RU2765108C1 - Method and hardware and software complex for pre-medical preliminary classifying multifactorial assessment of possibility of human auditory analyzer during mass preventive examinations of the population - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: present invention refers to the means and methods that provide automated execution of diagnostic procedures in terms of performing a pre-medical preliminary classifying multifactorial assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population. A method for pre-medical preliminary classifying multifactorial assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population is proposed, performed using a computing device connected to audio signal playback devices and containing stages in which: using a computing device, a primary test speech sequence (TSS) is formed, which is sentences consisting of the first number of words based on a matrix test; a noise competing sound is formed for primary TSS; the primary TSS is reproduced using audio signal playback devices made in the form of air and bone sound transmission headphones, while the TSS is reproduced simultaneously with the noise competing sound at the first signal-to-noise ratio using speech simulation based on a deep machine learning model; the user receives an oral response; performs automatic analysis of the user's oral response to TSS recognition by converting it into a text form and analyzing the correctness of the response using a machine learning model; moreover, on the basis of the performed analysis of the user's oral responses, a dynamic change in the complexity of the assessment is carried out, in which, according to the result of each automatic analysis, a change in the number of words in sentences forming the TSS and/or the signal/noise ratio of the reproduced signal is performed; an assessment of the possibility of the user's auditory analyzer is performed based on responses when reproducing a test speech sequence.

EFFECT: invention provides an automated pre-medical preliminary classifying assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population.

6 cl, 14 dwg, 3 tbl

Description

FIELD OF TECHNOLOGY

This technical solution relates to means and methods that provide automated execution of diagnostic procedures in terms of performing a pre-medical preliminary classifying multifactorial assessment of the possibility of a person's auditory analyzer during mass preventive examinations of the population.

BACKGROUND OF THE INVENTION

The main characteristic of any speech transmission channel, including the auditory tract of human speech perception, is the intelligibility of speech. To determine this characteristic in technical communication systems, a statistical method is used with the participation of a large number of listeners and speakers.

Speech intelligibility is understood as the relative or percentage of received (understood) speech elements out of the total number transmitted over the communication channel. The elements of speech are syllables, sounds, words, phrases, numbers. In accordance with them, syllabic, sound, verbal, semantic and numerical intelligibility are set. To measure intelligibility, articulation tables of syllables, sound combinations and words have been developed, taking into account their occurrence in Russian speech.

The term "social adequacy of hearing" refers to the ability of a person to perceive sound stimuli of varying complexity (including speech) and participate in a dialogue. People with a hearing level below "socially adequate" have difficulties in communicating with others, problems at work, at home. The so-called hearing loss stigma makes people hide their communication problem for decades!

The study of auditory function is carried out through two groups of methods: Subjective (psychoacoustic):

- study of hearing by speech with noise;

- study of hearing with tuning forks;

- subjective audiometry.

Objective:

- objective (computer) audiometry;

- acoustic reflexometry;

- tympanometry;

- from acoustic emission;

- unconditioned reflex reactions;

- conditioned responses to sound.

With all subjective methods of hearing research, the subject himself evaluates whether he hears the sound or not and informs the specialist about it in some way.

With objective methods of examination, the results obtained do not depend on the desire of the patient; in most cases, they are recorded using special equipment.

Unfortunately, the results of almost all types of hearing diagnostics, except for a direct assessment of hearing intelligibility by speech, describe the results of the examination in specific terms (decibels, audiograms, curve peak numbers, etc.) and do not provide the patient with objective information about his "real degree of social adequacy ". All this set of scientific terms does not give a person a direct answer: how well, or how badly, he hears and understands the “usual” speech of the interlocutor in the real noise that surrounds him daily. Numerous results of modern high-precision examinations are needed by specialists; a person with a hearing impairment needs to know only one thing, how well he understands the speech of the interlocutor under normal conditions, and further medical attention is required.

The simplest and most accessible method is the study of hearing by speech in a noise signal. The advantages of this method lie in its compliance with the main role of the auditory function in humans - to serve as a means of verbal communication.

In the study of hearing by speech, whispered and loud speech is used. Of course, both of these concepts do not include the exact dosage of the strength and pitch of the sound, however, there are still some indicators that determine the dynamic (power) and frequency response of whispered and loud speech.

In the study of hearing by speech, all speech material is pronounced in reserve air (inhale-exhale-speech). This contributes to equalizing the loudness upon presentation of all speech material in different persons.

An important circumstance in the study of hearing is the "muffling" of the unexplored ear. There are several ways to muffle: put cotton wool with petroleum jelly into the ear canal, insert a finger moistened with water into the ear canal, press the ear tragus into the ear canal, rub the back of the hand covering the ear with the other hand!

The main advantage of the study of hearing by speech is its "physiological comprehensibility" for the subject. The main problematic barriers to widespread adoption are:

- the impossibility of ensuring reproducibility of results for different testers, and for one tester at different times;

- relative duration and complexity of tests;

- the need to perform the study by specially trained medical personnel.

From the existing prior art, various approaches are known for assessing speech recognition. Known software DIRAC (htip://asm-tm.ru/7841-izmerenie-razborchivosti-rechi-v-po-dirac.html), which allows you to evaluate the acoustic environment of the room for recognizable speech inside it. However, this approach is not applicable to testing the auditory abilities of patients.

There is a method for diagnosing the level of hearing (patent RU2467691 C1, 11/27/2012), which uses the speech table of V. Voyachek, which is recorded in the memory of a digital device in the audio format mp3. After the patient repeats the words, the percentage of correctly repeated from the total number of words in the table is determined. In this case, the keyboard of the device is used to record the patient's response at the sound signal volumes of 10, 20 and 30% of the maximum headphone power, respectively. Then, a decrease in the percentage of speech intelligibility is detected. With a decrease in the percentage of speech intelligibility of less than 95%, the presence of hearing loss is judged by any size of headphone power.

The disadvantage of this solution is the lack of accuracy in assessing the intelligibility of speech by the patient, which consists in the lack of automation of the process of recognizing the patient's answers by translating the patient's answers from voice to text format to analyze the correctness of the words heard using artificial intelligence.

With the development of the state of the art in the field of technologies used for diagnostic purposes of various profiles, including for testing the quality of hearing, speech recognition and audiometry, the need to automate all key functions that can be accessed by the end user without the need to resort to services of specialized specialists, as well as simplifying the receipt of a primary picture of the state of the hearing organs, which allows solving the problem of obtaining assistance to the population in regions where there is no possibility of contacting a specialized specialist as such.

The WHO 2021 report (https://www.who.int/ru/news-room/fact-sheets/detail/deafness-and-hearing-loss) notes:

“More than 5% of the world's population, or 430 million people, need rehabilitation to address 'disabling' hearing loss (432 million adults and 34 million children). It is estimated that by 2050 more than 700 million people, or one in ten, will have disabling hearing loss.

"Disabling" is defined as hearing loss in the better ear that exceeds 35 decibels (dB). Nearly 80% of these people live in low- and middle-income countries. Hearing loss is more common among older people: more than 25% of people over the age of 60 suffer from this problem.

Effective interventions to reduce hearing loss at different stages of a person's life include:

Раннее выявление потери слуха и заболеваний уха имеет решающее значение для эффективного ведения пациентов.

Early detection of hearing loss and ear disease is critical to effective patient management.

Для этого необходим систематический скрининг с целью выявления болезней ушей и связанной с ними потери слуха среди следующих категорий людей, подверженных наибольшему риску:

This requires systematic screening to identify ear diseases and associated hearing loss among the following categories of people most at risk:

новорожденные и грудные дети, о дети дошкольного и школьного возраста.

newborns and infants, children of preschool and school age.

люди, подвергающиеся воздействию шума или химических веществ на работе.

people exposed to noise or chemicals at work.

люди, принимающие ототоксичные лекарственные препараты, о люди пожилого возраста»

people taking ototoxic drugs, o elderly people"

WHO experts have proposed a recommended scheme of measures for conducting mass preventive examinations. The procedure is based on sequentially performing a hearing test in three ways: whispered speech, screening air conduction audiometry, or speech intelligibility assessment for recognition of two-digit digits in noise.

Studies of whispered speech require the presence of a specially trained person who is able to provide the same volume of test words for a long time. With mass tests, when the number of subjects is in the region of 30-70 people / session, this is excluded. The results of such testing conducted by different specialists are often difficult to compare.

The complexity of the process is unacceptably high for practical application.

Screening audiometry in a limited frequency range (500 Hz - 4000 Hz) only by air conduction allows assessing only the level of "hearing" in a relatively calm environment, but does not determine the possibility of impaired speech recognition in noise, and also requires the mandatory personal participation of a medical worker.

The most "advanced" method of automated speech recognition in noise, using only two-digit numbers, also has a significant drawback - a significant effect of "recognition" of well-known combinations by the duration of the sound. It is used in the presence of signs of dementia in the elderly.

The participation of a specially trained specialist, the high labor intensity and low information content of the technique for use in subsequent diagnostic tests, which are already carried out by specialized specialists to make a medical diagnosis and develop recommendations for hearing correction, sharply reduce the practical value of the development.

Previously, to solve existing problems in the field of automated hearing assessment and speech recognition, the authors proposed a method and a hardware-software complex (HSC) that implements it, providing a pre-medical assessment of the quality of speech recognition and screening audiometry (RF patent No. 2743049).

This solution formed the basis of the claimed invention and offers a new research principle that uses a dynamically adaptable scenario for conducting multifactorial testing of a patient, which allows you to quickly and efficiently assess the capabilities of a person's auditory analyzer during mass preventive examinations of the population.

SUMMARY OF THE INVENTION

The claimed invention solves a technical problem in terms of providing a fast and efficient method for an automated pre-medical preliminary classifying assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population.

The technical result is to provide an automated pre-medical preliminary classifying assessment of the possibility of a person's auditory analyzer during the behavior of mass preventive examinations of the population.

An additional effect is to increase the reliability of determining the degree of presence or absence of hearing impairment, through the use of a dynamically adaptable scenario when testing a patient.

The claimed technical result is achieved using the method of pre-medical preliminary classifying multifactorial assessment of the possibility of a human auditory analyzer in the behavior of mass preventive examinations of the population, performed using a computing device connected to audio signal playback devices and containing the steps at which:

using a computing device

- form a primary test speech sequence (PTS), which is a sentence consisting of the first number of words based on the matrix test;

- form a noise competing sound for the primary TRP;

- primary TRP is reproduced using audio signal playback devices made in the form of air and bone conduction headphones, while TRP playback is performed simultaneously with the noise competing sound at the first signal-to-noise ratio using speech simulation based on a deep machine learning model;

- receive an oral response from the user;

- perform automatic analysis of the user's oral response to recognize the TRP, by converting it into a text form and analyzing the correctness of the answer using a machine learning model;

and

- based on the analysis of the user's oral responses, a dynamic change in the complexity of the assessment is carried out, in which, as a result of each automatic analysis performed, the number of words in the sentences forming the TRP and / or the signal-to-noise ratio of the reproduced signal is changed;

- performing an assessment of the capabilities of the user's auditory analyzer based on the responses when reproducing the test speech sequence.

In one of the particular embodiments of the method, when analyzing the user's oral responses, the number of correct word matches in the primary TRP is determined.

In another frequent embodiment of the method, the reduction in the number of words in the TRP is performed, provided that the number of correct matches is below a predetermined threshold value.

In another frequent implementation of the method, a volume reduction is performed in the TRP, provided that the number of correct matches is above a predetermined threshold value.

In another frequent embodiment of the method, amplification of the noise signal is performed in the TDP, provided that the number of correct matches is above a set threshold value.

The claimed technical result is also achieved by means of a hardware-software complex (HSC) for a pre-medical preliminary classifying multifactorial assessment of the possibility of a person's auditory analyzer during mass preventive examinations of the population, which contains a computing device and means of reproducing audio signals, in which:

the computing device is configured to

- forming a test speech sequence in the form of sentences consisting of the first number of words with a given volume level and in a given language based on a matrix test that supports the selected language, and the reproduction of the test speech sequence is carried out using speech simulation based on a deep machine learning model;

- selecting a noise competing sound for said test speech sequence of words;

- receiving an oral response from the user;

- automatic analysis of the user's oral response by recognizing each word and / or phrase of the test sequence of words with its transformation into a text form and analyzing the correctness of the answer;

- changes in the generated sentences, consisting of the second number of words, less than the value of the first number, in the case when the number of recognized words in the reproduced sentences is less than a specified threshold value;

- assessing the capabilities of the user's auditory analyzer based on the responses during the reproduction of the test speech sequence;

audio signal playback devices made in the form of air and bone conduction headphones, with the help of which the test speech sequence is reproduced, while the test speech sequence is reproduced in the form of phrases from the matrix test or individual words, accompanied by a competing noise signal.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 illustrates the general layout of an interactive HAC.

Fig. 2A illustrates a flow diagram of the claimed method.

Fig. 2B illustrates a flowchart for adaptively changing test difficulty.

Fig. 3A-3K illustrate an example test protocol.

Fig. 4 illustrates a general view of the computing device.

IMPLEMENTATION OF THE INVENTION

On FIG. 1 shows the general scheme of the interactive PAC (100). It includes a computing device (software), which is a tablet or smartphone running the iOS, Android or Windows operating system. The device (software) provides all the necessary functionality that provides interaction with the user (10) to assess the quality of speech recognition and screening audiometry.

The computing device (software) is manufactured on the basis of standard hardware, with the difference that its audio path undergoes a preliminary procedure of metrological calibration and billing of the loudness scale in dB, in order to comply with existing audiometric standards, for example, to ensure its operation in accordance with GOST R ISO 8253-3-2014 “Acoustics. Methods of audiometric tests".

Computing device (software) contains connected to it via a data transmission channel, a device for inputting speech information (111) and audio playback devices (112, 113).

The voice input device (111) can be in the form of a built-in or external microphone. The external version of the device (111) can be connected using any suitable connection principle, eg USB cable, Lightning connector, Bluetooth connection, etc.

Audio playback devices (112, 113) are two types of air (for example, Sennheiser hd 400s) and bone conduction (for example, Aftershock trekz titanium) headphones that provide the possibility of selective two-channel sound transmission to the user (101) generated by a computing device (110) . The earphones (112, 113) are calibrated with a check, for example using the Tester application, to match the output parameters of the product (volume, frequency) of the audiometric equipment. Calibration parameters must comply with international and / or national standards, for example, GOST R IEC 60645-1-2017. Calibration of sound reproduction devices (112, 113) is carried out using an artificial ear, for example, manufactured by B&K.

The computing device (110) can also be connected via a data network (120), for example, the Internet, to a remote server (130), which can store various information, including settings, user data, update packages data, parameters and information for performing tests, etc.

The computing device (110) generates all the necessary signals, sounds and graphic information for the user testing process (101) to evaluate speech recognition and obtain screening audiometry data.

On FIG. 2A is a description of the process for implementing the method (200) of a pre-classifying multivariate assessment of the capability of the human auditory analyzer using the AAA (100). The use of the claimed PAC allows for a pre-medical objectified and specific assessment of the capabilities of the auditory analyzer of a particular person, while this PAC and the algorithms used in it are fully automated and easy to use.

At the first stage (201), a test speech sequence (hereinafter referred to as TSP) is formed using the device (110), for which the user (101) enters a specialized software application on the device (110). The user (101) selects the language of the test sequence, for example, using the graphical interface of the application, or automatically by speaking a phrase suggested by the application for automatically recognizing the user's language (101). The application used for testing with the device (110) is based on artificial intelligence, in particular one or more machine learning models, for example using an artificial neural network trained to recognize the user's speech (101).

After recognizing the required language for the implementation of the formation of the TRP, the selection of the appropriate matrix test for the given language is performed. These types of tests are standardized and allow you to determine the degree of speech recognition ability in a noise signal (see, for example, Nuesse et al. Measuring Speech Recognition With a Matrix Test Using Synthetic Speech // Trends Hear. 2019 Jan-Dec; 23: 2331216519862982. Published online 2019 Jul 19. doi: 10.1177/2331216519862982). The TRP is formed from sentences consisting of a fixed number of words. As an example, the initial PR is formed from sentences consisting of five words, with the subsequent change in their number based on the responses of users undergoing testing.

Next, in step (202), setting the competing noise sound to be played simultaneously with the matrix test PRT is performed. Before starting the test, you can set the volume of the speech signal and the volume of the noise, the number of words in the test, turn off the left or right channel.

For the first time in the practice of mass screening examinations of the state of hearing, it was proposed to include in the mandatory testing of sound-conducting paths by bone conduction using stereo wireless headphones of bone conduction with a frequency range extended to 20,000 Hz as a transducer, specially calibrated with the tablet used.

The speech sequence uses unique sound (speech and noise) files of our own design. The list of these files can be changed and supplemented depending on the specific goals of the hearing test.

The generated sequence of words and phrases of the matrix test at step (203) is reproduced using a speech synthesizer built on the basis of machine learning models. At the same time, for more accurate testing for each user (101), the appropriate type of voice (female, male, children's, etc.) can be selected, as well as simultaneously testing in several languages, with the possibility of switching them, which is important for people living in a country with several official languages (eg Switzerland).

The speaker's speech is synthesized automatically with the specified volume level and speech rate. The full TRP matrix makes it possible to objectively assess not only the level of speech recognition in noise, but also the cognitive capabilities of a person (101).

Before starting the test, you can set some test modes, in particular:

выбрать проверяемое ухо (оба, правое или левое);

select the ear to be tested (both, right or left);

установить счетчик тестовых фраз;

set the counter of test phrases;

выбрать голос;

choose a voice;

установить соотношение громкости полезного сигнала и шума одним из типичных значений или вручную.

set the volume ratio of the useful signal and noise to one of the typical values or manually.

An important feature of the claimed solution is its implementation in terms of speech reproduction using a software synthesizer with specified parameters (male, female, child, hoarse, whisper, etc.), with a given volume level, speed, intelligibility of pronunciation of phonemes, words, sentences with observance of the intonational features of the given national language.

An important distinguishing feature of PAK (100) is the ability to assess speech intelligibility by bone conduction using headphones of the appropriate type, in particular, bone conduction stereo headphones (113) with an extended frequency range that have passed metrological calibration for compliance with GOST requirements. This makes it possible for the first time to assess the intelligibility of speech in people with diseases of the sound conduction system (otitis media of various forms), by applying speech signals through bone conduction directly to the cochlear system.

The PAC (100) provides a subsystem of the computing device (110) for calibrating audio equipment. The sound paths of the applied devices (110) of different types, different models, different air and bone headphones have different characteristics. In addition, the device (110) provides the ability to adjust the volume of the output signal only in conventional units from 0.0 to 1.0.

In audiology, it is customary to measure the level of an audio signal in decibels (dB). This level depends on the sound pressure level logarithmically. To convert conventional units of the output signal volume of the device (110) into dB, a method for calibrating audio equipment was developed, which must be carried out for each specific model of the device used (110), air (112) and bone headphones (113).

An automatic system is provided to protect the user's auditory system (101) from acoustic overload. The maximum listening level is limited to 95 dB at 1000-3000 Hz, with the ability to immediately mute the sound using the graphical interface of the application to perform the test running on the device (110).

By itself, the need to deliver listening levels in excess of 95 dB means that a person has severe hearing loss, requiring only urgent professional examination by a specialist in the field. This signal information is generated automatically.

Machine learning algorithms used to implement a software speech synthesizer can be, for example, Google Cloud AI & Machine Learning Products Speech-to-Text (https://cloud.google.com/speech-to-text). Google Cloud AI & Machine Learning Products Text-to-Speech (https://cloud.google.com/text-to-speech). Speech processing and analysis cloud services (MDG Cloud - speech synthesis and recognition technologies (speechpro.com)) or any other algorithms that can be trained for the purpose of implementing this technical solution as part of an interactive HSS.

At step (203), the application on the device (110) generates and plays the words that form the sentences of the TRP of the corresponding matrix test using air conduction headphones (112). Upon speaking the words and phrases using the speech synthesizer, the user responds (step 204) by interacting with the device interface (110) or by speaking each word and/or phrase of the test. The user's voice response is captured using a microphone (111) and converted into text format for analysis for the correctness of the spoken phrase. During the test, a list of all answer options as understood by the recognition system can be displayed at the bottom of the device screen (110).

The user's response (101) is evaluated by the device (110) using a machine learning model-based software module that translates the user's response (101) into text and compares it with the rendered word or phrase in the PRT.

The speech sequence is played in parallel with the competing noise sound to more accurately assess the degree of speech intelligibility by the user (101) in the emulated situation. Speech sound files and their corresponding text files can be stored on the computing device (110), which ensures the formation of an arbitrary combination of any speech sound file with any noise file without changing the testing program, as well as supplementing the list of files with arbitrary signals and noise.

After voicing one synthesized sentence, the user (101) repeats the heard sentence as he heard and understood it, after which, at step (205), the beginning of the speech response is automatically detected and compared word by word with the test variant, during which the number of incorrectly pronounced or missing words.

There are several tests of varying degrees of complexity for all age groups:

1. screening assessment of hearing intelligibility in children:

- age group from 7 to 14 years.

2. Balanced tests (two digits)

3. Tables of polysyllabic words containing all the phonemes of the Russian language (Grinberg G.I., Zinder L.R.)

4. Tables of phonemically balanced words (Neyman)

When conducting a speech intelligibility test in noise, it is usually recommended to use white or pink noise as interference, imitating the speech spectrum of the voices of a large group of people standing in large open spaces. In this case, the noise sound can be simulated by selecting from a wide range of different situations most suitable for the situation of the user being tested, for example, a meeting, workshop noises, construction sites, indoor crowd noise, and the like.

PAC (100) allows the user (101) to choose the verification mode in which he experiences the greatest discomfort in his daily life, in particular, the type of voice (male, female, children's, etc.), the volume level of the conversation in generally accepted terms: whispered speech, normal conversation volume, loud conversation in a group. In this case, the device (110) will automatically replace the selected “verbal” loudness description level with the corresponding metrologically confirmed listening intensity level in dB (whispered speech - 35 dB, normal speech - 50 dB, etc.).

When analyzing the responses of the tested users (101), at step (205), the level of their cognitive capabilities is assessed, which is taken into account in the subsequent adaptive tuning of the test at step (206). The evaluation of user responses (101) is carried out by the device (110) according to the international STI scale using the appropriate software logic based on a machine learning model.

As shown in FIG. 2B, a change in the test progress occurs when the recognized responses are calculated, i.e. words, during the playback of the primary TRP. If the user under test (101) scores below the threshold when recognizing words in the AT at the stage (2061), then the device program logic (110) generates a simplified AT (step (2062)), which reduces the number of words in sentences, for example , three instead of 5, or the noise signal becomes quieter and the overall playback signal of the TRP becomes louder. If the user under test (101) at step (2061) correctly recognizes all the words in the primary PR, or a sufficient number to obtain a score exceeding the threshold value, then at step (2063) a sophisticated test scenario is generated, in which the number of words in the generated PR can be increased, or the signal-to-noise ratio becomes louder.

The adaptive adjustment of testing in step (206) also allows the next level of cognitive complexity of the user's audio analyzer to be assessed (101). Also, if the test scenario becomes more complex, the following test suites can be used:

- two-digit numbers;

- a set of phonemically balanced Neumann words;

- a set of test words containing all the phonemes of the Russian language (Grinberg tables).

The above example is for Russian. The English version may use test material: Word recognition performance for Northwestern University Auditory Test No. 6 word. The German language version uses German speech test standards DIN 45626-1-1995, 45621-1-1995, 45621-3-1985, etc.

Consider an example of performing dynamic test adaptation in step (206).

The first test (primary TRP) starts the procedure of playing a shortened phrase test at the “personal conversation” volume level (normal volume level). The level of the competing speech signal is "-10" dB. The test is repeated 5 times with automatic recognition). In accordance with the assessment of the quality of speech perception (Table 1), the result of the initial test is determined.

Next, one of three possible options is automatically considered:

- "excellent" - good hearing is possible. Further tests should go on increasing the complexity of the acoustic and cognitive settings;

- "very bad" - possibly severe hearing loss. Perform further tests with simplification of acoustic and cognitive settings;

- intermediate options from "bad" to "good" are considered as possible correctable hearing impairments, and the most detailed testing is carried out.

Step testing can be carried out further in the following scenario: Step 1

Матричный тест, фразы из трех слов, 52 дБ;

Matrix test, three-word phrases, 52 dB;

Пять фраз;

Five phrases;

Розовый шум, 42 дБ.

Pink noise, 42 dB.

0 correct answers go to step 2

1-4 correct answers go to step 3

5 correct answers - go to step 4

Step 2

Двузначные числа, 52 дБ;

Double digits, 52 dB;

Пять чисел;

Five numbers;

Розовый шум, 42 дБ.

Pink noise, 42 dB.

0 correct answers - go to step 6

1-4 correct answers - go to step 6

Step 3

Матричный тест, фразы из трех слов, 52 дБ;

Matrix test, three-word phrases, 52 dB;

Пять фраз;

Five phrases;

Розовый шум, 42 дБ.

Pink noise, 42 dB.

0-9 correct answers in total with step 1 - go to step 5

10 correct answers - "No significant hearing loss", end of scenario Step 4

Матричный тест, фразы из пяти слов, 47 дБ;

Matrix test, five-word phrases, 47 dB;

Три фразы;

Three phrases;

Розовый шум, 42 дБ.

Pink noise, 42 dB.

1-2 correct answers - go to step 3

3 correct answers - "No hearing loss", end of scenario

Step 5

Тест Неймана, 52 дБ;

Neumann test, 52 dB;

Десять слов;

ten words;

Без шума.

No noise.

0-8 correct answers go to step 6

9-10 correct answers “No significant hearing loss found”, end of scenario Step 6

Тональная пороговая аудиометрия по воздушной проводимости (125-8000 Гц, автомаскировка);

Tone threshold air conduction audiometry (125-8000 Hz, automasking);

Тональная пороговая аудиометрия по костной проводимости (250-8000 Гц, без маскировки).

Tonal threshold audiometry by bone conduction (250-8000 Hz, without masking).

The analysis is carried out in accordance with the scale for assessing the quality of speech communication, which ensures the correct interpretation of speech messages by the listener, based on the standards ISO 9921:2003 * "Ergonomics. Assessment of speech communication" (ISO 9921:2003 "Ergonomics - Assessment of speech communication", IDT) and GOST P ISO 9921-2013.

Also, during testing, the error in the recognition of individual words is also regulated in terms of the percentage of correct answers to word recognition and the calculation of the corresponding score (Table 2).

Table 3 shows regulated and agreed typical speech loudness levels.

Based on the results of the adaptive change in the complexity of testing at stage (206), using the machine learning model at stage (207), users are classified (101), which characterizes the capabilities of the human auditory analyzer (101).

Preliminary classification of users (101) allows you to determine the main three groups in a pre-medical order:

1) "Green Zone" - the participants of this group showed speech intelligibility (phrases and test words) in noise with a score not lower than "good".

Members of this group of specialized medical care at the time of the survey is not required. It is recommended to undergo a second express examination in a year.

2) "Yellow zone" - the participants of this group showed speech intelligibility (phrases and test words) at an average level (insufficient speech adaptation). The possible level of hearing loss at basic speech frequencies by air conduction is more than 25 dB, which means a possible significant hearing loss. For the first time in the automatic version, a joint analysis of audiograms by air and bone conduction was performed to identify the nature of the violation: conductive, sensorineural or mixed type of violation. The analysis was performed for the following ranges:

низкочастотный (125 Гц - 1500 Гц);

low frequency (125 Hz - 1500 Hz);

основные речевые частоты (500 Гц - 4000 Гц);

main speech frequencies (500 Hz - 4000 Hz);

высокочастотный (3000 Гц - 8000 Гц)

high frequency (3000 Hz - 8000 Hz)

Participants in this group should be examined in more detail for subsequent hearing correction and rehabilitation.

Members of this group require a scheduled visit to a specialist to clarify the results (if necessary), make a diagnosis and develop a plan of necessary measures to correct hearing abilities.

3) "Red Zone" - participants in this group showed the level of speech intelligibility (test words) in noise at the level of "bad" and "very bad" and indicators of possible hearing loss during screening audiometry, indicating severe hearing loss or deafness.

These participants need an urgent examination by an otorhinolaryngologist to determine the direction of a more detailed medical diagnostic examination.

When conducting research, with the help of the proposed PAK (100), a general list of the sessions performed is formed in one file, which can be transferred to a PC by mail and loaded into Excel. The line is formed only if all the stages of the session are fully completed.

Each line in the file contains:

Идентификатор вычислительного устройства (110);

Computing device identifier (110);

Дату и время сохранения файла полного протокола;

Date and time when the full protocol file was saved;

Имя файла полного протокола;

Full log file name;

Почтовый индекс места проведения сеанса;

Postal code of the session location;

Пол пациента;

gender of the patient;

Возраст;

Age;

Группа срочности (зеленая, желтая или красная);

Urgency group (green, yellow or red);

Разборчивость фраз;

intelligibility of phrases;

Разборчивость тестовых слов;

legibility of test words;

Скрининговая аудиометрия:

Screening audiometry:

Степень снижения слуха по воздушной проводимости для правого и левого уха из списка:

The degree of hearing loss by air conduction for the right and left ear from the list:

Нет данных;

There is no data;

Норма;

Norm;

1 степень;

1 degree;

2 степень;

2 degree;

3 степень;

3 degree;

4 степень;

4 degree;

Глухота;

Deafness;

Вид тугоухости по речевым частотам для правого и левого уха из списка:

Type of hearing loss by speech frequencies for the right and left ear from the list:

Нет данных;

There is no data;

Норма;

Norm;

Кондуктивная;

Conductive;

Нейросенсорная;

Neurosensory;

Смешанная.

Mixed.

Linking the results to a geographic location (zip code, address or navigator coordinates) allows you to combine the results obtained in neighboring regions for further analytical processing.

If a representative sample is available, it is possible to predict the personnel and financial needs both by region and the country as a whole, and evaluate the effectiveness of the organizational and technical measures taken to improve the communication capabilities of the population of specific regions.

In accordance with the WHO international statistics on the prevalence of hearing loss, at least 75-80% of test examination participants can be expected to fall into the "green zone", which will significantly reduce the burden on existing specialists.

The claimed solution makes it possible to realize the practical possibility of a comparative analysis of the results of assessing speech intelligibility in two physiologically different ways of transmitting an audio signal to the cochlea: through air and bone conduction.

The difference in the results of speech audiometry for air and bone conduction may allow differentiating conductive and sensorineural hearing loss without resorting to complex audiometric studies.

Taking into account the “qualifying” specifics of the examination being carried out - a pre-medical method without the involvement of specialists in audiology and otorhinolaryngology, it is extremely important to ensure an adequate level of reliability of data on measuring the threshold values of audibility of test signals. Simultaneously with all tests, the level of ambient noise is automatically checked and recorded.

In this solution, an algorithm for the semantic analysis of audiograms that have passed a preliminary analysis for reliability and consistency has been developed and implemented, which, together with the results of the analysis of speech intelligibility tests, made it possible to more reliably attribute each subject to one of three classes: "green", "yellow", "red". ".

Based on the results of the testing, a detailed protocol of testing conditions, descriptions of the test materials used and real answers of patients with automatic assessment of the correctness of the answer using machine learning algorithms are automatically prepared for each patient. An example protocol is shown in Fig. 3A-3K. A set of digital protocols generated for the entire array of examined patients can be sent, for example, to the regional telemedicine system or another organization.

On FIG. 4 shows a general example of a computing device (300) that can be used to implement the devices included in the HAC (100), such as a computing device (110). In general, the device (300) contains components such as: one or more processors (301), at least one random access memory (302), persistent data storage (303), input/output interfaces (304), I/O At (305), networking tools (306).

The processor (301) of the device performs the basic computing operations necessary for the operation of the device (300) or the functionality of one or more of its components. The processor (301) executes the necessary machine-readable instructions contained in the main memory (302).

The memory (302) is typically in the form of RAM and contains the necessary software logic to provide the required functionality. The data storage means (303) can be in the form of HDD, SSD disks, raid array, network storage, flash memory, optical storage media (CD, DVD, MD, Blue-Ray disks), etc. The tool (303) allows long-term storage of various types of information, such as request processing history (logs), user IDs, audio files, and the like.

Interfaces (304) are standard means for connecting and operating various types of devices (300), such as USB, RS232, RJ45, LPT, COM, HDMI, PS/2, Lightning, FireWire, etc. The choice of interfaces (304) depends on the specific implementation of the device (300), which can be a personal computer, mainframe, server cluster, thin client, smartphone, laptop, and the like.

As means of I/O data (305) can be used: a keyboard, a joystick, a display (touchscreen), a projector, a touchpad, a mouse, a trackball, a light pen, speakers, a microphone, and the like.

Means of networking (306) are selected from a device that provides network data reception and transmission, for example, an Ethernet card, WLAN/Wi-Fi module, Bluetooth module, BLE module, NFC module, IrDa, RFID module, GSM modem, etc. With the help of the tool (306) the organization of data exchange over a wired or wireless data transmission channel, for example, WAN, PAN, LAN (LAN), Intranet, Internet, WLAN, WMAN or GSM, is provided.

The components of the device (300) are typically connected via a common data bus or any other type of communication that allows the elements of the device (300) to interact.

In these application materials, a preferred disclosure of the implementation of the claimed technical solution was presented, which should not be used as limiting other, private embodiments of its implementation, which do not go beyond the scope of the requested legal protection and are obvious to specialists in the relevant field of technology.

Claims (23)

1. A method for pre-medical preliminary classifying multifactorial assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population, performed using a computing device connected to audio signal playback devices, and comprising the steps of: using a computing device - form a primary test speech sequence (PTS), which is a sentence consisting of the first number of words based on the matrix test; - form a noise competing sound for the primary TRP; - primary TRP is reproduced using audio signal playback devices made in the form of air and bone conduction headphones, while TRP playback is performed simultaneously with the noise competing sound at the first signal-to-noise ratio using speech simulation based on a deep machine learning model; - receive an oral response from the user; - perform automatic analysis of the user's oral response to recognize the TRP by converting it into text form and analyzing the correctness of the answer using a machine learning model; and - based on the analysis of the user's oral responses, a dynamic change in the complexity of the assessment is carried out, in which, as a result of each automatic analysis performed, the number of words in the sentences forming the TRP and / or the signal-to-noise ratio of the reproduced signal is changed; - perform an assessment of the capabilities of the user's auditory analyzer based on the responses when reproducing the test speech sequence. 2. The method according to claim 1, characterized in that when analyzing the user's oral responses, the number of correct word matches in the primary TRP is determined. 3. The method according to claim 2, characterized in that the reduction in the number of words in the TRP is performed, provided that the number of correct matches is below a predetermined threshold value. 4. The method according to claim 2, characterized in that the volume is reduced in the TRP, provided that the number of correct matches is above a predetermined threshold value. 5. The method according to claim 2, characterized in that the amplification of the noise signal in the TRP is performed, provided that the number of correct matches is higher than the set threshold value. 6. Hardware-software complex (HSC) for pre-medical preliminary classifying multifactorial assessment of the possibility of a human auditory analyzer during mass preventive examinations of the population, containing a computing device and means of reproducing audio signals, in which: the computing device is configured to - forming a test speech sequence in the form of sentences consisting of the first number of words with a given volume level and in a given language based on a matrix test that supports the selected language, and the reproduction of the test speech sequence is carried out using speech simulation based on a deep machine learning model; - selecting a noise competing sound for said test speech sequence of words; - receiving an oral response from the user; - automatic analysis of the user's oral response by recognizing each word and / or phrase of the test sequence of words with its transformation into a text form and analyzing the correctness of the answer; - changes in the generated sentences, consisting of the second number of words, less than the value of the first number, in the case when the number of recognized words in the reproduced sentences is less than a specified threshold value; - assessment of the capabilities of the user's auditory analyzer based on the responses during the reproduction of the test speech sequence; audio signal playback devices made in the form of air and bone conduction headphones, with the help of which the test speech sequence is reproduced, while the test speech sequence is reproduced in the form of phrases from the matrix test or individual words, accompanied by a competing noise signal.

Images