JP7408096B2 - Evaluation device and evaluation program - Google Patents

Description

Article 30, Paragraph 2 of the Patent Act applies Publication date August 21, 2020 Publication (publication name, volume number, issue number, relevant pages, publisher/publisher, etc.) Acoustical Society of Japan 2019 Autumn Research Presentation Collection of papers, p. 1071-1072 Publisher: Acoustical Society of Japan (General Incorporated Association) [Publications, etc.] Event date: September 4-6, 2020 (Announcement date: September 5, 2021) Meeting name, venue: Japan Onkyo Academic Conference 2019 Autumn Research Presentation Ritsumeikan University Gakuwako/Kusatsu Campus [Publications, etc.] Publication date June 10, 2020 Publications (Publication name, volume number, issue number, relevant pages, publisher/publisher, etc.) ) Innovation in Medicine and Healthcare, Proceedings of 8th KES-InMed 2020 p. 171-177 Publisher: KES International [Publications, etc.] Event date: June 17 to 19, 2020 (Announcement date: June 17, 2020) Meeting name, venue KES-INMED-20 Online Held at

The present invention relates to an evaluation device and an evaluation program for evaluating the presence or absence of food residue in a test subject's throat.

Dysphagia generally leads to aspiration, and aspiration causes aspiration pneumonia. Early detection of decline in swallowing function is important to prevent progression of dysphagia. Conventionally, many methods for evaluating the decline in swallowing function have been proposed. For example, a method called swallowing contrast examination (VF) or swallowing endoscopy (VE) is an evaluation method using radiation or a microscope. A method called cervical auscultation is a method for evaluating dysphagia by auscultating swallowing sounds or breathing sounds. Furthermore, there is also known a device that allows the user to confirm that swallowing has been performed, such as the device described in Patent Document 1 below.

Japanese Patent Application Publication No. 2018-007723

Since the conventional VF method or VE method described above is highly invasive, there is a need for a simple evaluation method that does not place a burden on the subject. Cervical auscultation is a method in which a doctor listens to the patient's breathing sounds or voices before and after swallowing, and distinguishes the acoustic features (choking sounds, bubbling sounds, wet hoarseness, etc.) contained in these sounds to diagnose the presence or absence of dysphagia. Since this is a method of evaluation based on the subjectivity of the doctor, accurate evaluation is difficult for those who are not experienced doctors. Further, even with the device described in Patent Document 1, it is not possible to evaluate the deterioration of swallowing function.

Therefore, the present invention has been made in view of such problems, and provides an evaluation device and an evaluation program that can easily and stably evaluate the presence or absence of food residue, which is a sign of a decline in swallowing function. The challenge is to provide the following.

One aspect of the present invention is an evaluation device that evaluates the presence or absence of food residue in the throat of a subject, and includes at least one processor, and the at least one processor is configured to utter a sound while the subject's head is facing forward. When the subject's head is turned sideways, a first audio signal is obtained using a microphone, and when the subject's head is turned sideways, a second audio signal is obtained using a microphone. frequency-analyzing each of the signal and the second audio signal to obtain a first spectral signal and a second spectral signal, and comparing distances and threshold values of the first spectral signal and the second spectral signal. Evaluates and outputs the presence or absence of food residue.

Alternatively, another aspect of the present invention is an evaluation program for evaluating the presence or absence of food residue in the throat of a test subject, the program comprising: an acquisition unit that acquires a first audio signal using a microphone; an acquisition unit that acquires a second audio signal using a microphone when the subject's head is uttered with the head turned sideways; and an analysis unit that frequency-analyzes each of the second audio signals to obtain a first spectrum signal and a second spectrum signal, and compares the distance between the first spectrum signal and the second spectrum signal with a threshold value. By doing so, it functions as an evaluation unit that evaluates and outputs the presence or absence of food residue.

According to any of the above aspects, the first audio signal is a recording of audio from a subject with the head facing forward, and the second audio signal is recording audio from the subject with the head facing sideways. A first spectral signal obtained by frequency analysis from the first audio signal and a second spectral signal obtained by frequency analysis from the second audio signal. Based on the comparison result between the distance and the threshold value, the presence or absence of food residue in the subject's throat is evaluated and output. This makes it possible to non-invasively and stably evaluate the presence or absence of food residue, which is one sign of a decline in swallowing function.

In the above aspect, it is preferable that at least one processor evaluates the presence or absence of food residue based on distance in a predetermined frequency band. It is also preferable that the predetermined frequency band includes 4.5 kHz. The present inventors have found that when there is no food residue, anti-resonance characteristics appear in the frequency band around 4.5 kHz in the spectrum of the audio signal when facing the front of the subject. By utilizing such characteristics and including the frequency of 4.5 kHz in the predetermined frequency band, the presence or absence of food residue can be evaluated more stably.

The at least one processor also acquires a third audio signal using a microphone when the subject's head is turned sideways and then turned back to the front, and the third audio signal is subjected to frequency analysis. Preferably, the presence or absence of food residue is reevaluated and output by acquiring a third spectral signal and comparing the distance between the first spectral signal and the third spectral signal with a threshold value. . The present inventors found that the spectrum of the audio signal acquired with the subject's head turned to the side and then back to the front is different from the spectrum of the audio signal acquired with the subject facing forward in the absence of food residue. We found that the signal spectra have similar characteristics. By utilizing such a phenomenon and re-evaluating using the distance between the first spectrum signal and the third spectrum signal, the presence or absence of food residue can be evaluated more stably.

Further, the at least one processor acquires a fourth audio signal using a microphone when the subject's head before ingesting food is uttered with the subject's head facing forward, and A fifth audio signal is obtained using a microphone when uttering a sound with the device facing sideways, and frequency analysis is performed on each of the fourth audio signal and the fifth audio signal to obtain a fourth spectral signal and By obtaining a fifth spectral signal and comparing the distance between the first spectral signal and the second spectral signal and a threshold value determined from the distance between the fourth spectral signal and the fifth spectral signal, It is also suitable to evaluate the presence or absence. In this case, the threshold value can be efficiently and appropriately set based on the distance between the spectral signals in the two states when there is no food residue, and the presence or absence of food residue can be evaluated more stably.

Furthermore, the at least one processor acquires a sixth audio signal using the microphone when the subject utters the subject's head in a frontal state after shaking the head sideways multiple times, and performs frequency analysis on the sixth audio signal. It is also preferable that the presence or absence of food residue is re-evaluated and output by acquiring a sixth spectral signal and comparing the distance between the first spectral signal and the sixth spectral signal with a threshold value. . We found that the spectrum of the audio signal acquired with the subject facing forward after multiple sideways head shakes was the same as that obtained with the subject facing forward in the absence of food residue. We found that the characteristics are similar to the spectrum of the audio signal. By utilizing such a phenomenon and re-evaluating using the distance between the first spectrum signal and the sixth spectrum signal, the presence or absence of food residue can be evaluated more stably.

According to one aspect of the present invention, it is possible to easily and stably evaluate the presence or absence of food residue, which is a sign of a decline in swallowing function.

FIG. 1 is a block diagram showing a schematic configuration of an evaluation device 1 according to an embodiment. 2 is a diagram showing a hardware configuration of evaluation control 1 in FIG. 1. FIG. 2 is a flowchart showing an operation procedure in evaluation processing by the evaluation device 1 of FIG. 1. FIG. 2 is a flowchart showing an operation procedure in evaluation processing by the evaluation device 1 of FIG. 1. FIG. 2 is a diagram showing waveforms of a first spectrum signal _{SF and second spectrum signals S R and} S _L acquired by the spectrum acquisition unit 13 of FIG. 1. FIG. FIG. 2 is a block diagram showing the configuration of an evaluation program according to an embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description, the same elements or elements having the same function will be denoted by the same reference numerals, and redundant description will be omitted.

FIG. 1 is a block diagram showing a schematic configuration of an evaluation device 1 according to an embodiment. As shown in FIG. 1, an evaluation device 1 is a device for evaluating the presence or absence of food residue in the piriform fossa in the throat of a user (subject). Note that the food intake to be evaluated by the evaluation device 1 also includes water intake. The evaluation device 1 is configured to be able to receive an audio signal, which is an analog signal, from a close-talking microphone M, which is an acoustic microphone worn close to the user's mouth, via a cable. It has a function to execute evaluation processing using audio signals. However, the evaluation device 1 may be configured to be able to receive the audio signal from the close-talk type microphone M using a wireless signal such as Bluetooth (registered trademark) or wireless LAN. The close-talk type microphone M is a detection device that is used close to the user's mouth and generates an audio signal by detecting vibrations in the air near the mouth in response to speech. However, the close-talk type microphone M may be replaced with any other type of microphone as long as it is an acoustic microphone that can detect vocalizations as vibrations through the atmosphere, or may be replaced with a sound-collecting microphone such as a pin microphone or vocal microphone. good.

Here, the evaluation device 1 includes, as functional components, an instruction signal output section 11, a voice acquisition section (acquisition section) 12, a spectrum acquisition section (analysis section) 13, a distance calculation section 14, an evaluation section 15, and an evaluation section. It is configured to include a result output section 16.

FIG. 2 is a block diagram showing the hardware configuration of the evaluation device 1. As shown in FIG. 2, the evaluation device 1 is realized by an arithmetic device 50 typified by a smartphone, a tablet terminal, a computer terminal, or the like. The arithmetic device 50 physically includes a CPU (Central Processing Unit) 101 which is a processor, a RAM (Random Access Memory) 102 or a ROM (Read Only Memory) 103 which is a recording medium, a communication module 104, an input/output device, etc. , each of which is electrically connected internally. The input/output device 105 is a keyboard, a mouse, a display device, a touch panel display device, a speaker, or the like. Each functional unit of the evaluation device 1 described above operates the communication module 104, input/output device 105, etc. under the control of the CPU 101 by loading the evaluation program of the embodiment onto hardware such as the CPU 101 and the RAM 102. This is realized by operating the program and reading and writing data in the RAM 102.

The evaluation device 1 is composed of at least one arithmetic device 50, but may be composed of two or more arithmetic devices 50. When a plurality of arithmetic devices 50 are used, one evaluation device 1 is logically constructed by connecting these arithmetic devices 50 via a communication network such as the Internet or an intranet.

Hereinafter, returning to FIG. 1, the functions of each functional unit of the evaluation device 1 will be described in detail.

The instruction signal output unit 11 outputs an instruction signal to a user when evaluating the user. That is, when acquiring the user's voice signal, the instruction signal output unit 11 outputs an instruction signal that instructs the user to maintain a head posture and speak. Examples of the format for outputting the instruction signal include outputting text information on a screen using a display device or the like, or audio output using a speaker or the like. Specifically, the instruction signal output unit 11 outputs a first instruction signal that instructs to utter a predetermined sound (for example, a steady vowel) after turning the head forward after ingesting food; A second instruction signal instructs the head to turn sideways (to the right or left) and then emit a predetermined sound, and a second instruction signal instructs the patient to turn the head sideways after ingesting food, then turn it forward, and then emit a predetermined sound. The third instruction signal instructs the patient to turn the head forward before ingesting food and then emit a predetermined sound.The fourth instruction signal instructs the patient to emit a predetermined sound after turning the head sideways before ingesting food. After the start of the evaluation process, a fifth instruction signal that instructs the user to make a predetermined sound, or a sixth instruction signal that instructs the user to utter a predetermined sound after shaking the head sideways multiple times after ingesting the food, is transmitted in a predetermined order and in a predetermined manner. Output at the timing of. Here, the user's head posture controlled by the instruction from the instruction signal output unit 11 is the head posture based on the frontal orientation of the user's upper body.

The audio acquisition unit 12 receives the user's audio signal from the close-talk type microphone M as the first to sixth audio signals in accordance with the output timing of each of the first to sixth instruction signals by the instruction signal output unit 11. Acquire (receive). That is, the audio acquisition unit 12 acquires the first audio signal generated when the user utters a voice after ingesting food with the user's head facing forward by outputting the first instruction signal. Furthermore, the audio acquisition unit 12 outputs the second instruction signal to acquire a second audio signal when the user utters a voice while the user's head is turned sideways after ingesting food. In addition, the audio acquisition unit 12 outputs the third instruction signal to acquire a third audio signal when the user after ingesting food turns his head sideways and then returns it to the front and utters a sound. do. Furthermore, the audio acquisition unit 12 outputs the fourth instruction signal to acquire a fourth audio signal generated when the user utters a voice with the user's head facing forward before ingesting food. Furthermore, the audio acquisition unit 12 outputs the fifth instruction signal to acquire a fifth audio signal generated when the user utters a voice while the user's head is turned sideways before ingesting food. Furthermore, the audio acquisition unit 12 outputs the sixth instruction signal to acquire a sixth audio signal generated when the user utters a voice after shaking the user's head sideways multiple times after ingesting food.

The spectrum acquisition unit 13 converts the first to sixth audio signals from the time domain to the frequency domain by performing frequency analysis on each of the first to sixth audio signals acquired by the audio acquisition unit 12. The first to sixth spectral signals, which are power spectra converted to , are obtained. For example, the spectrum acquisition unit 13 acquires the first to sixth spectrum signals by Fourier transforming the first to sixth audio signals.

The distance calculation unit 14 calculates a distance that is the basis for evaluating food residue. Specifically, the distance calculation unit 14 calculates the average value (or maximum value or minimum value) of the distance between the first spectrum signal and the second spectrum signal in a predetermined frequency band as the distance. This predetermined frequency band is, for example, 4.0 kHz to 5.0 kHz, and is a frequency band that includes at least 4.5 kHz. The distance calculation unit 14 calculates, for example, the average value (or maximum value, minimum value) of distances at a plurality of frequencies obtained by dividing a predetermined frequency band at regular intervals. At this time, when calculating the distance, the distance calculation unit 14 may weight the frequency (for example, a frequency around 4.5 kHz). Similarly, the distance calculating unit 14 calculates the distance between the first spectrum signal and the third spectrum signal, the distance between the fourth spectrum signal and the fifth spectrum signal, or the distance between the first spectrum signal and the third spectrum signal, or the distance between the first spectrum signal and the third spectrum signal. Calculate the distance between the signal and the sixth spectral signal.

In addition to the method of directly calculating the distance between spectra (for example, log-Euclidean distance between spectra) as described above, the distance calculation unit 14 uses a method generally used for pattern identification (for example, support vector machine (SVM)). )) may be used.

The evaluation unit 15 evaluates whether food remains in the user's piriform fossa based on the distance calculated by the distance calculation unit 14. Here, the principle of evaluation by the evaluation section 15 will be explained. The present inventors used a vocal tract model to simulate and analyze a state in which food remains in a user's pyriform fossa, a region next to the entrance to the esophagus. It has been found that acoustic characteristics corresponding to the anti-resonance state around .5 kHz appear. Specifically, it was found that the minimum value around 4.5 kHz tends to disappear when food residue occurs. On the other hand, the inventor found that even when the user twists the head from side to side, the piriform fossa is compressed, similar to when a vocal tract model is used to simulate food residue in the piriform fossa. It was also found that acoustic features appeared. The evaluation unit 15 performs evaluation using these acoustic features.

Specifically, the evaluation unit 15 compares the distance between the first spectrum signal and the second spectrum signal with a predetermined threshold, and if the distance is less than or equal to the predetermined threshold, the evaluation unit 15 It is evaluated that there is food residue. At this time, the predetermined threshold value is determined in advance based on the distance between the fourth spectrum signal and the fifth spectrum signal acquired before food intake. For example, a value of a certain percentage of this distance is set as the threshold value. Further, the predetermined threshold value may be a value set in advance by an operator of the evaluation device 1 or the like. On the other hand, if the distance between the first spectrum signal and the second spectrum signal exceeds a predetermined threshold, the evaluation unit 15 evaluates that there is no food remaining in the piriform fossa.

Furthermore, the evaluation unit 15 compares the distance between the first spectrum signal and the third spectrum signal with another threshold value, and if the distance is less than the threshold value, the evaluation unit 15 determines that there is no food residue in the piriform fossa. If the distance exceeds the threshold, it is reevaluated that there is food residue in the piriform fossa. This evaluation utilizes the fact that when a user with food residue turns his or her head sideways and then returns it to the front, the food residue tends to be removed. The evaluation unit 15 then generates a final evaluation result that takes into account the evaluation results using the first and second spectrum signals and the re-evaluation results using the first and third spectrum signals. For example, when the evaluation result and the re-evaluation result match, the evaluation unit 15 generates the matched result as the final evaluation result. Furthermore, the evaluation unit 15 may include accuracy information in the final evaluation result depending on whether or not the evaluation result and the re-evaluation result match.

The evaluation result output unit 16 outputs the final evaluation result generated by the evaluation unit 15. For example, the evaluation result output unit 16 outputs the evaluation result of the presence or absence of food residue in the piriform fossa and its accuracy information on a screen such as text information using a display device or the like, or outputs audio using a speaker or the like. Output by

Next, the flow of operations in the evaluation process of the evaluation device 1 described above will be explained. 3 and 4 are flowcharts showing operational procedures in evaluation processing by the evaluation device 1.

The evaluation process regarding food residue in the user's piriform fossa is started each time the user or the like inputs an instruction to the evaluation device 1. First, referring to FIG. 3, the fourth instruction signal and the fifth instruction signal are sequentially outputted by the instruction signal output unit 11 at a predetermined time interval before the user takes food ( Step S101). As a result, the user utters the predetermined sound with the user's head facing forward, and then utters the predetermined sound with the user's head facing sideways. Accordingly, the audio acquisition unit 12 sequentially acquires the fourth audio signal and the fifth audio signal (step S102).

Next, the instruction signal output unit 11 sequentially outputs the first instruction signal, the second instruction signal, and the third instruction signal with a predetermined time interval after the user takes food. (Step S103). As a result, the user utters a predetermined sound with the user's head facing forward, then utters the predetermined sound with the user's head facing sideways, and then A predetermined sound is emitted when the device is turned back to the front. Accordingly, the audio acquisition unit sequentially acquires the first audio signal, the second audio signal, and the third audio signal (step S104).

Furthermore, the spectrum acquisition unit 13 performs frequency analysis on the first to fifth audio signals acquired so far, and as a result, first to fifth spectrum signals are acquired (step S105). Thereafter, the distance calculation unit 14 calculates the distance between the fourth spectrum signal and the fifth spectrum signal in a predetermined frequency band, and a value of a certain percentage of the distance is set as the first threshold TH ₁ ( Step S106).

Next, the distance calculation unit 14 calculates the distance between the first spectrum signal and the second spectrum signal in a predetermined frequency band (step S107). Then, the evaluation unit 15 determines whether the distance exceeds the first threshold _TH1 (step S108). As a result of the determination, if the distance exceeds the first threshold TH ₁ (step S108; Yes), the evaluation unit 15 evaluates that "no food remains in the user's piriform fossa" (step S109). . On the other hand, if the distance is less than or equal to the first threshold _TH1 (step S108; No), the evaluation unit 15 evaluates that "food remains in the user's piriform fossa" (step S110).

4, the distance calculation unit 14 calculates the distance between the first spectrum signal and the third spectrum signal in a predetermined frequency band (step S111). Then, the evaluation unit 15 determines whether the distance exceeds a preset second threshold _TH2 (step S112). As a result of the determination, if the distance exceeds the second threshold _TH2 (step S112; Yes), the evaluation unit 15 re-evaluates that "there is food residue in the user's piriform fossa" (step S113). ). On the other hand, if the distance is less than or equal to the second threshold _TH2 (step S112; No), the evaluation unit 15 re-evaluates that there is "no food residue in the user's piriform fossa" (step S114).

Then, the evaluation unit 15 generates a final evaluation result based on the evaluation result and the re-evaluation result, and the evaluation result output unit 16 outputs the final evaluation result (step S115). As the final evaluation result, the above evaluation result is output together with accuracy information. As the accuracy information, for example, information indicating the accuracy of the evaluation, such as "high" or "low", is output depending on the match/mismatch between the evaluation result and the re-evaluation result.

FIG. 5 shows the waveforms of the first spectrum signal S _F and the second spectrum signals S _R and S _L acquired by the spectrum acquisition unit 13 when the user utters the vowel /e/. The second spectral signal S _R indicates a waveform obtained when the user is facing right, and the second spectral signal S _L indicates a waveform obtained when the user is facing left. In this example, in the first spectrum signal _SF , a remarkable minimum point appears near 4.2 kHz. On the other hand, in the second spectrum signals S _R and S _L acquired with the user facing sideways, the minimum point around 4.2 kHz has disappeared. By evaluating the difference in waveform between the first spectral signal S _F and the second spectral signals S _R and S _L , the evaluation unit 15 determines whether food remains in the user's piriform fossa. It can be evaluated that there is no.

Next, with reference to FIG. 6, an evaluation program for causing a computer to function as the evaluation apparatus 1 will be described.

The evaluation program P1 includes a main module P10, an instruction signal output module P11, a voice acquisition module P12, a spectrum acquisition module P13, a distance calculation module P14, an evaluation module P15, and an evaluation result output module P16.

The main module P10 is a part that centrally controls the operation of the evaluation device 1. The functions realized by executing the instruction signal output module P11, the audio acquisition module P12, the spectrum acquisition module P13, the distance calculation module P14, the evaluation module P15, and the evaluation result output module P16 are the instruction signal output unit 11, The functions are similar to those of the audio acquisition section 12, spectrum acquisition section 13, distance calculation section 14, evaluation section 15, and evaluation result output section 16.

The evaluation program P1 is provided by, for example, a computer-readable recording medium such as a CD-ROM, DVD, or ROM, or a semiconductor memory. Furthermore, the evaluation program P1 may be provided via a network as a computer data signal superimposed on a carrier wave.

According to the above-mentioned evaluation device 1, the first audio signal is a recording of the user's voice with the user's head facing forward, and the second audio signal is the audio of the user with the user's head facing sideways. A first spectral signal obtained by frequency analysis from the first audio signal and a second spectral signal obtained by frequency analysis from the second audio signal. Based on the comparison result between the distance and the threshold value, the presence or absence of food residue in the user's piriform fossa is evaluated and output. This makes it possible to non-invasively and stably evaluate the presence or absence of food residue, which is one sign of a decline in swallowing function. In particular, if the evaluation device 1 is used, the presence or absence of food residue can be evaluated using audio signals obtained by having the same user speak over a short period of time, and there is no need to collect large amounts of learning data from disabled people in advance. . Thereby, simple and stable evaluation can be achieved without being affected by changes in equipment, environment, or user's physical condition. As a result, the evaluation device 1 can contribute to early detection of deterioration of swallowing function. In other words, it is possible to implement screening at a stage before noticeable symptoms such as choking appear in the user.

The evaluation device 1 of this embodiment evaluates the presence or absence of food residue based on the distance in a predetermined frequency band, particularly in a frequency band including 4.5 kHz. In this way, by including the frequency around 4.5 kHz where acoustic features due to food residue in the piriform fossa are likely to appear, the presence or absence of food residue can be evaluated more stably.

Furthermore, the evaluation device 1 of this embodiment re-evaluates the presence or absence of food residue by comparing the distance between the first spectrum signal and the third spectrum signal and the threshold value. The inventors found that the spectrum of the audio signal acquired with the user turning their head to the side and then back to the frontal position is the same as the spectrum of the audio signal acquired with the user facing forward in the absence of food residue. We found that the acoustic features are close to the spectrum of the signal. By utilizing such a phenomenon and re-evaluating using the distance between the first spectrum signal and the third spectrum signal, the presence or absence of food residue can be evaluated more stably.

Moreover, the evaluation device 1 of this embodiment sets the first threshold value TH ₁ from the distance between the fourth spectrum signal and the fifth spectrum signal. In this case, the threshold value can be efficiently and appropriately set based on the distance between the spectral signals in the two states when there is no food residue, and the presence or absence of food residue can be evaluated more stably.

Although various embodiments of the present invention have been described above, the present invention is not limited to the above embodiments, and may be modified or applied to other things without changing the gist of each claim. It may be something.

For example, when performing re-evaluation, the evaluation unit 15 of the evaluation device 1 uses the distance between the first spectrum signal and the sixth spectrum signal in a predetermined frequency band, and compares the distance with a prescribed threshold value. Therefore, the presence or absence of food residues may be re-evaluated. We found that the spectrum of the audio signal acquired with the user facing forward after multiple sideways head shakes was the same as that obtained with the user facing forward in the absence of food residue. We found that the characteristics are similar to the spectrum of the audio signal. By utilizing such a phenomenon and re-evaluating using the distance between the first spectrum signal and the sixth spectrum signal, the presence or absence of food residue can be evaluated more stably.

DESCRIPTION OF SYMBOLS 1... Evaluation device, 12... Audio acquisition part (acquisition part), 13... Spectrum acquisition part (analysis part), 14... Distance calculation part, 15... Evaluation part, M... Close-talking type microphone, P1... Evaluation program.

Claims (7)

An evaluation device for evaluating the presence or absence of food residue in the throat of a subject,
comprising at least one processor;
The at least one processor includes:
Obtaining a first audio signal using a microphone when uttering with the subject's head facing forward;
Obtaining a second audio signal using the microphone when making the subject speak with the subject's head turned sideways,
frequency-analyzing each of the first audio signal and the second audio signal to obtain a first spectral signal and a second spectral signal;
Evaluating and outputting the presence or absence of the food residue by comparing the distance of the first spectral signal and the second spectral signal with a threshold value;
Evaluation device. The at least one processor includes:
Evaluating the presence or absence of the food residue based on the distance in a predetermined frequency band;
The evaluation device according to claim 1. the predetermined frequency band includes 4.5kHz;
The evaluation device according to claim 2. The at least one processor includes:
Obtaining a third audio signal using the microphone when the subject's head is turned sideways and then turned back to the front and utters,
frequency-analyzing the third audio signal to obtain a third spectrum signal;
Re-evaluating and outputting the presence or absence of the food residue by comparing the distance of the first spectral signal and the third spectral signal with a threshold value;
The evaluation device according to any one of claims 1 to 3. The at least one processor includes:
Obtaining a fourth audio signal using a microphone when the subject's head is facing forward before ingesting food, and when the subject is made to speak,
Obtaining a fifth audio signal using the microphone when uttering a sound with the subject's head turned sideways before ingesting food;
frequency-analyzing each of the fourth audio signal and the fifth audio signal to obtain a fourth spectral signal and a fifth spectral signal;
The presence or absence of the food residue is evaluated by comparing the distance between the first spectrum signal and the second spectrum signal with a threshold determined from the distance between the fourth spectrum signal and the fifth spectrum signal. ,
The evaluation device according to any one of claims 1 to 4. The at least one processor includes:
Obtaining a sixth audio signal using the microphone when the subject's head is uttered while facing forward after shaking the head sideways multiple times;
frequency-analyzing the sixth audio signal to obtain a sixth spectrum signal;
Re-evaluating and outputting the presence or absence of the food residue by comparing the distance of the first spectral signal and the sixth spectral signal with a threshold value;
The evaluation device according to any one of claims 1 to 5. An evaluation program for evaluating the presence or absence of food residue in the throat of a subject,
computer,
an acquisition unit that acquires a first audio signal using a microphone when the subject's head is uttered with the head facing forward;
an acquisition unit that acquires a second audio signal using the microphone when the subject's head is turned sideways and the subject is uttered;
an analysis unit that frequency-analyzes each of the first audio signal and the second audio signal to obtain a first spectrum signal and a second spectrum signal; and the first spectrum signal and the second spectrum. an evaluation unit that evaluates and outputs the presence or absence of food residue by comparing the distance of the signal with a threshold;
An evaluation program that functions as an evaluation program.

Images