RU2722108C1 - Method for objective screening for detection of hearing disorders (method for accelerated assessment of auditory function in infants) - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention refers to medicine, particularly to otorhinolaryngology and audiology, and can be used in population survey, especially of young children, for detecting problems with hearing organs. Method for accelerated assessment of auditory function in patients (especially infants) involves recording of short-latent acoustical evoked potentials (SLAEP) with presentation of stimuli of different frequency and intensity, wherein SLAEP registration criterion is V peak imaging with minimum level of stimulus. Frequency-specific stimuli are located in different frequency intervals covering the whole investigated range of hearing, and their number is usually from 2 to 5. Stimuli are presented in the order defined by a sequence of maximum length (MLS) (for each frequency range) containing to cycles of duration of not less than 6 ms, where k equals 63, 127 or 255. Stimuli in each frequency sequence are presented simultaneously, and the corresponding sequences are shifted such that the distance between the stimuli for the neighboring frequency bands is not less than 2.5 ms, and for any other frequency bands of not less than 0.8 ms, then the recorded SLAEP signal is averaged over the entire length of the MLS and folded for all frequency intervals.

EFFECT: method enables reducing the time of diagnosis, increasing the accuracy of detecting the presence and latency of the peaks and, as a result, improving reliability and accuracy of diagnosis.

3 cl, 2 ex, 3 dwg

Description

The invention relates to medicine, in particular to otorhinolaryngology and audiology, and can be used to examine the population, especially young children, to identify problems with hearing organs.

The introduction into clinical practice of objective electrophysiological methods for determining auditory function can significantly improve the diagnosis of patients with hearing impairment of various origins. Often, such methods are the only option for studying hearing in children of the neonatal, infant and early childhood years.

The methodology for recording short-latency auditory evoked potentials (CSWP) is basic in the diagnosis of hearing impairment in children (Tavartkiladze G.A. Clinical Audiology Guide - Moscow: Medicine, 2013, 676 pp., Ill.), The second one (diagnostic) is based on it stage of universal audiological screening.

The CSWP signal is the electrical response of the auditory nerve cells, as measured by surface electrodes located on the scalp in response to an acoustic stimulus. This signal is very small, usually it does not exceed 0.4 μV, despite the fact that in an awake patient the amplitude of the background electrophysiological activity in the frequency range of VSWR can be about 10 μV. In a sleeping patient, the amplitude of background activity is noticeably smaller, however, to isolate the signal, one still has to use averaging of several thousand realizations, which significantly slows down the study.

In the study of hearing by registering HEPP in most cases, acoustic click stimulation is used (Levin S.V. Use of auditory evoked potentials in modern audiological research: Abstract of thesis. Candidate of Medical Sciences. - St. Petersburg, 2009). This technique is easy to implement, well studied, but has several drawbacks. When assessing thresholds for auditory perception by CWP, obtained in response to a click, a threshold estimate is obtained for the frequencies that are least affected by the pathology. One can only guess about the thresholds in the region of the most affected frequencies by indirect signs. Therefore, it is advisable to use a methodology for assessing auditory perception thresholds in at least octave frequency bands.

To obtain this estimate, the registration of the VSWR is performed in response to a tonal premise, which is a fragment of a pure tone with one or another envelope. The choice of the optimal envelope depends on the required resolution in frequency and threshold and is the subject of discussion. In particular, tonal premises with changing instantaneous tone frequencies, the so-called narrow-band “chirp” (Elberling, C .; Don, M. Auditory brainstem responses to a chirp stimulus designed from derived-band latencies in normal-hearing subjects J, have found application Acoust. Soc. Am. 124, 5, 2008; 3022-3037.), Providing a slightly larger amplitude and a clear response at near-threshold levels of stimulation. However, such stimulation does not solve the problem of the duration of the study, since it only slightly increases the amplitude of the response compared to the response to the filtered click. At the same time, there is a significant scatter of estimates of the effectiveness of this method in the literature, which may be associated with difficulties in selecting optimal tonal stimuli.

An important role in the interpretation of the survey results is played by the assessment of the latency of the peaks of the VSWR. However, the latency measured during click stimulation depends on several factors at once. The measured potential is the sum of the potentials excited by different frequencies of the signal spectrum. Due to delays in the propagation of the acoustic signal along the basilar membrane of the organ of Corti, receptor cells responsible for different frequencies do not generate a nerve response at the same time, and the delay increases as it approaches low frequencies. As a result, the peaks of potential obtained from the low-frequency regions turn out to be blurry, and, as a rule, fall out of the analysis. Visually, only the latency of the most high-frequency region, which gives an answer at a given intensity, is observed. Thus, the measured latency depends both on the propagation delays of the signal along the neurons of the auditory nerve, and on the profile of the dependence of the threshold of auditory perception on frequency. There are various methods for obtaining auditory evoked potentials from individual areas of the cochlea (Burkard R.F., Don M., Eggermont J.J. Auditory Evoked Potentials: Basic Principles and Clinical Application / Lippincott Williams & Wilkins, 2007). However, their implementation requires such time costs and accuracy that they are used exclusively in scientific research.

In 1981, Eysholdt and Schneiner (Eysholdt U., Schreiner C. Maximum length sequences - a fast method for measuring brain-stem-evoked responses Audiology 1982; 21: 242-250.) Proposed using maximal sequence stimulation for recording the VSWR ( PMD). The use of PMD made it possible to increase the frequency of presentation of stimuli many times, preserving the response form with some accuracy. However, further studies have shown (Picton TW, Champagne SC, Kellett AJ Human auditory evoked potentials recorded using maximum length sequences. Electroencephalogr Clin Neurophysiol 1992; 84: 90-100) that with an increase in the frequency of presentation, the amplitude of the response begins to decrease rapidly, and artifacts that appear reduce its diagnostic value.

RF patent No. 2481788 proposes a method for conducting a study to assess auditory function in young children, which includes recording short-latency auditory evoked potentials (VSWR) upon presentation as a stimulus of broadband sound clicks delivered through the earphone during natural or medical sleep. The frequency of presentation of stimuli is 21 per second, the lower limit of the amplifier bandwidth is 100 Hz, the upper limit is 3000 Hz, the analysis window is 20 ms, and the number of averagings is 2000. According to the invention, the study is carried out in two stages, and at the first stage, the registration of the VSWR is carried out by upward method with successive twofold presentation of stimuli of each loudness - 20 dB, 40 dB, 60 dB, 80 dB, 90 dB, 103 dB, after which at the second stage they perform registration of the VSWR upon presentation as a stimulus modulated in amplitude and frequency - “chirp” a signal with a frequency of 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz with the same stimulation parameters, while at both stages of the study, the criterion for recording the VSWR is visualization of the V peak with a minimum stimulus level. The disadvantages of the method include the duration of the examination (the presence of two stages and the need to set sufficient statistics), which limits its clinical use, as well as insufficient accuracy, especially with a low level of incentives.

The specified patent can be selected as the closest analogue to the claimed invention. The present invention is directed to the formation of such a sequence of stimuli that would reduce accumulation time and, due to a larger number of responses, improve the signal-to-noise ratio.

The technical result from the use of the claimed invention is: reduction of diagnostic time; increasing the accuracy of determining the presence and latency of peaks and, as a result, improving the reliability and accuracy of diagnostics.

The claimed technical result is achieved by the fact that the method of accelerated assessment of auditory function in patients (especially young children) involves recording short-latency auditory evoked potentials (VSWR) upon presentation of stimuli of various frequencies and intensities, while the criterion for registering VSWR is visualization of the V peak with a minimum level of stimulus . Frequency-specific stimuli are located in different frequency intervals (covering the entire studied range of hearing), and their number is usually from 2 to 5. The stimuli of each frequency interval are presented in the order determined by the maximum length sequence (PMD), containing at least 25-second measures 6 ms, where k is 63, 127 or 255. The stimulus sequences of all frequency intervals are presented simultaneously, but are shifted in such a way that the distance between the stimuli for adjacent frequency intervals is not less than 2.5 ms, and for any other frequency intervals - not less than 0 , 8 ms, then the recorded VSWR signal is averaged over the entire length of the PMD, after which responses to stimuli of each frequency interval are extracted from it.

It is preferable to use octave frequency bands as frequency intervals, and Chirp as frequency-specific stimuli.

The maximum length sequences (PMD, M-sequences or MLS in the English literature) are called pseudorandom binary, i.e. consisting of zeros and ones, sequences of length N = 2 ^m -1, which can be obtained on a binary shift register of length m with appropriately wound feedbacks. Stimulation and accumulation of data are performed with a constant period of time, conditionally divided into measures of equal duration. The number of measures during the stimulation period is equal to the length of the sequence, and the duration of the measure can be chosen less than the response length. The stimulus is transmitted to the object under study only at certain measures determined by PMD.

To obtain an answer to a single stimulus, it is necessary to cyclically shift the response of the studied system by the value of the cycle duration as many times as there are cycles in the sequence, and add the shifted answers with a sign determined by the sequence used. Averaging the response to individual periods of stimulation can be performed both before calculating the response to a single stimulus, and after calculation. Problems in the use of PMD are that with a decrease in the interval between stimuli, the magnitude of the response to the stimulus begins to decrease, and the form of the response begins to change, which significantly reduces both the amplitude of the restored response and its diagnostic significance. The maximum increase in amplitude at the optimum frequency is not so great, and the shape distortion is too noticeable (Picton TW, Champagne SC, Kellett AJ Human auditory evoked potentials recorded using maximum length sequences. Electroencephalogr Clin Neurophysiol 1992; 84: 90-100, Burkhard R., Shi Y. Hecox KE Brainstem auditory evoked responses elicited by maximum length sequences: Effects of simultaneous masking noise. // J Acoust Soc Am 1990; 87: 1656-1664, Thornton ARD, Slaven A. Auditory brainstem responses recorded at fast stimulation rates using maximum length sequences. // Br J Audiol 1993; 27: 205-210).

The proposed method is based on the fact that the weakening and distortion of the response appears locally and independently for each area of the organ of Corti. The response to a narrow-band stimulus is much less distorted by stimuli that are spaced apart in frequency, which means that these stimuli can be arranged in a stimulating sequence much closer in time without significant distortion of the response form. The inventors constructed a mixture of stimuli in such a way that distortion of the response form to the stimulus as a result of exposure to both sufficiently far stimuli of the same frequency range and stimuli of other frequency ranges remains acceptable for the purposes of examining patients. Since the optimal interstimulus intervals turn out to be significantly less than the duration of the analyzed response, PMD stimulation is used to isolate the responses to each type of stimulus. Obtained during the implementation of the proposed method, the answers are used as well as in the prototype (RF patent No. 2481788). Further treatment and rehabilitation is carried out in accordance with the diagnosed degree of hearing loss.

The essence of the invention is illustrated by the following figures:

In FIG. 1 presents the principle of the simultaneous registration of several responses of the HACP;

In FIG. Figure 2 shows the dependence of the peak amplitude V on the time shift within the clock cycle between two PMDs;

In FIG. Figure 3 shows the dependences of frequency-specific potentials simultaneously recorded for each volume level.

EXAMPLES OF IMPLEMENTATION

Example 1

The capabilities of the claimed invention were demonstrated in a series of experiments in which volunteers aged 20 to 58 years participated with hearing within the age norm. The goal was to select the optimal interstimulus intervals and to demonstrate the possibility of simultaneously registering five VSWPs in normally hearing adults.

In FIG. 1 shows the principle of registering multiple responses using two sequences of 7 measures in length. The sequence period consists of measures of equal duration. The upper three curves conditionally show the simulating sequence, the answer to it, and the restored answer. On the ordinate axis - signal values in arbitrary units, on the abscissa axis - time. The bottom three curves show the same operations for a mixture of two PMD with a different type of stimulus. Part of the measures determined by the PMD contains a stimulus, the remaining measures are empty. The response to such a sequence of stimuli contains responses to individual stimuli, which, as a rule, overlap. The procedure for reconstructing the response consists in cyclic shifts of data averaged over the length of the period by the value of the clock cycle and their addition with one or another sign determined by the recovery sequence. As a result, it is possible to restore the response form even if individual responses overlap.

If we add another sequence by cyclically shifting it by a certain amount, then when processing the signal, the responses to individual sequences will be restored in different places of the analysis era, determined by the magnitude of the introduced shift. If individual answers have different amplitudes, then not only does the amplitude of the reconstructed response decrease, but artifacts also appear that are shifted by a multiple of a tact measure and distort the shape of the result. In this example, the interstimulus interval of PMD is chosen large enough so that masking by stimuli of the same sequence is negligible. Stimuli of different sequences can be located much closer to each other, while the calculation is made on the fact that stimuli of different frequency ranges to a lesser extent affect their responses.

In experiments with volunteers, tonal stimuli were formed as a result of digital filtering of a click in the forward and reverse directions with Butterworth filters of the 7th (LF) and 5th (HF) orders so that the signal spectrum was flat within one octave at zero phase shifts.

The central frequencies of the stimuli were 0.5; 1; 2; 4 and 8 kHz. The stimulus amplitude of each frequency in the mixture was adjusted to +6, 0, -6, -12 and -18 dB, respectively. The level of the mixture intensity was considered the excess over the threshold of auditory perception of tonal stimuli averaged over the participants of the experiment with a frequency of 1 and 2 kHz, adjusted for the results of the tonal threshold audiometry of this participant.

We used PMD with a length of 63 cycles with a minimum interstimulus interval of 8 ms. For comparison, in a number of experiments, stimulation was performed by clicking or tone sending with a constant inter-stimulus interval of 24 ms.

The stimulus sequences calculated on a computer were loaded into the memory of the signal processor of the Tucker & Davis System 3 electrophysiological complex and presented via the EARtone 3A telephone. The amplitude was adjusted by the PA5 hardware attenuator. Potentials were recorded in the abduction from the forehead and mastoid process by cup electrodes, amplified by the biological amplifier Medusa of the System 3 complex and transferred to a computer. All data was saved without averaging to files for further processing. During processing, the signal in the frequency range 100 Hz – 3 kHz was extracted by 3rd order Butterworth filters, then a weighted average of the responses was performed. To estimate noise, averaging and further processing were performed alternately in two buffers, each with a length of one PMD period. Next, the responses were restored by convolution with the recovery sequence and the responses to individual stimuli were sampled from the corresponding sections of the buffer.

Three series of experiments were conducted. At first, the VSWPs recorded during stimulation by tone transmission with a constant frequency and in PMD mode were compared. It was shown that under the selected stimulation conditions, the waveform in the two modes differs slightly. Next, we studied the dependence of the distortion of the response on the time shift between the sequences upon stimulation with two PMDs with central stimulus frequencies differing by two octaves. In the last series, CVRs were recorded per click with a constant frequency of presentation and to a mixture of 5 PMD. To reduce the impact on the result of changes in the state of the subject, all experiments were performed in blocks, in each of which all the compared types of stimulation, differing in the use of PMD or shift, were applied in turn. Each species was presented for about 30 s, while the data obtained within 2 s after switching to a new type of stimulation were not used. The duration of the experiment ranged from 90 to 150 minutes, during which time blocks with different levels of stimulation were presented several times in turn. Before the experiment, an ENT examination of the subject was carried out, impedanometry and tonal threshold audiometry. In the last series of experiments, the threshold of subjective perception of the tonal stimulus was also determined when it was presented with a constant frequency immediately before the registration of CVS. All experiments were carried out by the program, operator participation was reduced only to the installation of electrodes and a telephone. Experiments on the mutual influence of two PMDs with central frequencies of stimuli differing by two octaves demonstrated that the maximum change in both responses occurs when the shift between sequences within a measure is close to zero, i.e. when separate stimuli of two sequences are presented simultaneously. In this case, the greatest mutual influence was observed between stimuli with a frequency of 1 kHz and 4 kHz. In FIG. Figure 2 shows an example of the effect of a shift between these stimuli on the amplitude V of the peak of the VSWR. In this case, we track the change in the amplitude of the response to one of the sequences of stimuli, calling the second PMD a masker.

On the whole, a compromise delay value of 1 ms was made based on these data, while a higher-frequency stimulus followed a lower-frequency one. Thus, shifts within the cycle for stimuli with a frequency of 0.5; 1; 2; 4 and 8 kHz were 0, 4, 1, 5, and 2 ms, respectively. A scheme of 0, 5, 2, 7, and 4 ms is also possible, in which the time shift between stimuli of neighboring ranges is 3 ms, and between stimuli separated by the same range is 2 ms.

In FIG. Figure 3 shows the results of registration of HEPC at several levels of stimulation in one experiment in a subject with normal hearing. The left-hand column shows the RER obtained in response to a click when presented with a constant frequency, and in the other five - responses to 5 frequency-specific stimuli, and for each volume level shown on the left, all five responses were recorded simultaneously. All data were obtained in one experiment lasting 2 hours. At the same time, it took about 12 minutes to get an answer both to a click of each intensity (left) and to 5 tonal premises presented in the general mixture (each row in 5 panels on the right). Thus, the method allows the registration of five frequency-specific potentials in a time comparable to the registration of the response to a click. (The ability to simultaneously register 5 responses allows in most cases to reduce the total examination time by about 4 times). The diagnostic significance of such responses is much higher than that of a response to a click. In addition to the possibility of assessing the dependence of the auditory perception threshold on frequency, the accuracy of peak latency assessment is significantly increased, and, accordingly, the reliability of differential diagnostics based on latency measurements is increased. At the same time, it remains possible to calculate the response to a click with some accuracy in order to compare it with other measurements.

Example 2

The inventive method was tested on simulating models with the following parameters:

The stimuli were presented in the order determined by the sequence of maximum length, containing 63 measures with a duration of 8.1 ms. Chirp type stimuli were used with a frequency of 500 Hz, 1 kHz, 2 kHz, 4 kHz and 8 kHz. The stimuli in the mixture were presented simultaneously, and the sequences corresponding to them were cyclically shifted by 0; 12.625; 20.25; 26.875 and 6.5 cycles, respectively. After averaging over the length of the sequence, the response was restored by 62-fold cyclic shift of the response by the tact duration and summing the shifted responses with a sign determined by the used PMD. After that, the responses to individual incentives that were 0 from the beginning of the period were selected from the sum; 12.625; 20.25; 26.875 and 6.5 cycles, respectively.

The proposed method can be implemented on existing systems for recording auditory evoked potentials that can transmit an arbitrary stimulus of duration during the accumulation period. For this, a mixture of stimuli is loaded into the stimulator buffer, the response to the first stimulation period is excluded from the accumulation, and the average response is processed in order to single out a single response, after which fragments are separated from the beginning by the same amount by which they were cyclically shifted during synthesis a stimulating mixture of a sequence of corresponding stimuli.

Claims (3)

1. A method of examining patients to assess their auditory function, including recording short-latency auditory evoked potentials (VSWR) upon presentation of stimuli of various frequencies and intensities, while the criterion for registering VSWR is visualization of the V peak at a minimum level of stimulus, characterized in that it is frequency-specific stimuli located in different frequency intervals that cover the entire studied frequency range, and the number of which is from 2 to 5, are presented in the order determined by the maximum length sequence (PMD), containing k cycles with a duration of at least 6 ms, where k is 63, 127 or 255, while the sequences of stimuli of all frequency intervals are presented simultaneously, and are shifted in such a way that the distance between the stimuli for adjacent frequency intervals is not less than 2.5 ms, and for any other frequency intervals not less than 0.8 ms, then the recorded signal averaged over the entire length of the PMD and warehouses for all frequency intervals, after which the response is restored by k-1-fold cyclic shift of the response by the duration of the tact and summing the shifted responses with a sign determined by the used sequence of maximum length, after which the responses to individual stimuli that are separated from the beginning of the period are selected from the sum by the amount by which the sequences of the corresponding stimuli were cyclically shifted. 2. The method of conducting the survey according to claim 1, characterized in that octave frequency bands are taken as frequency intervals. 3. The method of conducting the examination according to claim 1, characterized in that the use of frequency-specific stimuli such as chirp (Chirp).

Images