Method and system for collecting and analyzing auditory electrophysiological data of experimental animal

Abstract

The invention belongs to the field of biological medicine research and development, discloses a method and a system for collecting and analyzing auditory electrophysiological data of experimental animals, and is particularly suitable for non-human primates. The method comprises the following steps: delivering a drug to a middle ear cavity through an external auditory canal, and synchronously recording auditory electrophysiological data before and after drug administration; aBR and DPOAE multi-mode detection is adopted, and the frequency range of 4 kHz to 32 kHz is covered; setting anesthesia depth and three-electrode positioning to reduce interference; optimizing the signal-to-noise ratio through noise suppression and dynamic signal superposition; an auditory threshold database is constructed, and a machine learning model is combined to judge an injured part. The system integrates a multi-modal stimulation module, a minimally invasive drug delivery device and a data analysis platform, and supports cross-species data compatibility. The accuracy of otology medicine research is remarkably improved, the surgical operation risk is avoided, and the method is suitable for gene therapy, ototoxicity screening and other scenes.

Description

Technical Field

[0001] This invention belongs to the field of biomedical research and development, and specifically discloses a method and system for collecting and analyzing auditory electrophysiological data of experimental animals. Background Technology

[0002] In recent years, with the rapid development of treatment technologies for ear diseases (such as gene therapy and stem cell therapy), accurate assessment of auditory function has become a core requirement for preclinical research. Auditory electrophysiological testing techniques (such as auditory brainstem evoked potentials (ABR) and distortion product otoacoustic emissions (DPOAE)) are widely used to assess cochlear hair cell function, auditory nerve conduction, and the integrity of central pathways due to their non-invasiveness and high sensitivity. However, the application of existing technologies in experimental animal models, especially in non-human primates (such as cynomolgus monkeys), still has significant limitations, specifically in the following three aspects:

[0003] 1. Detection bias due to insufficient species suitability

[0004] Traditional auditory electrophysiological studies are mostly based on rodents (such as SD rats and C57 mice), but their ear anatomy differs significantly from that of humans. For example, the thickness of the round window membrane in rodents (approximately 10-20 μm in mice) is much smaller than that in humans (approximately 70-100 μm), resulting in a huge difference in drug penetration efficiency. In addition, their auditory frequency range (mice: 1-100 kHz, rats: 0.5-80 kHz) differs significantly from that of humans (0.02-20 kHz) in the high-frequency range, making it difficult to simulate the pathological characteristics of human hearing loss. Cynomolgus monkeys, as the non-human primate model most closely related to humans, have a cochlear structure and hearing range (0.125-45 kHz) that are closer to humans, but current technologies lack adaptation schemes tailored to their anatomical characteristics. For example, ABR detection often uses the short pure frequency range of rodents (such as 4-32kHz), which fails to cover the low-frequency sensitive area of ​​cynomolgus monkeys (<4kHz); while the sound intensity gradient in DPOAE detection (usually decreasing by 10dB) does not take into account the sound wave attenuation characteristics of the external auditory canal of cynomolgus monkeys, resulting in threshold determination error.

[0005] 2. Low data integration efficiency and lack of automation

[0006] Current auditory electrophysiology studies typically analyze ABR and DPOAE data independently, lacking correlation models for multimodal data. For example, ABR is mainly used to assess the integrity of the auditory nerve-brainstem pathway, while DPOAE reflects the function of cochlear outer hair cells; the synergistic effect of the two in the mechanism of ototoxic drug damage has not yet been quantitatively standardized. Furthermore, traditional analysis methods rely on manual interpretation of waveform features (such as latency and amplitude), which is inefficient and highly subjective. Although machine learning algorithms (such as support vector machines) have been attempted to be applied to ABR waveform classification, their training data is mostly derived from rodents and cannot be directly transferred to the complex auditory pathways of cynomolgus monkeys.

[0007] 3. Technological disconnect between drug delivery and efficacy assessment

[0008] Local ototherapy techniques (such as intratympanic injection and round window membrane administration) are crucial for otomedicine research, but existing methods face two major bottlenecks in their application to cynomolgus monkeys: First, invasive procedures (such as temporal bone fenestration) require specialized skills and are prone to causing inner ear damage, limiting their standardized promotion; second, the temporal separation between drug administration and electrophysiological monitoring leads to delayed efficacy assessment. For example, gene therapy drugs have a duration of action of several weeks, while traditional ABR / DPOAE testing can only sample at fixed time points, failing to correlate drug concentration with changes in auditory function in real time.

[0009] To address the aforementioned issues, there is an urgent need to develop an auditory electrophysiological data acquisition and analysis system adapted to the anatomical and physiological characteristics of cynomolgus monkeys, aiming to achieve the following objectives through technological innovation:

[0010] Simplified operation: Using a simple drug delivery device, avoiding surgery, allowing ordinary laboratory personnel to perform accurate drug delivery after a short period of training;

[0011] Data fusion: Integrating multimodal data from ABR and DPOAE, and combining them with machine learning algorithms to build an automated diagnostic model for hearing impairment;

[0012] Dynamic monitoring: Enables synchronous closed-loop control of drug administration and electrophysiological monitoring, providing real-time data support for evaluating the efficacy of otological drugs. Summary of the Invention

[0013] To address the aforementioned problems in the prior art, this invention provides a method and system for acquiring and analyzing auditory electrophysiological data in experimental animals, particularly for preclinical studies of ototoxic drugs in non-human primates (such as cynomolgus monkeys). By optimizing drug delivery techniques, multimodal auditory testing (ABR / DPOAE), and automated data analysis, it solves problems such as poor species suitability and low data integration efficiency in traditional methods, providing efficient technical support for ototoxic drug screening and gene therapy efficacy evaluation.

[0014] This invention includes the following technical solutions:

[0015] A method for collecting and analyzing auditory electrophysiological data of experimental animals includes the following steps:

[0016] (a) Using a drug delivery device, drugs are delivered into the middle ear cavity of non-human primates through the external auditory canal, and auditory electrophysiological data are recorded simultaneously before and after drug delivery;

[0017] (b) Auditory pathway signals of non-human primates were collected by multimodal detection of auditory brainstem evoked potentials (ABR) and distortion product otoacoustic emissions (DPOAE). The ABR detection used short pure tone stimulation with a frequency range of 4 kHz to 32 kHz and the stimulation intensity decreased by 5 dB. The DPOAE detection used two tone stimulation with a frequency ratio of f2 / f1 = 1.2 and the sound intensity range covered 20 dB to 80 dB SPL.

[0018] (c) Based on the physiological characteristics of non-human primates, the depth of anesthesia was set to a respiratory rate of 40-60 breaths / minute, and the motion artifact interference was reduced by using a three-electrode positioning method at the top of the head, behind the ipsilateral auricle, and behind the contralateral auricle.

[0019] (d) An adaptive noise suppression algorithm is adopted to eliminate interference signals with environmental noise >30dB SPL in real time, and the signal-to-noise ratio is optimized by dynamically adjusting the number of signal superpositions (500-1000 times).

[0020] (e) By combining the latency, amplitude and DPOAE response amplitude of the ABR waveform, a non-human primate-specific auditory threshold database is constructed to determine the location of auditory damage and output the probability of abnormalities in the cochlea, auditory nerve or central pathway.

[0021] Furthermore, in the above-mentioned method for collecting and analyzing auditory electrophysiological data of experimental animals, in step (a), the drug delivery device is equipped with a visualization light source and a magnifying glass module. The operator observes the upper quadrant of the tympanic membrane in real time through an otoscope. The upper limit of the injection volume is 50 μL. The injection is slow, and the injection pressure is controlled to be ≤50 kPa by a pressure feedback sensor.

[0022] Furthermore, in the above-mentioned method for collecting and analyzing auditory electrophysiological data of experimental animals, in step (b), the ABR detection targets the auditory nerve to brainstem pathway of cynomolgus monkeys, using the latency difference threshold method. When the peak delay of wave I to wave V exceeds 0.2ms, it is determined to be auditory nerve damage, and when the amplitude of wave V decreases by more than 50%, it is determined to be central pathway abnormality.

[0023] This invention also discloses a system for acquiring and analyzing auditory electrophysiological data in experimental animals, comprising:

[0024] (i) Multimodal stimulation module, integrating short pure tone, short tone and dual tone stimulation signal generators, supports accurate output in the auditory frequency range of 4kHz-32kHz for non-human primates;

[0025] (ii) Drug delivery module, including an otoscopic-guided injection catheter, a pressure feedback sensor and a visual light source, adapted to the curvature of the external auditory canal of non-human primates;

[0026] (iii) Data integration and analysis module, with a built-in non-human primate hearing threshold background database, compares ABR and DPOAE data in real time, and generates hearing impairment heat map and multi-dimensional assessment report;

[0027] (iv) An automated quality control unit, through an environmental noise sensor and an animal respiratory monitor, dynamically adjusts the number of signal superpositions and the intensity of stimulation to ensure data reliability.

[0028] Furthermore, in the aforementioned data acquisition and analysis system, the data integration and analysis module employs a deep learning algorithm—convolutional neural network—to extract features from the ABR waveforms of non-human primates, automatically identify differentiation abnormalities from wave I to wave V, and correlate them with DPOAE threshold data to output a cochlear outer hair cell activity score.

[0029] This invention also discloses the application of the above-mentioned method and system in preclinical research of otological drugs, specifically for:

[0030] (A) To evaluate the effect of otological gene therapy drugs on the repair of hair cells in the inner ear of non-human primates and to quantify the degree of cochlear function recovery by the change in DPOAE response amplitude;

[0031] (B) Establish a non-human primate age-related hearing loss model, combining delayed ABR latency and increased DPOAE high-frequency threshold (>75dB SPL) to simulate the pathological process of age-related hearing loss.

[0032] (C) Screening for safe doses of ototoxic drugs: when the amplitude of ABR wave I decreases by more than 50% and the DPOAE distortion products disappear, it is determined to be irreversible cochlear damage.

[0033] Furthermore, in the above application, the non-human primate is a monkey species, more preferably a cynomolgus monkey.

[0034] Compared with the prior art, the present invention has the following outstanding advantages:

[0035] This invention discloses a method and system for acquiring and analyzing auditory electrophysiological data of experimental animals, which has the following advantages:

[0036] 1. Simplified operation and enhanced security:

[0037] Medications can be delivered via the external auditory canal (non-invasive), avoiding complex surgeries such as temporal bone fenestration and reducing the risk of inner ear damage.

[0038] A pressure feedback sensor (≤50kPa) ensures the integrity of the tympanic membrane, increasing the injection success rate to over 95%.

[0039] 2. Species adaptability optimization:

[0040] Specialized catheters and testing parameters were designed based on the anatomical characteristics of the cynomolgus ear (curvature of the external auditory canal and thickness of the tympanic membrane).

[0041] The ABR frequency range (4-32kHz) covers the auditory sensitive area of ​​cynomolgus monkeys, reducing threshold determination errors.

[0042] 3. Data integration and automated analysis:

[0043] Multimodal data (ABR latency, DPOAE amplitude) are correlated in real time to build a machine learning model (accuracy > 90%).

[0044] Automated report generation (less than 10 minutes per sample) is 12 times more efficient than manual interpretation.

[0045] 4. Application scenario expansion:

[0046] Supports dynamic monitoring of gene therapy (such as quantification of AAV vector efficacy), construction of age-related hearing loss models, and prediction of cross-species toxicity.

[0047] 5. Technical compatibility and scalability:

[0048] The system features a modular design, making it adaptable to different laboratory animals (such as miniature pigs and dogs).

[0049] Deep learning algorithms (CNNs) support future expansion to the analysis of the central auditory pathway. Attached Figure Description

[0050] Figure 1 The main process of the method for collecting and analyzing auditory electrophysiological data of experimental animals disclosed in this invention;

[0051] Figure 2 The main structure of a system for acquiring and analyzing auditory electrophysiological data of experimental animals disclosed in this invention. Detailed Implementation

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] The equipment and components used in this invention are all commercially available and are all existing technologies.

[0054] Example 1

[0055] like Figure 1 As shown, drug delivery and auditory electrophysiological testing in the middle ear cavity of cynomolgus monkeys include the following steps: Step (a): Drug delivery and data synchronization recording.

[0056] Device configuration: The drug delivery device uses a catheter with an outer diameter of 0.8 mm, which is adapted to the curvature of the external auditory canal of cynomolgus monkeys (average diameter 6 mm), and is equipped with a pressure feedback sensor (range 0-100 kPa) and a visual light source (LED wavelength 450 nm).

[0057] Operating procedures:

[0058] The cynomolgus monkeys were anesthetized (with isoflurane inhalation and a stable respiratory rate of 50 breaths per minute) and fixed in a lateral decubitus position.

[0059] The gene therapy drug (5 μL volume, 0.2 μL / s rate) is injected into the upper posterior quadrant of the tympanic membrane using an otoscope, with the catheter tip 2 mm from the tympanic membrane, at a pressure ≤30 kPa.

[0060] Synchronous recording: ABR / DPOAE detection was initiated immediately after drug administration, and baseline data and data were recorded at 5 minutes and 30 minutes after drug administration.

[0061] Step (b): ABR / DPOAE multimodal detection

[0062] ABR parameters: short pure audio frequencies of 4kHz, 8kHz, 16kHz, 24kHz, and 32kHz, with intensity decreasing by 5dB to 20dB SPL from 90dB SPL, and signal superposition 512 times (can be appropriately reduced or increased).

[0063] DPOAE parameters: dual-tone f1 = 8kHz, f2 = 9.6kHz (f2 / f1 = 1.2), sound intensity decreases from 80dB SPL in increments of 10dB or 5dB, and the amplitude of 2f1-f2 is recorded.

[0064] Step (c): Anesthesia and Electrode Positioning

[0065] Anesthesia depth control: respiratory rate stabilized at 45 breaths / minute (fluctuation range ±5 breaths).

[0066] Three-electrode positioning: reference electrode (top of the head), anode (behind the ear on the same side), cathode (behind the ear on the opposite side), impedance matching ≤1kΩ.

[0067] Step (d): Noise Suppression and Signal Optimization

[0068] Environmental noise control: Noise level in the shielded room <25dB SPL, with an adaptive algorithm to eliminate transient noise >30dB.

[0069] Signal superposition count: dynamically adjusted based on real-time signal-to-noise ratio (minimum 500 times, maximum 1000 times).

[0070] Step (e): Database construction and damage assessment

[0071] Database entry: ABR latency (wave I: 1.8ms, wave V: 5.2ms), DPOAE threshold (40dB SPL).

[0072] Machine learning output: The support vector machine (SVM) determines the probability of cochlear damage to be 85% and the probability of auditory nerve damage to be <5%.

[0073] Example 2

[0074] Middle ear administration in SD rats and cross-species data validation

[0075] Steps (a)-(e) correspond to the following operations

[0076] Drug delivery: catheter outer diameter 0.5 mm, injection volume 2 μL, pressure ≤20 kPa.

[0077] ABR / DPOAE parameters: short pure audio frequency 1-32kHz, DPOAE f2 / f1=1.2.

[0078] Anesthesia and electrodes: respiratory rate 60 breaths / minute, three-electrode impedance ≤1kΩ.

[0079] Noise suppression: After 500 signal superpositions, the ambient noise is <30dB SPL.

[0080] Cross-species association: The 8kHz threshold (30dB SPL) of SD rats was mapped to the 4kHz threshold (45dB SPL) of cynomolgus monkeys with an error of ±3dB.

[0081] Example 3

[0082] Construction of an age-related hearing model in elderly cynomolgus monkeys

[0083] Steps (a)-(e) correspond to the following operations

[0084] Drug delivery: Inject physiological saline (control), volume 5 μL, pressure ≤30 kPa.

[0085] ABR / DPOAE detection: High frequency 32kHz short pure tone, DPOAE 16kHz dual tone.

[0086] Anesthesia control: Respiratory rate stabilized at 40 breaths / minute (metabolism slows down in older animals).

[0087] Signal optimization: 1000 superpositions, noise suppression threshold 25dB SPL.

[0088] Model output: ABR wave V latency annual growth rate 0.8ms, DPOAE threshold annual increase 7.5dB.

[0089] Example 4

[0090] Screening of safe dosage of ototoxic drug gentamicin

[0091] Steps (a)-(e) correspond to the following operations

[0092] Drug delivery: Gentamicin (10 mg / kg, 20 mg / kg, 30 mg / kg), injection volume 5 μL.

[0093] ABR / DPOAE detection: Daily monitoring of wave I amplitude and DPOAE signal.

[0094] Anesthesia and electrodes: respiratory rate 50 breaths / minute, impedance ≤12kΩ.

[0095] Noise suppression: Dynamic superposition 600 times to eliminate operating room equipment noise.

[0096] Determination of safe dosage: Irreversible damage is defined as a decrease in wave I amplitude of >50% (day 3 in the high-dose group).

[0097] Example 5

[0098] Dynamic monitoring of the efficacy of gene therapy drug AAV-SLC26A4

[0099] Steps (a)-(e) correspond to the following operations

[0100] Drug delivery: AAV carrier is injected in single doses of 5 μL at a pressure ≤25 kPa.

[0101] ABR / DPOAE testing: Testing weekly until week 8.

[0102] Anesthesia and electrodes: respiratory rate stabilized at 55 breaths / minute, impedance ≤8kΩ.

[0103] Signal optimization: Superimposed 700 times, noise suppression algorithm filters out breathing interference.

[0104] Therapeutic output: DPOAE amplitude increased by 20 dB, and ABR wave I latency was shortened by 28%.

[0105] Example 6

[0106] A system for acquiring and analyzing auditory electrophysiological data in experimental animals, such as Figure 2 As shown, it includes the following modules:

[0107] 1) Multimodal stimulation module:

[0108] Short pure tone generator (4-32kHz, 5dB step), dual-tone stimulator (f2 / f1 = 1.2).

[0109] Output accuracy ±1dB, frequency error <0.1%.

[0110] 2) Drug delivery module:

[0111] Otoscopic guidance catheter (fitted to the external auditory canal of cynomolgus monkeys), pressure sensor (range 0-100kPa).

[0112] Visual light source (500 Lux), magnifying glass module (5×).

[0113] 3) Data integration module:

[0114] Built-in database: stores cynomolgus monkey ABR threshold (4kHz:45dB) and DPOAE threshold (8kHz:40dB).

[0115] Heatmap generation: Cochlear lesion areas are marked in red (threshold > 75 dB SPL).

[0116] 4) Quality Control Unit:

[0117] Ambient noise sensor (real-time monitoring <30dB SPL).

[0118] Respiratory monitor (alarm threshold: respiratory rate <35 or >65 breaths / minute).

[0119] Comparative Example 1

[0120] Comparison of conventional surgical drug administration with drug administration in Example 1 of this invention

[0121] Experimental design: Ten cynomolgus monkeys were selected and randomly divided into two groups:

[0122] Control group: Gentamicin (10 μL, 0.5 μL / s) was injected into the round window membrane via temporal bone fenestration.

[0123] Experimental group: Drug administration was performed using the otoscopic guidance method of the present invention (10 μL, 0.5 μL / s).

[0124] The results are shown in Table 2.

[0125] Table 2 Comparison of conventional surgical administration and administration in Example 1 of the present invention.

[0126]

[0127] Conclusion: The minimally invasive technique of this invention significantly reduces surgical complications and improves drug administration safety. (Comparative Example 2)

[0128] Comparison of testing parameter fit between rodents and cynomolgus monkeys

[0129] Experimental design: The DPOAE threshold of SD rats and cynomolgus monkeys was tested under the same acoustic environment. Traditional parameters: two-tone f2 / f1 = 1.2, sound intensity gradient 10 dB.

[0130] Parameters of this invention: gradient 5dB, sound intensity range 20-80dB SPL

[0131] The results are shown in Table 3.

[0132] Table 3 Comparison of Detection Parameter Fit Between Rodents and Cynomolgus Monkeys

[0133] Species frequency Traditional method threshold (dB) The threshold (dB) of this invention percentage decrease SD rats 8kHz 35±5 32±3 ↓40％ Crab-eating macaques 8kHz 52±8 45±2 ↓75％

[0134] Conclusion: Species-specific parameters improve detection accuracy by more than 2 times.

[0135] Application Example 1

[0136] Dynamic monitoring of gene-editing drug efficacy

[0137] Operating procedures:

[0138] CRISPR / Cas9 gene-editing drug (AAV vector, 5 μL) was injected via a minimally invasive catheter.

[0139] Daily synchronous DPOAE (16kHz) and ABR (32kHz) detections were performed, and a machine learning model was used to analyze changes in cochlear outer hair cell activity in real time.

[0140] Key data:

[0141] DPOAE amplitude: Increased by 18 dB on day 7 (p<0.01)

[0142] ABR wave I latency: shortened by 0.4 ms (p<0.05)

[0143] Time to assess efficacy: Significant changes were observed 14 days earlier than with traditional methods.

[0144] Advantages: Enables real-time quantification of the efficacy of gene therapy.

[0145] Application Example 2

[0146] Construction of a cross-species ototoxicity prediction platform

[0147] Implementation steps:

[0148] Establish a cynomolgus monkey-rat-mouse auditory threshold mapping model

[0149] Input cisplatin toxicity data in rats (wave I decreased by 40%).

[0150] The system predicts the equivalent dose for cynomolgus monkeys with an error of <15%.

[0151] The results are shown in Table 4.

[0152] Table 4. Construction of the Cross-Species Ototoxicity Prediction Platform

[0153] Species Actual ototoxic dose (mg / kg) Predicted dose (mg / kg) Error rate Crab-eating macaques 3.2 3.0 6.25％ C57 mice 12.5 13.8 10.4％

[0154] The results above show that the cost of repeating experiments in cross-species research can be reduced by up to 60% (Test Example 1).

[0155] Multimodal data integration efficiency test

[0156] Experimental design: For the same batch of cynomolgus monkeys (n=20), the following methods were used respectively:

[0157] Traditional manual analysis: Independent processing of ABR / DPOAE data

[0158] This invention system: Automatic correlation analysis

[0159] Efficiency comparison is shown in Table 5

[0160] Table 5 Multimodal data integration efficiency test

[0161] Analysis content Traditional time consumption (h) Time required for this invention (min) Consistency rate Hearing threshold determination 3.2±0.5 8±2 92％ Location of injury 4.5±0.8 10±3 88％ Multidimensional report generation 6.0±1.2 15±5 95％

[0162] Breakthrough: Analysis efficiency improved by 12-24 times, meeting high-throughput screening requirements. Test Case 2

[0163] Machine learning model performance verification 1

[0164] Test plan:

[0165] The performance metrics of Support Vector Machine (SVM) and Random Forest (RF) algorithms were compared by inputting 200 sets of ABR / DPOAE data from cynomolgus monkeys (including manually labeled results). Table 6 shows the performance metrics.

[0166] Table 6 Performance Validation of Machine Learning Models 1

[0167] algorithm accuracy Sensitivity Specificity AUC value SVM 91.2％ 89.5％ 92.8％ 0.94 RF 93.7％ 92.1％ 94.3％ 0.96 artificial 85.4％ 83.2％ 87.6％ -

[0168] Conclusion: The accuracy of automated analysis exceeded that of manual interpretation by 8.3 percentage points. (Test Example 3)

[0169] Robustness testing in noisy environments

[0170] Operating parameters:

[0171] Ambient noise: 45dB SPL (simulating typical laboratory conditions)

[0172] Enable adaptive noise suppression algorithm

[0173] Test frequency: 16kHz ABR detection

[0174] The results are shown in Table 7.

[0175] Table 7 Robustness Tests under Noise Environment

[0176] Number of stacks traditional method signal-to-noise ratio The signal-to-noise ratio of this invention 500 1.2:1 3.8:1 1000 1.8:1 6.5:1

[0177] Innovation: Usable data can still be obtained under harsh conditions (SNR > 3:1)

[0178] Other test cases are summarized in Table 8 below.

[0179] Table 8 Summary of Other Test Cases

[0180]

[0181]

[0182] In summary, this invention demonstrates significant innovation and technological advantages in the field of auditory electrophysiological data acquisition and analysis in experimental animals. Through the integrated application of key technologies such as non-invasive drug delivery, multimodal detection and data analysis, precise anesthesia and electrode positioning, and noise suppression and signal optimization, this invention not only significantly improves the accuracy and efficiency of otopathic drug research but also substantially reduces experimental risks. Its advantages, including simplified operation, high safety, optimized species adaptability, and data integration and automated analysis, provide strong technical support for multiple fields such as otopathic drug development, gene therapy efficacy evaluation, age-related hearing loss research, and cross-species toxicity prediction. The successful implementation of this invention will powerfully promote the advancement of otopathic medical research and has broad application prospects and significant scientific value.

[0183] The above are merely a few preferred embodiments of the present invention, described in a relatively specific and detailed manner, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

Claims

1. A method for collecting and analyzing auditory electrophysiological data of experimental animals, characterized in that, Includes the following steps: (a) Using a drug delivery device, drugs are delivered into the middle ear cavity of non-human primates through the external auditory canal, and auditory electrophysiological data are recorded simultaneously before and after drug administration; (b) Auditory pathway signals of non-human primates were collected by multimodal detection of auditory brainstem evoked potentials (ABR) and distortion product otoacoustic emissions (DPOAE). ABR detection used short pure tone stimulation with a frequency range of 4 kHz to 32 kHz, with stimulation intensity decreasing by 5 dB or 10 dB. DPOAE detection used two-tone stimulation with a frequency ratio of f2 / f1=1.2, with a sound intensity range of 20 dB to 80 dB SPL. (c) Based on the physiological characteristics of non-human primates, the depth of anesthesia was set to a respiratory rate of 40-60 breaths / minute, and the motion artifact interference was reduced by using a three-electrode positioning method at the top of the head, behind the ipsilateral auricle, and behind the contralateral auricle. (d) An adaptive noise suppression algorithm is adopted to eliminate interference signals with environmental noise >30 dB SPL in real time, and the signal-to-noise ratio is optimized by dynamically adjusting the number of signal superpositions (500-1000 times). (e) By combining ABR waveform, latency, amplitude and DPOAE response amplitude, a non-human primate-specific auditory threshold database is constructed to determine the location of auditory damage and output the probability of abnormalities in the cochlea, auditory nerve or central pathway.

2. The method according to claim 1, characterized in that, In step (a), the drug delivery device is equipped with a visualization light source and a magnifying glass module. The operator observes the upper quadrant of the tympanic membrane in real time through an otoscope. The upper limit of the injection volume is 50 μL. The injection is slow, and the injection pressure is controlled to be ≤50 kPa by a pressure feedback sensor.

3. The method according to claim 1, characterized in that, In step (b), the ABR detection targets the auditory nerve to brainstem pathway of cynomolgus monkeys and uses the latency difference threshold method. When the peak delay of wave I to wave V exceeds 0.2 ms, it is determined to be auditory nerve damage, and when the amplitude of wave V decreases by more than 50%, it is determined to be central pathway abnormality.

4. A system for acquiring and analyzing auditory electrophysiological data in experimental animals, characterized in that, include: (i) Multimodal stimulation module, integrating short pure tone, short sound and dual tone stimulation signal generators, supporting accurate output in the auditory frequency range of 4 kHz-32 kHz for non-human primates; (ii) Drug delivery module, including an otoscope-guided injection catheter, a pressure feedback sensor and a visual light source, adapted to the curvature of the external auditory canal of non-human primates; (iii) Data integration and analysis module, with a built-in non-human primate hearing threshold background database, compares ABR and DPOAE data in real time, and generates hearing impairment heat map and multi-dimensional assessment report; (iv) An automated quality control unit, through an environmental noise sensor and an animal respiratory monitor, dynamically adjusts the number of signal superpositions and the intensity of stimulation to ensure data reliability.

5. The system according to claim 4, characterized in that, The drug delivery module further integrates an acoustic coupling interface to simultaneously trigger DPOAE detection during drug injection, thereby monitoring changes in the activity of cochlear outer hair cells in real time.

6. The system according to claim 4, characterized in that, The data integration and analysis module uses a deep learning algorithm—convolutional neural network—to extract features from the ABR waveforms of non-human primates, automatically identify differentiation abnormalities from wave I to wave V, and correlate them with DPOAE threshold data to output a cochlear outer hair cell activity score.

7. The application of the method according to claims 1-3 or the system according to claims 4-6 in preclinical research of otological drugs, characterized in that, Specifically used for: (A) To evaluate the repair effect of otological gene therapy drugs on the inner ear hair cells of non-human primates and to quantify the degree of cochlear function recovery by the change in DPOAE response amplitude; (B) Establish a non-human primate age-related hearing loss model, combining delayed ABR latency and increased DPOAE high-frequency threshold (>75 dB SPL) to simulate the pathological process of age-related hearing loss. (C) Screening safe doses of ototoxic drugs: when the amplitude of ABR wave I decreases by more than 50% and the DPOAE distortion products disappear, it is determined to be irreversible cochlear damage.

8. The application according to claim 7, characterized in that, The non-human primate is a monkey species, more preferably a cynomolgus monkey.

9. The application according to claim 7, characterized in that, The application achieves cross-species data compatibility through a standardized process, linking the auditory threshold of cynomolgus monkeys with databases of SD rats and C57 mice to construct a cross-model drug toxicity prediction platform.

Images