CN-121987158-A - Monitoring and evaluating system and method for postoperative recurrent laryngeal nerve function damage and recovery

CN121987158ACN 121987158 ACN121987158 ACN 121987158ACN-121987158-A

Abstract

The invention discloses a system and a method for monitoring and evaluating postoperative recurrent laryngeal nerve function damage and recovery, and relates to the technical field of laryngeal nerve function evaluation. The invention converts postoperative cervical muscle tone signals into multidimensional computable characteristics reflecting nerve discharging rhythms, muscle mechanics modes and nerve-muscle conduction coupling relations by constructing a pre-operation individualized neuromuscular vibration fingerprint base line of a patient, and carries out structural comparison under an identical stable topological reference, thereby realizing the conversion from subjective observation to objective quantification of the laryngeal return nerve function state, dynamically tracking the stable maintenance degree of a nerve control structure under the non-invasive and continuously wearable conditions, distinguishing neurogenic abnormality, myogenic change and normal function recovery trend, obviously reducing the influence of individual anatomical difference, sounding habit and environmental noise on an evaluation result, and improving the sensitivity, repeatability and long-term trend judgment capability of postoperative function monitoring.

Inventors

ZHU YI
LENG XUEFENG
Peng deke
DONG JIANMING
YANG JIE
LUO XINYI

Assignees

四川省肿瘤医院
四川康源医创科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (8)

1. A monitoring and evaluating system for postoperative recurrent laryngeal nerve function injury and recovery, comprising: The preoperative acquisition module is used for acquiring preoperative polymorphic laryngeal mechanical myoacoustic signals of different physiological states of a patient before operation through the cervical multi-point array myoacoustic sensor; The time-frequency analysis module is used for carrying out time-frequency structure analysis on the preoperative polymorphic laryngeal mechanical myoacoustic signals and constructing an individual neuromuscular vibration fingerprint baseline model; The postoperative acquisition module is used for continuously acquiring neck vibration signals of the patient at different postoperative detection points; the separation characteristic module is used for extracting an active muscle contraction vibration component, a body movement interference component and a respiration background vibration component of each neck vibration signal by utilizing blind source separation and spatial filtering to obtain postoperative muscle sound signal sequences of each time point; The mapping module is used for processing each postoperative muscle tone signal sequence through short-time Fourier transformation, establishing a nerve-muscle-vibration structure mapping relation and generating a current nerve control feature vector of each time point; The index calculation module is used for calculating the comprehensive stability maintaining index at each time point based on the current nerve control feature vector of each time point and the individual nerve muscle vibration fingerprint baseline model for projection; the damage classification module is used for generating a damage evaluation feature vector based on the current nerve control feature vector and each comprehensive stability retention index, calculating a damage deviation index, and obtaining a monitoring evaluation result through a preset functional state classification rule.
2. The system for monitoring and assessing the impairment and recovery of post-operative recurrent laryngeal nerve function of claim 1, wherein the processing of the time-frequency analysis module comprises: short-time Fourier transform or continuous wavelet transform is carried out on each preoperative polymorphic laryngeal mechanical myoacoustic signal to obtain a corresponding time-frequency representation matrix: based on the periodic characteristics of the time-frequency representation matrix, corresponding rhythm characteristics, modal characteristics and coupling characteristics are calculated and integrated to generate a multidimensional coupling characteristic vector; And mapping the multidimensional coupling feature vector to an n-dimensional feature space, and taking the corresponding stable topological area as an individual neuromuscular vibration fingerprint baseline model.
3. The system for monitoring and assessing the impairment and recovery of post-operative recurrent laryngeal nerve function of claim 2, wherein the rhythmic features comprise coefficients of variation and pulse group synchronization indices, the modal features comprise energy concentration and frequency drift, and the coupling features comprise time-frequency energy ridge line continuity and phase consistency.
4. The system for postoperative recurrent laryngeal nerve function impairment and recovery monitoring and assessment according to claim 1, wherein the separation characteristic module comprises: correcting the neck vibration signals and calculating a cross correlation coefficient matrix, a phase difference matrix and a space covariance matrix among different channels; generating a linear hybrid model based on each cross-correlation coefficient matrix, each phase difference matrix and each space covariance matrix, and solving by utilizing blind source separation to obtain an independent source mechanical myoacoustic component set; and calculating the main frequency distribution range, the rhythm period stability, the space weight distribution and the phase propagation direction of each component in the independent source mechanical myoacoustic component set, determining the active muscle contraction vibration component, the body movement interference component and the respiratory background vibration component, and generating a postoperative myoacoustic signal sequence.
5. The system for postoperative recurrent laryngeal nerve function impairment and recovery monitoring and assessment according to claim 4, wherein the calculating of the cross-correlation coefficient matrix, the phase difference matrix and the spatial covariance matrix between different channels comprises: correcting the neck vibration signal to generate a multi-channel vibration signal matrix; Calculating a cross correlation coefficient matrix among different channels in the multi-channel vibration signal matrix; calculating a phase difference matrix among different channels in the multi-channel vibration signal matrix; and calculating a spatial covariance matrix between different channels in the multi-channel vibration signal matrix.
6. The system for monitoring and assessing the impairment and recovery of post-operative recurrent laryngeal nerve function of claim 1, wherein the processing of the index calculation module comprises: Respectively projecting each current nerve control feature vector to a topological space of an individual nerve muscle vibration fingerprint baseline model to generate a corresponding postoperative feature mean value vector, a postoperative covariance matrix and a normalized postoperative feature vector, wherein the normalized postoperative feature vector comprises postoperative rhythm features, postoperative modal features and postoperative coupling features; calculating corresponding rhythm maintenance degree, modal maintenance degree and coupling consistency based on each normalized postoperative feature vector and individual neuromuscular vibration fingerprint baseline model; based on each rhythm retention, each modality retention, and each coupling consistency, a comprehensive stability retention index at different time points is calculated, respectively.
7. The system for monitoring and assessing the impairment and recovery of post-operative recurrent laryngeal nerve function of claim 6, wherein the processing of the impairment classification module comprises: performing differential operation on the normalized postoperative feature vector and the mean vector of the individual neuromuscular vibration fingerprint baseline model to generate a multidimensional deviation vector; Combining all comprehensive stability maintaining indexes and the rhythm maintaining degree, the mode maintaining degree and the coupling consistency thereof to generate a function maintaining vector; Calculating a damage deviation index based on the multidimensional deviation vector and the function maintenance vector; combining the multidimensional deviation vector and the function maintaining vector, inputting the multidimensional deviation vector and the function maintaining vector into a damage mode classifier, and outputting a damage leading mode; and obtaining a monitoring evaluation result through a preset functional state grading rule based on the damage deviation index and the damage leading mode.
8. A method for monitoring and evaluating postoperative recurrent laryngeal nerve function damage and recovery, for realizing the system for monitoring and evaluating postoperative recurrent laryngeal nerve function damage and recovery according to any one of claims 1 to 7, comprising: acquiring preoperative polymorphic laryngeal mechanical myoacoustic signals of different physiological states of a patient before operation through a cervical multipoint array myoacoustic sensor, and performing frequency structure analysis to construct an individual neuromuscular vibration fingerprint baseline model; continuously collecting neck vibration signals of a patient at different post-operation detection points, and separating an active muscle contraction vibration component, a body movement interference component and a respiratory background vibration component by using blind source separation and spatial filtering to obtain post-operation muscle tone signal sequences at all time points; processing each postoperative muscle tone signal sequence through short-time Fourier transform, establishing a nerve-muscle-vibration structure mapping relation, generating a current nerve control feature vector of each time point, projecting by combining an individual nerve muscle vibration fingerprint baseline model, and calculating a comprehensive stability retention index at each time point; Based on the current nerve control feature vector and each comprehensive stability maintaining index, generating a damage evaluation feature vector, calculating a damage deviation index, and obtaining a monitoring evaluation result through a preset functional state grading rule.

Description

Monitoring and evaluating system and method for postoperative recurrent laryngeal nerve function damage and recovery Technical Field The invention relates to the technical field of laryngeal nerve function assessment, in particular to a system and a method for monitoring and assessing the damage and recovery of laryngeal return nerve function after operation. Background Recurrent laryngeal nerves (recurrent LARYNGEAL NERVE, RLN) are key nerves that govern laryngeal muscle movement, and their injury is one of the common complications of thoracic and cervical surgery (e.g., esophageal cancer, thyroid cancer, lung cancer, etc.). The postoperative nerve injury may cause patients to have symptoms such as hoarseness, cough and water choking, dyspnea, etc., seriously affect life quality, and may delay subsequent radiotherapy and chemotherapy or rehabilitation treatment. Therefore, the method has important clinical significance in early, continuous and noninvasive dynamic monitoring of the postoperative recurrent laryngeal nerve function. Currently commonly used methods for laryngeal return function assessment include laryngoscopy, laryngeal Electromyography (EMG), acoustic analysis, percutaneous laryngeal ultrasound, and the like. The laryngoscopy can directly observe vocal cord movement, is a gold standard for diagnosis, has strong operation invasiveness, depends on professionals and cannot realize continuous monitoring, can record the electrical activity of innervating muscles by using a laryngeal electromyography, has traumata in the electrode implantation process, is easy to interfere with signals, limits the continuous application after operation, can reflect sounding states by using acoustic analysis and voice evaluation, is greatly influenced by speaking habits and background noise, is difficult to accurately capture nerve functions in the resting state, and has limitations in sensitivity and specificity due to the fact that percutaneous laryngeal ultrasound is noninvasive and real-time, but is influenced by experience of operators and individual anatomical differences (such as calcification of thyroid cartilage). Disclosure of Invention The invention aims to provide a monitoring and evaluating system and method for postoperative recurrent laryngeal nerve function damage and recovery, which are used for solving the technical problems that the prior art does not quantify a reference system corresponding to the control state of the preoperative neuromuscular of a patient and cannot realize objective, continuous and individual evaluation of the type and recovery degree of recurrent laryngeal nerve function damage. In order to achieve the above object, the embodiment of the present invention provides the following technical solutions: a monitoring and assessment system for postoperative recurrent laryngeal nerve function damage and recovery, comprising: The preoperative acquisition module is used for acquiring preoperative polymorphic laryngeal mechanical myoacoustic signals of different physiological states of a patient before operation through the cervical multi-point array myoacoustic sensor; The time-frequency analysis module is used for carrying out time-frequency structure analysis on the preoperative polymorphic laryngeal mechanical myoacoustic signals and constructing an individual neuromuscular vibration fingerprint baseline model; The postoperative acquisition module is used for continuously acquiring neck vibration signals of the patient at different postoperative detection points; the separation characteristic module is used for extracting an active muscle contraction vibration component, a body movement interference component and a respiration background vibration component of each neck vibration signal by utilizing blind source separation and spatial filtering to obtain postoperative muscle sound signal sequences of each time point; The mapping module is used for processing each postoperative muscle tone signal sequence through short-time Fourier transformation, establishing a nerve-muscle-vibration structure mapping relation and generating a current nerve control feature vector of each time point; The index calculation module is used for calculating the comprehensive stability maintaining index at each time point based on the current nerve control feature vector of each time point and the individual nerve muscle vibration fingerprint baseline model for projection; the damage classification module is used for generating a damage evaluation feature vector based on the current nerve control feature vector and each comprehensive stability retention index, calculating a damage deviation index, and obtaining a monitoring evaluation result through a preset functional state classification rule. Further, the processing procedure of the time-frequency analysis module comprises: short-time Fourier transform or continuous wavelet transform is carried out on each preoperative polymorphic laryngeal mechanical