memory, optical data storage device, etc.

As described above, the disclosed embodiments have been described with reference to the attached drawings. A person skilled in the art to which this disclosure pertains will understand that the present disclosure may be practiced in forms different from the disclosed embodiments without changing the technical idea or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be construed as limiting.

Claims (15)

In the primary progressive aphasia type classification device based on task performance results,
a memory configured to store task-related content;
A microphone configured to collect voice information from the user;
A display unit configured to display screen information; and
Outputting the task-related content on the display unit to induce the user to speak,
A primary progressive aphasia type classification device comprising a processor that classifies the user's primary progressive aphasia type based on the user's speech information collected through the microphone. According to paragraph 1,
The content is a primary progressive aphasia type classification device, characterized in that it includes at least one of text consisting of content, images, sentences, or paragraphs that induce conversation between the user and the evaluator. According to paragraph 2,
The processor,
A task performance result corresponding to the content inducing the conversation consisting of the sentence, a task performance result corresponding to the image, a task performance result corresponding to the text consisting of the paragraph, a task performance result corresponding to the text consisting of the sentence, A primary progressive aphasia type classification device characterized in that it classifies the type of primary progressive aphasia by assigning higher weights in that order. According to paragraph 3,
The processor,
Total number of utterances extracted from the above utterance information, total number of syllables, number of syllables per utterance, number of words per utterance, total pause frequency, pause frequency between utterances, pause frequency within utterances (between words + within words), pause frequency between words, Primary progressive aphasia based on at least one of the following: intra-word pause frequency, overall pause duration, inter-utterance pause time, intra-utterance pause time (inter-word + intra-word pause), inter-word pause time, intra-word pause time, speech rate, and articulation rate. Primary progressive aphasia type classification device, characterized in that it performs type classification. According to paragraph 4,
The processor,
The user's primary progressive aphasia is one of nonfluent/agrammatic variant of PPA (nfvPPA), semantic variant of PPA (svPPA), logopenic variant of PPA (lvPPA), and normal PPA. Primary progressive aphasia type classification device characterized by classifying types. According to clause 5,
The processor,
Based on the results of the task corresponding to the image above,
A primary progressive aphasia type classification device, characterized in that it determines whether the user's primary progressive aphasia type is normal. According to clause 6,
The processor,
Based on the results of the task corresponding to the text consisting of the above paragraph,
A primary progressive aphasia type classification device, characterized in that it determines whether the user's primary progressive aphasia type is nfvPPA. In clause 7,
The processor,
In the above speech information, silent sections of 100 ms (0.1 seconds) or more within and between speeches are set as pauses,
A primary progressive aphasia type classification device, characterized in that the user's primary progressive aphasia type is classified as one of svPPA and lvPPA based on at least one of the total pause frequency and pause time within the speech extracted from the speech information. In the method of classifying primary progressive aphasia based on task performance results performed by the device,
Outputting task-related content on a display unit to induce the user to speak;
Collecting user speech information through a microphone; and
A primary progressive aphasia type classification method comprising classifying the user's primary progressive aphasia type based on the user's speech information. According to clause 9,
The method for classifying primary progressive aphasia, characterized in that the content includes at least one of text consisting of content, images, sentences, or paragraphs that induce conversation between the user and the evaluator. According to clause 10,
The step of classifying the above types is,
A task performance result corresponding to the content inducing the conversation consisting of the sentence, a task performance result corresponding to the image, a task performance result corresponding to the text consisting of the paragraph, a task performance result corresponding to the text consisting of the sentence, A primary progressive aphasia type classification method characterized by classifying the type of primary progressive aphasia by assigning higher weights in that order. According to clause 11,
The step of classifying the above types is,
Total number of utterances extracted from the above utterance information, total number of syllables, number of syllables per utterance, number of words per utterance, overall pause frequency, pause frequency between utterances, pause frequency within utterances (between words + within words), pause frequency between words, Primary progressive aphasia based on at least one of the following: intra-word pause frequency, overall pause duration, inter-utterance pause time, intra-utterance pause time (inter-word + intra-word pause), inter-word pause time, intra-word pause time, speech rate, and articulation rate. A primary progressive aphasia type classification method characterized by performing type classification. According to clause 12,
The step of classifying the above types is,
The user's primary progressive aphasia is one of nonfluent/agrammatic variant of PPA (nfvPPA), semantic variant of PPA (svPPA), logopenic variant of PPA (lvPPA), and normal PPA. Primary progressive aphasia type classification method characterized by classifying types. According to clause 13,
The step of classifying the above types is,
Based on the results of the task corresponding to the image above,
A primary progressive aphasia type classification method characterized by determining whether the user's primary progressive aphasia type is normal. According to clause 14,
The step of classifying the above types is,
Based on the results of the task corresponding to the text consisting of the above paragraph,
A primary progressive aphasia type classification device, characterized in that it determines whether the user's primary progressive aphasia type is nfvPPA.