A method for detecting the severity of a patient with pneumothorax based on voiceprint features

Abstract

The application discloses a kind of pneumothorax patient illness severity detection methods based on voiceprint features, it is related to pneumothorax patient detection technical field, including the following steps: after the position of pneumothorax patient mutation, based on the real-time monitoring signal of thoracic cavity sound field constructs thoracic cavity sound energy distribution map, through to the joint analysis of sound wave propagation delay and amplitude gradient, extract the spatial drift trajectory of sound energy concentration area in thoracic cavity;Thoracic cavity sound energy distribution map is obtained using spatial drift trajectory, time series stratified analysis is carried out to thoracic cavity sound energy data.The application realizes the real-time capture and stable analysis of thoracic cavity sound field under the position mutation by constructing thoracic cavity sound energy distribution map and extracting spatial drift trajectory, reduces sound wave reflection interference, improves the accuracy and continuity of pneumothorax voiceprint detection;Meanwhile, through turn-back inhibition, phase rearrangement and energy weight mapping, make sound field energy rebalance, realize the dynamic consistency of lesion positioning accuracy and illness assessment result, enhance the reliability of detection.

Description

Technical Field

[0001] This invention relates to the field of pneumothorax patient detection technology, specifically to a method for detecting the severity of pneumothorax in patients based on voiceprint characteristics. Background Technology

[0002] Voiceprint-based assessment of pneumothorax severity utilizes acoustic signals such as breath sounds, vesicular sounds, and pleural resonances. Through acoustic feature extraction and pattern recognition techniques, it analyzes the degree of lung tissue collapse and changes in pleural gas distribution caused by pneumothorax, thereby assessing the severity of the condition. Specifically, the system uses a multi-point acquisition array on the chest to acquire breath sound signals from different locations in real time. Acoustic features of the lungs are extracted using voiceprint parameters such as spectral analysis, Mel-frequency cepstral coefficients (MFCC), short-time energy, and fundamental frequency perturbation rate. Machine learning models or deep neural networks are then used to model these voiceprint features, comparing them with a database of normal lung sounds to identify energy attenuation, frequency drift, and abnormal resonance patterns in the pneumothorax area. Based on the amplitude and spatial distribution of these voiceprint features, the extent of pneumothorax, the degree of lung re-expansion, and the compressive trend can be non-invasively determined, enabling dynamic grading of the condition and providing doctors with rapid, quantitative, and repeatable diagnostic evidence.

[0003] The existing technology has the following shortcomings: During voiceprint detection in patients with pneumothorax, sudden changes in patient position cause rapid migration of the previously stable free gas layer within the pleural cavity. This results in dynamic overlap and reflection superposition effects on the propagation paths of sound waves within the pleural cavity. Due to abrupt changes in acoustic impedance at the free gas layer interface, this dynamic migration process creates multipath interference structures in local areas, leading to abnormal concentration and nonlinear amplification of sound wave energy within specific frequency bands. These transient interference peaks interfere with the amplitude and phase stability of the voiceprint signal, making it difficult for the detection system to accurately distinguish between real lesion reflections and gas migration spurious signals during timing calculations and sound source localization. This results in significant errors in lesion boundary identification, voiceprint energy distribution distortion, and severity assessment.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for detecting the severity of pneumothorax in patients based on voiceprint features, so as to solve the problems in the background art mentioned above.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting the severity of pneumothorax in patients based on voiceprint features, comprising the following steps: Step 1: After a sudden change in the body position of a pneumothorax patient, a chest cavity acoustic energy distribution map is constructed based on the real-time monitoring signal of the chest cavity acoustic field. By jointly analyzing the sound wave propagation delay and amplitude gradient, the spatial drift trajectory of the acoustic energy concentration area in the chest cavity is extracted to form the time and space reference for subsequent sound wave propagation path reconstruction. Step 2: Using the spatial drift trajectory obtained from the chest cavity acoustic energy distribution map, perform time series layered analysis on the chest cavity acoustic energy data, separate and map the sound wave propagation paths in different time periods, and extract the energy interference features of the multi-path superposition area based on the separation and mapping results, providing a positioning basis for the stabilization processing of the chest cavity sound field. Step 3: Based on the obtained energy interference characteristics, establish a chest cavity sound field reflection suppression model. Transiently constrain the high-energy interference peaks through delayed phase modulation and energy attenuation control, so that the sound wave propagation path in the chest cavity is redistributed, forming a relatively stable sound wave propagation region. Step 4: Based on the formed sound wave propagation region, the phase sequence of the chest cavity sound wave signal is synchronously rearranged, and the adjacent propagation paths are continuously registered using a time-sliding window to compensate for the temporal misalignment caused by the migration of the free gas layer, thereby maintaining the amplitude and phase continuity of the sound wave signal. Step 5: Based on the synchronously rearranged chest cavity acoustic wave signal, construct a chest cavity acoustic energy weight mapping, redistribute the energy weights of different propagation paths to the chest cavity acoustic field coordinate space, so that the acoustic energy distribution after the gas layer migration is stabilized again within the solution range, thereby improving the positioning accuracy of the pneumothorax lesion area and dynamically consistent with the assessment results of the severity of the condition.

[0007] Preferably, the steps of constructing a chest cavity acoustic energy distribution map and extracting the spatial drift trajectory of the acoustic energy concentration area within the chest cavity after a sudden change in the patient's position in pneumothorax include: The system continuously collects respiratory sounds, vesicular sounds, and chest resonance signals from the chest wall surface using a multi-point acoustic sensor array attached to different areas of the patient's chest. The system also performs delayed synchronization processing on the signals from each collection point, using the instant of sudden change in body position as the time reference. After time synchronization is completed, the propagation delay and amplitude gradient of sound waves at each acquisition point are jointly analyzed to establish a preliminary spatial mapping relationship of chest cavity sound energy in order to identify energy concentration areas and energy sparse areas caused by gas layer migration. Based on the temporal changes in the acoustic energy distribution in the chest cavity, the continuous time frames are analyzed frame by frame to track the displacement changes of the region with the highest energy density in spatial coordinates and extract the spatial drift trajectory of the region where the acoustic energy in the chest cavity is concentrated. By combining the drift trajectory with the overall structure of the chest cavity acoustic energy distribution map, a temporal and spatial benchmark for the reconstruction of the chest cavity acoustic wave propagation path is formed, providing a unified reference for subsequent sound field stabilization processing.

[0008] Preferably, when extracting the spatial drift trajectory of the concentrated acoustic energy region in the thoracic cavity, the acoustic energy concentration center of the previous time frame is used as the reference point of the current time frame by adopting an energy centroid correlation method between adjacent time frames. This ensures the continuity of the drift trajectory in the time dimension and the spatial correspondence, thereby accurately reflecting the migration direction and range of the free gas layer in the thoracic cavity after a sudden change in body position.

[0009] Preferably, the steps for performing time-series hierarchical analysis of chest cavity acoustic energy data using the spatial drift trajectory obtained from the chest cavity acoustic energy distribution map include: Based on the obtained spatial drift trajectory, the chest cavity acoustic energy distribution map is layered along the time axis so that each time layer corresponds to a stable acoustic field state during the gas layer migration process, and the time layer boundary is determined by the position change of the spatial drift trajectory. Spatial mapping and separation of the sound wave propagation path within each time layer are performed, and the drift trajectory is used as a spatial reference line to distinguish between the main propagation path and the secondary propagation path, thereby obtaining the separation and mapping results of the chest cavity sound wave propagation. Based on the separation mapping results, spatial overlay analysis is performed on the overlapping areas between different paths to extract the energy interference features of chest cavity acoustic energy and identify local energy concentration areas; The obtained energy interference features are fused with the spatial drift trajectory in the chest cavity acoustic energy distribution map to establish a spatiotemporal correlation between the energy interference distribution and the gas layer migration trajectory, thus forming the positioning basis for the chest cavity acoustic field stabilization processing.

[0010] Preferably, when fusing the energy interference features with the spatial drift trajectory in the chest cavity acoustic energy distribution map, the spatial location of the energy interference region is matched with the time node of the drift trajectory, so that each energy interference feature corresponds to the motion state of the gas layer in the chest cavity during a specific time period, thereby forming a spatial positioning result of the dynamic change of the chest cavity acoustic field, providing an accurate spatiotemporal reference for subsequent chest cavity acoustic field reflection suppression and sound wave propagation path reconstruction.

[0011] Preferably, the steps for establishing a chest cavity sound field reflection suppression model based on the obtained energy interference characteristics include: Based on the obtained energy interference characteristics, the regions with abnormally concentrated energy in the chest cavity sound field are identified and spatially calibrated, and the distribution range of the high-energy interference region in the chest cavity coordinate system is determined. Based on the temporal variation law in the energy interference characteristics, the propagation delay characteristics of sound waves in the interference region are phase modulated to make the multipath reflected waves produce a controllable phase shift in time, so as to weaken the formation of local energy peaks. Based on the acoustic energy distribution results after phase modulation, energy attenuation control is performed on the high-energy part in the interference region to make the local acoustic energy diffuse evenly in time and space, and to avoid excessive energy accumulation leading to sound field imbalance. Based on the chest cavity acoustic energy distribution map after energy attenuation control, the chest cavity sound wave propagation path is reconstructed, so that the chest cavity sound field changes from a multi-path interference state to a stable state with balanced energy distribution and continuous propagation path.

[0012] Preferably, during the energy attenuation control process, based on the continuity of the energy gradient in each region of the chest cavity acoustic energy distribution map, a smooth transition is implemented in the region where the energy density exceeds the normal acoustic energy threshold, so that the energy diffuses evenly from the interference center to the surrounding region, thereby achieving energy redistribution when reconstructing the chest cavity acoustic wave propagation path, and making the sound wave propagation form a stable channel along the acoustic impedance gradient of the chest cavity tissue.

[0013] Preferably, the step of synchronously rearranging the phase sequence of the chest cavity sound wave signal based on the formed sound wave propagation region includes: Based on the chest cavity sound field state after energy return suppression, the sound wave signal is collected in a partition and the phase sequence is extracted in the stable sound wave propagation area. A time-continuous mapping of the chest cavity sound field is established with a unified time reference. Based on the phase difference between different propagation regions, the time reference of the chest cavity sound wave signal is adjusted so that the propagation sequence of sound waves in different regions is consistent with the propagation path of chest cavity sound waves, in order to compensate for the local phase shift caused by gas layer migration. By using a time-sliding window to perform continuous registration of adjacent propagation paths, the phase difference between adjacent paths remains stable during the time-sliding process, thus achieving a continuous time transition of chest cavity sound wave propagation. The amplitude and phase of the chest cavity acoustic wave signal after time-sliding window processing are uniformly rearranged to maintain the continuity and energy balance of the chest cavity acoustic wave signal in time and space, providing a stable acoustic basis for subsequent acoustic energy weighting mapping.

[0014] Preferably, during the continuous registration process of the time-sliding window, the phase trend and amplitude change trend of the chest cavity sound wave propagation area are used as the registration basis. By dynamically adjusting the time step of the sliding window, the phase connection of adjacent propagation paths on the time axis is kept smooth, thereby further improving the temporal continuity and spatial alignment accuracy of the chest cavity sound wave signal after synchronous rearrangement.

[0015] Preferably, the step of constructing a chest cavity acoustic energy weighting map based on the synchronously rearranged chest cavity acoustic wave signal includes: Based on the chest cavity acoustic wave signal after synchronous rearrangement, the acoustic energy of different propagation paths is extracted and quantified according to the spatial distribution characteristics of the chest cavity acoustic field, and a spatial mapping basis for chest cavity acoustic energy is established. Based on the principle of spatial continuity of the chest cavity sound field, the energy weights of different propagation paths are redistributed so that energy diffuses evenly from the high-energy region to the low-energy region, forming a continuous sound energy gradient and maintaining the correspondence between energy weights and time series. Based on the updated acoustic energy state, a chest cavity acoustic energy weight mapping is constructed, and the energy weights of each propagation path are marked in the chest cavity acoustic field coordinate space to form an acoustic energy weight mapping diagram with time continuity and spatial stability. Using thoracic acoustic energy weighting mapping as an energy reference framework, the location and severity of pneumothorax lesions are dynamically assessed, enabling the determination of lesion extent and the assessment of disease progression trends.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention achieves real-time capture and spatiotemporal correlation of dynamic changes in the thoracic cavity sound field by constructing a chest cavity acoustic energy distribution map after a sudden change in body position and extracting the spatial drift trajectory of the concentrated acoustic energy region. This allows for the quantitative expression of changes in the sound wave propagation path caused by the migration of the free gas layer within the thoracic cavity. By performing time-series hierarchical analysis and energy interference feature extraction on the sound wave propagation path, the dynamic instability of the sound field is transformed into a controllable spatial localization process. This effectively reduces the overlap and energy drift of sound wave reflections caused by changes in body position, ensuring continuous and stable voiceprint detection under complex body position conditions and improving the accuracy and repeatability of acoustic identification of pneumothorax lesions.

[0017] This invention establishes a chest cavity sound field reflex suppression model and implements phase synchronization rearrangement and energy weight mapping to redistribute chest cavity sound wave propagation in a dynamic environment and maintain energy equilibrium, ensuring the continuity of sound waves in time and space. The chest cavity sound field processed by this method possesses stable propagation characteristics, maintaining a stable sound energy distribution during gas layer migration. This enables high-precision localization of the lesion area and dynamic, consistent assessment of the severity of the condition, providing doctors with quantifiable and continuous evidence of disease changes and improving the diagnostic reliability of pneumothorax detection. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of a method for detecting the severity of pneumothorax in patients based on voiceprint features, according to the present invention. Detailed Implementation

[0020] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0021] This invention provides, for example Figure 1 The method for detecting the severity of pneumothorax in patients based on voiceprint features, as shown, includes the following steps: Step 1: After a sudden change in the body position of a pneumothorax patient, a chest cavity acoustic energy distribution map is constructed based on the real-time monitoring signal of the chest cavity acoustic field. By jointly analyzing the sound wave propagation delay and amplitude gradient, the spatial drift trajectory of the acoustic energy concentration area in the chest cavity is extracted to form the time and space reference for subsequent sound wave propagation path reconstruction. The specific implementation method for this step is as follows: After a sudden change in the patient's position during pneumothorax, a multi-point acoustic sensor array attached to different regions of the patient's chest continuously acquired respiratory sounds, vesicular sounds, and pleural resonances from the chest wall surface. Each sensor unit was distributed in different anatomical locations of the chest (including the subclavian, midaxillary, subscapular, and parasternal regions) to synchronously capture acoustic energy changes at different spatial locations in the sound field during the positional change. During acquisition, the acoustic signals were continuously recorded over several respiratory cycles with the instant of the patient's positional change as the time reference to ensure a complete sequence of acoustic field changes during the gas layer migration process. At this time, the acoustic signals were still in a mixed state, containing reflected and scattered components from the interfaces of normal lung tissue, the collapsed area, and the free gas layer. To effectively identify the energy distribution relationships between these components, the acoustic signals at each sensor point were time-delayed and synchronized to correspond to a unified time reference frame, thus ensuring that acoustic energy information at different locations in the subsequent acoustic energy distribution map could be compared in the same time dimension.

[0022] After achieving time synchronization of the acoustic signals, a preliminary spatial mapping relationship of the thoracic cavity acoustic energy is established by jointly analyzing the propagation delay and amplitude gradient of the signals at each acquisition point. Specifically, the propagation delay of the acoustic waves received at each sensing point is correlated with the change in acoustic energy amplitude at that location, forming a continuous field of acoustic energy variation within the thoracic cavity. The propagation delay reflects the relative length of the propagation path of the acoustic waves between the thoracic tissue and the gas layer, while the amplitude gradient reflects the energy attenuation law of acoustic energy during propagation due to reflection, scattering, and absorption. By jointly analyzing these two parameters, acoustic characteristic regions corresponding to different propagation paths can be distinguished, thereby initially identifying energy concentration areas and energy sparse areas caused by gas layer migration. In this process, by comparing the temporal trends of acoustic energy changes between different parts of the thoracic cavity, it can be observed that after a change in body position, the acoustic energy values ​​of some upper sensing points gradually increase, while the acoustic energy values ​​of lower sensing points decrease accordingly. This indicates that the free gas layer within the thoracic cavity is undergoing a positional redistribution.

[0023] After obtaining the initial spatial mapping of the chest cavity acoustic energy, the acoustic energy distribution at the instant of the sudden change in body position is used as the starting point. Subsequent time-series signals are then analyzed frame by frame to extract the spatial drift trajectory of the acoustic energy concentration region. Specifically, following the sequence of consecutive time frames, the region with the highest energy density in the chest cavity acoustic energy distribution map is tracked, and the continuous displacement changes of this high-energy region in spatial coordinates are recorded, thus obtaining the spatial trajectory of gas layer migration. This spatial drift trajectory reflects the migration direction and range of the gas layer over time, and can intuitively reflect the dynamic evolution of the chest cavity acoustic field structure after a sudden change in body position. To ensure the continuity and spatial correspondence of the trajectory, an energy centroid correlation method is used between adjacent time frames, using the acoustic energy concentration center of the previous time frame as the reference point for the current time frame, so that the drift trajectory forms a smooth spatial curve in the time dimension. In this way, the entire process of the free gas layer migrating from one chest cavity region to another can be clearly displayed, and the main direction of change due to the movement of the gas interface in the chest cavity sound wave propagation path can be determined.

[0024] After obtaining the spatial drift trajectory of the concentrated acoustic energy region in the chest cavity, the overall structure of the drift trajectory and the chest cavity acoustic energy distribution map is combined to form the temporal and spatial reference for subsequent sound wave propagation path reconstruction. Specifically, the drift trajectory is used as a dynamic reference line for the chest cavity sound field, and the spatial position of this trajectory at different time points is used as anchor points to correspondingly mark the acoustic energy distribution state of each region of the chest cavity, giving the chest cavity acoustic energy distribution map temporal continuity and spatial alignment characteristics. In this way, the reference model for sound wave propagation path can be dynamically adjusted after sudden changes in body position, allowing the sound wave propagation path to be corrected and optimized based on this temporal and spatial reference in subsequent path reconstruction and sound field stabilization processing. At this point, the chest cavity acoustic energy distribution map not only reflects the spatial dynamic process of gas layer migration, but also provides a unified temporal coordinate and spatial reference framework for subsequent layered mapping of sound wave propagation paths and extraction of energy interference features. Through this process, the dynamic characteristics of acoustic energy in the chest cavity after changes in body position in pneumothorax patients can be accurately captured and characterized, thereby providing stable and reliable basic sound field information for voiceprint feature analysis, and thus improving the accuracy of pneumothorax lesion range identification and disease severity assessment.

[0025] The above process, through continuous acquisition, joint analysis, temporal tracking, and spatial calibration, realizes a complete technical path from sudden changes in body position to chest cavity acoustic energy distribution and then to drift trajectory extraction. This enables the chest cavity acoustic field to be accurately reconstructed under dynamic conditions, providing sufficient basic conditions for subsequent acoustic field stabilization and disease analysis.

[0026] Step 2: Using the spatial drift trajectory obtained from the chest cavity acoustic energy distribution map, perform time series layered analysis on the chest cavity acoustic energy data, separate and map the sound wave propagation paths in different time periods, and extract the energy interference features of the multi-path superposition area based on the separation and mapping results, providing a positioning basis for the stabilization processing of the chest cavity sound field. The specific implementation method for this step is as follows: Based on the obtained spatial drift trajectory, the chest cavity acoustic energy distribution map is layered according to time series. Taking the instant of sudden change in body position as the starting point, the continuous change of chest cavity acoustic energy is segmented along the time axis, so that each time layer corresponds to a stable sound field state during the gas layer migration process. In this process, the spatial drift trajectory is used as the dominant basis for time layering. Its positional changes at different time points are used to determine the boundary of each time layer, so that there is both spatial continuity and distinguishability of acoustic energy state between adjacent time layers. In this way, the dynamic evolution of chest cavity acoustic energy can be transformed into a series of independently resolvable sound field layers in the time dimension. Each layer reflects the spatial distribution characteristics of the chest cavity sound wave propagation path within a specific time period. The time series layered data processed in this way not only preserves the spatiotemporal evolution characteristics of the chest cavity sound field, but also provides a clear time frame for subsequent path separation.

[0027] After completing the temporal stratification of the chest cavity acoustic energy, the sound wave propagation paths within each time layer are spatially mapped and separated. Specifically, the acoustic energy distribution state of each time layer in the chest cavity acoustic energy distribution map is correlated with the corresponding position of the spatial drift trajectory. Using the drift trajectory as a spatial reference line, the sound wave propagation paths within the chest cavity are identified and mapped layer by layer. Since gas layer migration causes some propagation paths to overlap and revert, by continuously tracking the spatial gradient of the acoustic energy distribution within the same time layer, the main propagation path and secondary propagation paths formed by reflection and scattering can be clearly distinguished. For propagation paths with spatially overlapping regions, by comparing their acoustic energy amplitude variation trends and spatial correspondence with the drift trajectory, these paths can be separated in the mapping, so that each propagation path has an independent spatial expression within the same time layer. The path separation mapping results obtained in this way can effectively reveal the independent distribution law of chest cavity acoustic wave propagation in different time layers, providing an accurate spatial basis for the subsequent identification of energy interference characteristics.

[0028] After separating and mapping the sound wave propagation paths, energy interference features are extracted and described for the overlapping regions between different paths within the thoracic cavity. By performing spatial superposition analysis on the separation and mapping results, local energy concentration phenomena in the intersection regions of different propagation paths within the thoracic cavity are identified. Since these overlapping regions often correspond to multipath interference structures formed during the migration of the free gas layer, their sound energy exhibits periodic enhancement or attenuation characteristics in local space. In this process, using the sound energy change trend at the same spatial location between adjacent time layers as a reference, regions with high energy change rates within continuous time layers are identified as potential energy interference regions. In this way, a set of clear interference feature regions can be formed on the spatial distribution map, each region reflecting the interaction between different propagation paths of sound waves within the thoracic cavity. Furthermore, by comparing the continuity of these energy interference features in the time dimension, the influence range of gas layer migration on the sound wave propagation structure can be inferred, providing benchmark information for subsequent sound field stabilization processing.

[0029] After obtaining the energy interference features, these features are fused with the spatial drift trajectory in the chest cavity acoustic energy distribution map to form the localization basis for stabilizing the chest cavity acoustic field. Specifically, the spatial location of the energy interference region is matched with the time node of the drift trajectory, so that each interference feature corresponds to the motion state of the gas layer in the chest cavity during a specific time period. Through this matching process, the spatiotemporal correlation between the energy interference distribution and the gas layer migration trajectory in the chest cavity acoustic field can be established, thereby realizing the spatial localization of the dynamic changes in the chest cavity acoustic field. This localization basis can not only reveal the formation mechanism of the overlapping areas of acoustic energy in different propagation paths, but also indicate the unstable areas of sound wave propagation in the chest cavity, providing a direct spatial reference for subsequent sound field reflection suppression and propagation path reconstruction. Through the above process, the time series hierarchical analysis, path separation mapping, and energy interference feature extraction of chest cavity acoustic energy are organically combined, enabling the systematic revelation of the complex spatiotemporal changes in chest cavity sound wave propagation after sudden changes in body position in pneumothorax patients, and providing a sufficient technical foundation for achieving dynamic stabilization of the chest cavity acoustic field.

[0030] By implementing the above steps, the entire process of analyzing chest cavity acoustic energy data, from temporal stratification to spatial separation and then to energy interference identification, can be realized, enabling a precise description of the complex changes in the sound field caused by gas layer migration.

[0031] Step 3: Based on the obtained energy interference characteristics, establish a chest cavity sound field reflection suppression model. Transiently constrain the high-energy interference peaks through delayed phase modulation and energy attenuation control, so that the sound wave propagation path in the chest cavity is redistributed, forming a relatively stable sound wave propagation region. The specific implementation method for this step is as follows: Based on the obtained energy interference characteristics, regions with abnormally concentrated energy in the chest cavity sound field are identified and spatially calibrated. By spatially superimposing the energy interference characteristics with the chest cavity acoustic energy distribution map, local areas with abnormally high acoustic energy density within the chest cavity can be clearly identified. These regions typically correspond to multipath reflection overlap areas generated after the migration of the free gas layer. At this point, sound waves within the chest cavity form complex mutual interference between different propagation paths, causing some acoustic energy to accumulate in a short period of time and exhibit a high-energy state in localized areas. To suppress this unbalanced energy distribution, it is necessary to first accurately define the spatial boundaries of these high-energy regions and record their distribution range in the chest cavity coordinate system. This identification process provides the basis for subsequent delay phase modulation and energy attenuation control, enabling subsequent processing to target interference-concentrated regions without affecting other normal propagation paths.

[0032] After spatial calibration of the energy concentration region, phase modulation is applied to the sound wave propagation delay characteristics within the interference region based on the temporal variation patterns in the energy interference features. This causes a controllable phase shift in time of the multipath reflected waves within the chest cavity, thereby reducing their superposition effect. Specifically, the relative propagation order of the primary propagating wave and secondary reflected wave identified in the interference region is fine-tuned using delayed phase modulation. This causes a slight temporal separation in the sound waves that originally arrived and superimposed simultaneously, thus weakening the formation of local energy peaks. This modulation process, based on the propagation delay difference provided by the energy interference features, maintains the continuity of the entire sound field in its spatial structure, ensuring that phase modulation only affects the interference region without disrupting the overall spatiotemporal consistency of the chest cavity sound field. In this way, the superposition energy of local high-energy interference peaks is buffered, making the propagation process of chest cavity sound waves more stable.

[0033] After phase modulation, energy attenuation control is applied to the high-energy portion within the interference region based on the modulated acoustic energy distribution, thereby achieving a redistribution of local acoustic energy. By applying attenuation control to the portion of the energy interference feature where the energy density exceeds the normal acoustic energy threshold, the acoustic energy is gradually released in time and space, preventing excessive energy accumulation in a short period that could lead to sound field imbalance. This energy attenuation control process takes the phase-modulated acoustic energy state as input and, based on the continuity of the energy gradient in each region of the chest cavity acoustic energy distribution map, implements a smooth transition for high-energy regions, allowing energy to diffuse uniformly from the interference center to the surrounding areas. In this way, the local acoustic energy concentration within the chest cavity can be effectively reduced, avoiding repeated reflections of sound waves at the gas layer interface that generate new return paths, thus promoting the overall balance of the chest cavity sound field. Simultaneously, the implementation of energy attenuation control also allows the multipath waves originally present in the interference region to gradually converge to a state of similar energy, creating a balanced energy environment for the redistribution of chest cavity sound wave propagation paths.

[0034] After phase modulation and energy attenuation control, the sound wave propagation path within the chest cavity was reconstructed based on the updated chest cavity acoustic energy distribution map, stabilizing the sound wave propagation within the chest cavity. By comparing the trends in chest cavity acoustic energy distribution before and after energy attenuation processing, it was found that the original high-energy interference region gradually transformed into a region with uniform energy distribution, alleviating the spatial overlap of sound wave propagation paths. At this point, using the updated chest cavity acoustic energy distribution as a reference, the sound wave propagation direction in different regions of the chest cavity was recalibrated, allowing the sound waves, which were originally deflected by the free gas layer, to propagate again according to the acoustic impedance gradient of the chest cavity tissue structure, forming a relatively stable propagation channel. Through this process, the chest cavity sound field transitioned from a multi-path interference state to a stable state with uniform energy distribution and continuous propagation paths. This stable sound field not only facilitates subsequent phase synchronization and amplitude continuity processing of sound wave signals but also provides a reliable acoustic environment for the extraction of acoustic signature features in the lesion area.

[0035] It should be noted that: The chest cavity sound field reflection suppression model is essentially a sound field description and control mechanism that constrains and modulates multipath reflections and energy superposition phenomena within the chest cavity. Its target is the abnormal sound wave propagation region caused by the migration of the free gas layer after a sudden change in body position. Based on energy interference characteristics, the model divides the chest cavity sound field into stable propagation regions and interference enhancement regions, and specifically adjusts the propagation behavior of sound waves within the interference regions. Specifically, the model does not simply filter out sound waves, but rather reorganizes the superposition relationship between sound waves along different paths, considering the propagation path, phase relationship, and energy distribution within the chest cavity, thus suppressing the reflection propagation structure originally formed by multipath reflections. Through this model, the originally complex reflected and superimposed sound field within the chest cavity can be transformed into a stable sound field structure with a clear propagation direction and energy gradient, thereby providing a consistent acoustic environment for subsequent voiceprint feature extraction.

[0036] The establishment of the chest cavity sound field reflection suppression model is based on energy interference characteristics and chest cavity acoustic energy distribution maps. It is achieved by constructing the spatial structure, propagation relationships, and energy states of the interference region layer by layer. First, the location and spatial range of the high-energy interference region within the chest cavity are determined based on the energy interference characteristics, and these regions are designated as the key areas of influence in the model. Simultaneously, the chest cavity acoustic energy distribution map serves as the overall sound field background, forming a spatial coordinate reference system. Based on this, the time delay and spatial distribution relationships between the propagation paths within the interference region are further analyzed, distinguishing between the primary propagation path and secondary reflection paths, and establishing the correspondence between paths, enabling the model to reflect the structural characteristics of multi-path propagation. Subsequently, based on the aforementioned spatial and path relationships, constraint rules for delay phase modulation and energy attenuation control are introduced, mapping the sound wave propagation process within the interference region to a "phase-adjustable, energy-controllable" propagation state. This prevents the sound wave from forming concentrated enhancement during path superposition, instead gradually dispersing along the spatial gradient. Through this process, the model completes the construction from "interference structure identification" to "propagation relationship expression" and then to "energy regulation rule definition", enabling it to comprehensively describe the dynamic behavior of the chest cavity sound field under changes in body position.

[0037] In practical applications, the thoracic acoustic field reflection suppression model primarily functions in the dynamic adjustment phase of the acoustic wave propagation path within the thoracic cavity. First, the model is applied to the identified energy interference region. Based on the path relationships and phase modulation rules established in the model, the propagation timing of acoustic waves within the interference region is fine-tuned, causing the originally synchronously superimposed multipath acoustic waves to separate in time, thereby reducing the degree of local energy accumulation. Subsequently, according to the energy attenuation control rules in the model, spatial diffusion adjustment is implemented in the high-energy region, allowing acoustic energy to gradually transition along the thoracic structure to the surrounding area, avoiding the formation of new reflection superposition centers. Based on this, the adjusted acoustic energy distribution is remapped to the thoracic acoustic field coordinate space, and the acoustic wave propagation path is recalibrated, ensuring it propagates along a direction with relatively gentle acoustic impedance changes, thus forming a stable propagation channel. Through the continuous action of this model, the thoracic acoustic field can maintain a clear propagation path and balanced energy distribution during changes in body position, ensuring the continuity of the acoustic signal in both time and space. This provides a stable foundation for subsequent phase synchronization rearrangement and acoustic energy weighting mapping, ultimately improving the reliability of pneumothorax lesion localization and disease assessment.

[0038] The aforementioned "high-energy interference region" refers to the area where, after a sudden change in the patient's position during pneumothorax, the migration of the free gas layer within the pleural cavity causes a change in the position of the acoustic impedance interface. This results in sound waves being reflected, folded back, and superimposed across multiple propagation paths, creating a region of abnormally concentrated acoustic energy in a localized space. This region typically exhibits phase superposition when sound waves reach the same spatial location along different paths, causing a rapid increase in energy concentration and forming a sound energy density distribution significantly higher than the surrounding area. This high-energy state does not originate from a single propagation path but is the result of the interaction of multiple sound waves, exhibiting spatial locality and temporal transientity, accompanied by phase instability and uneven energy distribution. This region often corresponds to locations where drastic changes occur at the interface between the gas layer and tissue within the pleural cavity; it is a concentrated area of ​​overlapping and folding sound wave propagation paths and a major source of acoustic signal distortion and spurious interference.

[0039] Through the above steps, the chest cavity sound field can be gradually restored from a state of complex interference and energy imbalance to a state of clear propagation path and stable energy distribution after a sudden change in body position. This process, through the orderly connection of spatial calibration, phase modulation, energy attenuation and path reconstruction, effectively suppresses the chest cavity sound field reflection effect, laying a stable acoustic foundation for subsequent phase synchronization rearrangement and acoustic energy weighting mapping of sound wave signals.

[0040] Step 4: Based on the formed sound wave propagation region, the phase sequence of the chest cavity sound wave signal is synchronously rearranged, and the adjacent propagation paths are continuously registered using a time-sliding window to compensate for the temporal misalignment caused by the migration of the free gas layer, thereby maintaining the amplitude and phase continuity of the sound wave signal. The specific implementation method for this step is as follows: Based on the chest cavity acoustic field state after energy reflection suppression, acoustic signal acquisition and phase sequence extraction are performed in the formed stable acoustic wave propagation region. Since gas layer migration causes slight spatial changes in the acoustic wave propagation path in different regions of the chest cavity, it is necessary to divide the chest cavity acoustic energy distribution into zones based on the stable propagation region. The phase sequence of the acoustic signal in each region is extracted through time-series sampling, quantifying the phase change relationship between adjacent regions. This step ensures that the acoustic signal of each propagation region in the overall chest cavity acoustic field is recorded with the same time reference, thus providing a unified time coordinate system for subsequent synchronous rearrangement. Through this process, the phase state of the acoustic waves in each region of the chest cavity within the same time period can be accurately obtained, and a continuous temporal mapping of the chest cavity acoustic field after abrupt changes in body position can be established.

[0041] After extracting the phase sequence of the acoustic signal, preliminary time synchronization processing is performed on the chest cavity acoustic signal based on the phase differences between each propagation region. By comparing the arrival order and phase change direction of acoustic signals in adjacent regions, temporal misalignment regions caused by gas layer migration can be identified. This temporal misalignment manifests as acoustic wave propagation in some regions advancing or lagging behind other regions of the chest cavity, resulting in a discontinuity in the overall acoustic phase sequence. To address this issue, the temporal correspondence between phase sequences is used to adjust the time reference of the acoustic waves in each region, rearranging the acoustic signals from different regions on the time axis to ensure their propagation order aligns with the acoustic wave propagation path within the chest cavity. During this process, the stable propagation region obtained in the previous step serves as a reference frame, ensuring that the synchronization adjustment only affects the local phase shift caused by gas layer migration without disrupting the overall propagation pattern of the chest cavity sound field.

[0042] Building upon the initial synchronization of the phase sequence, a time-sliding window is further utilized to perform continuous registration of adjacent propagation paths. By setting a sliding window on the time axis, the acoustic signals in adjacent regions are gradually aligned within a time range, resulting in a smoother phase transition between adjacent propagation paths. Specifically, as the sliding window moves along the time direction, the phase and amplitude change trends of adjacent propagation paths within the window are continuously compared, ensuring that the phase difference between adjacent paths remains within a stable range during the sliding process. This continuous registration process effectively eliminates subtle time delay differences caused by gas layer migration, allowing the propagation of acoustic waves within the thoracic cavity to exhibit a continuous transition state in the time dimension. Through the dynamic effect of the time-sliding window, the phase jumps originally caused by irregular movement of the thoracic cavity gas layer can be gradually smoothed out, thereby restoring the temporal consistency of the thoracic cavity sound field and ensuring that the propagation sequence of the thoracic cavity acoustic signals is coordinated with the energy transfer path.

[0043] After completing the continuous registration of adjacent propagation paths, the chest cavity acoustic wave signals processed by the time-sliding window are uniformly rearranged in amplitude and phase. Specifically, using the overall acoustic energy distribution of the chest cavity acoustic wave propagation area as a reference, the registered phase sequences are recombined into a continuous acoustic wave time-series signal, ensuring that each propagation path maintains temporal continuity and energy balance in phase and amplitude. In this process, by aligning the acoustic wave signals in the propagation area spatially, each propagation path is temporally connected to its adjacent paths, avoiding signal discontinuities or amplitude abrupt changes caused by transient drift of the gas layer. The synchronously rearranged acoustic wave signal can fully reflect the continuous propagation state of the chest cavity sound field after a change in body position, ensuring that the acoustic energy presents an orderly transmission relationship within the chest cavity space. At this point, the phase continuity and amplitude stability of the chest cavity acoustic wave signal are restored, providing an accurate and repeatable acoustic data basis for subsequent acoustic energy weighting mapping and quantitative assessment of disease severity.

[0044] Through the execution of the above consecutive steps, from acoustic phase sequence extraction, time synchronization processing, continuous registration to phase rearrangement, a complete time-domain compensation and continuous processing workflow for thoracic cavity acoustic signals is formed. This workflow can effectively offset the effects of gas layer migration on the time drift and phase misalignment caused by the propagation of thoracic cavity acoustic waves, ensuring that the thoracic cavity sound field maintains temporal and spatial consistency under dynamic conditions, providing a stable and reliable signal basis for the extraction of acoustic signature features of pneumothorax lesions.

[0045] Step 5: Based on the synchronously rearranged chest cavity acoustic wave signal, construct a chest cavity acoustic energy weight mapping, redistribute the energy weights of different propagation paths to the chest cavity acoustic field coordinate space, so that the acoustic energy distribution after the gas layer migration is stabilized again within the solution range, thereby improving the positioning accuracy of the pneumothorax lesion area and the dynamic consistency of the disease severity assessment results. The specific implementation method for this step is as follows: Based on the synchronously rearranged chest cavity acoustic wave signal, the acoustic energy of different propagation paths is extracted and quantified according to the spatial distribution characteristics of the chest cavity sound field. Since the continuity of chest cavity acoustic waves in time and phase was achieved in the previous stage, the acoustic wave signals of each propagation path have a clear correspondence in spatial coordinates. At this point, based on the chest cavity sound field coordinate space, the acoustic wave amplitude of each propagation path and its energy integral over time are extracted as acoustic energy parameters, so that each propagation path corresponds to an independent energy quantization value. In this way, a multidimensional acoustic energy distribution matrix can be formed in the chest cavity space, reflecting the acoustic energy intensity state at various locations inside the chest cavity after the gas layer migration. Simultaneously, based on the spatial location of the propagation paths determined in the previous step, these acoustic energy parameters are mapped to the chest cavity tissue structure, thereby establishing a spatial mapping basis for chest cavity acoustic energy and providing a data source for subsequent energy weight allocation.

[0046] After obtaining the quantification results of the chest cavity acoustic energy, the energy weights of different propagation paths are redistributed according to the principle of spatial continuity of the chest cavity sound field. This step uses the chest cavity sound field coordinate space as a carrier to adjust the acoustic energy values ​​of each propagation path according to their spatial position relationship. Specifically, when gas layer migration causes an increase in acoustic energy density in some areas and a decrease in acoustic energy density in adjacent areas, the energy weights are redistributed in the spatial coordinates to allow energy to diffuse evenly from excessively high-energy areas to sparsely energy areas, thereby forming a continuous and smooth acoustic energy gradient. This process maintains the overall energy conservation of the chest cavity and improves the spatial distribution balance of acoustic energy. To ensure the accuracy of energy allocation, the phase continuity after synchronous rearrangement is used as a time reference during the energy weight redistribution process to ensure that the change in energy weights corresponds to the time series of sound wave propagation, thereby avoiding interference from temporal inconsistencies in spatial energy mapping. Through this process, local energy abrupt changes in the chest cavity sound field are mitigated, the energy relationship between sound wave propagation paths is re-established, and the spatial balance of the chest cavity sound field is restored.

[0047] After the spatial redistribution of energy weights, a chest cavity acoustic energy weight map is constructed based on the updated acoustic energy state. Specifically, using the chest cavity acoustic field coordinate space as a two-dimensional or three-dimensional reference frame, the energy weights of different propagation paths are marked on corresponding coordinate points according to their spatial positions, forming a weight map that reflects the distribution trend of chest cavity acoustic energy. This weight map not only describes the redistribution of acoustic energy after gas layer migration but also reveals the energy flow direction of sound wave propagation within the chest cavity. By observing the spatial structure of the weight map, the concentration area, transition area, and attenuation area of ​​chest cavity acoustic energy can be intuitively identified, thereby determining the range of chest cavity acoustic field changes caused by gas layer migration. Since the energy weights of each propagation path are synchronized in time and rearranged spatially, the acoustic energy structure presented in the weight map has temporal continuity and spatial stability, which can be used to accurately locate the acoustic energy abnormality area in the chest cavity caused by gas layer migration. Through the establishment of this map, the energy drift caused by multipath interference in the chest cavity acoustic field is effectively constrained, and the acoustic energy distribution is re-stabilized within the solution range, providing an accurate acoustic reference for voiceprint identification of lesion areas.

[0048] After obtaining the chest cavity acoustic energy weighting map, this map is used as an energy reference frame for the chest cavity acoustic field to locate the pneumothorax lesion and dynamically assess its severity. Specifically, the chest cavity acoustic energy weighting map is spatially compared with the previous chest cavity acoustic energy distribution map to identify areas where the energy weight is consistently below normal levels. These areas are then correlated with the gas layer migration trajectory to determine the location and extent of the lesion. Analysis of the energy change rate in the weighting map further indicates the degree of compression and re-expansion trend of the lesion area. When gas layer migration causes a continuous decrease in local acoustic energy weight that fails to return to normal after redistribution, it indicates severe lung tissue collapse in that area; conversely, areas where the energy weight recovers quickly and is evenly distributed reflect better lung tissue re-expansion. In this way, the chest cavity acoustic energy weighting map is not only used for spatial localization but also serves as a dynamic basis for assessing the severity of the condition, ensuring consistency and continuity of pneumothorax detection results across different body positions and time periods.

[0049] Through the implementation of the above steps, from acoustic energy extraction and quantification, energy weight redistribution, weight mapping construction to lesion localization and assessment, the energy reconstruction process of the thoracic cavity sound field after changes in body position was realized. This process effectively solves the problem of acoustic energy imbalance caused by the migration of the free gas layer, stabilizing the thoracic cavity sound field energy back within the calculation range, ensuring the coordination between the sound wave propagation path and energy distribution, thereby improving the spatial localization accuracy of the pneumothorax lesion area and the dynamic reliability of the assessment of the severity of the condition.

[0050] This invention achieves real-time capture and spatiotemporal correlation of dynamic changes in the thoracic cavity sound field by constructing a chest cavity acoustic energy distribution map after a sudden change in body position and extracting the spatial drift trajectory of the concentrated acoustic energy region. This allows for the quantitative expression of changes in the sound wave propagation path caused by the migration of the free gas layer within the thoracic cavity. By performing time-series hierarchical analysis and energy interference feature extraction on the sound wave propagation path, the dynamic instability of the sound field is transformed into a controllable spatial localization process. This effectively reduces the overlap and energy drift of sound wave reflections caused by changes in body position, ensuring continuous and stable voiceprint detection under complex body position conditions and improving the accuracy and repeatability of acoustic identification of pneumothorax lesions.

[0051] This invention establishes a chest cavity sound field reflex suppression model and implements phase synchronization rearrangement and energy weight mapping to redistribute chest cavity sound wave propagation in a dynamic environment and maintain energy equilibrium, ensuring the continuity of sound waves in time and space. The chest cavity sound field processed by this method possesses stable propagation characteristics, maintaining a stable sound energy distribution during gas layer migration. This enables high-precision localization of the lesion area and dynamic, consistent assessment of the severity of the condition, providing doctors with quantifiable and continuous evidence of disease changes and improving the diagnostic reliability of pneumothorax detection.

[0052] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for detecting the severity of pneumothorax in patients based on voiceprint features, characterized in that, Includes the following steps: Step 1: After a sudden change in the body position of a pneumothorax patient, a chest cavity acoustic energy distribution map is constructed based on the real-time monitoring signal of the chest cavity acoustic field. By jointly analyzing the sound wave propagation delay and amplitude gradient, the spatial drift trajectory of the acoustic energy concentration area in the chest cavity is extracted. Step 2: Using the spatial drift trajectory obtained from the chest cavity acoustic energy distribution map, perform time series hierarchical analysis on the chest cavity acoustic energy data, separate and map the sound wave propagation paths in different time periods, and extract the energy interference features of the multi-path superposition area based on the separation and mapping results. Step 3: Based on the obtained energy interference characteristics, establish a chest cavity sound field reflection suppression model. Transiently constrain the high-energy interference peak through delayed phase modulation and energy attenuation control, so that the sound wave propagation path in the chest cavity is redistributed, forming a sound wave propagation region. Step 4: Based on the formed sound wave propagation region, the phase sequence of the chest cavity sound wave signal is synchronously rearranged, and the adjacent propagation paths are continuously registered using a time-sliding window to compensate for the time-domain misalignment caused by the migration of the free gas layer. Step 5: Based on the synchronously rearranged chest cavity acoustic wave signal, construct a chest cavity acoustic energy weight mapping, redistribute the energy weights of different propagation paths to the chest cavity acoustic field coordinate space, so that the acoustic energy distribution after the gas layer migration is stabilized again within the solution range.

2. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 1, characterized in that, The steps for constructing a chest cavity acoustic energy distribution map and extracting the spatial drift trajectory of the acoustic energy concentration area within the chest cavity after a sudden change in the body position of a pneumothorax patient include: The system continuously collects respiratory sounds, vesicular sounds, and chest resonance signals from the chest wall surface using a multi-point acoustic sensor array attached to different areas of the patient's chest. The system also performs delayed synchronization processing on the signals from each collection point, using the instant of sudden change in body position as the time reference. After time synchronization is completed, the propagation delay and amplitude gradient of sound waves at each acquisition point are jointly analyzed to establish a preliminary spatial mapping relationship of chest cavity sound energy in order to identify energy concentration areas and energy sparse areas caused by gas layer migration. Based on the temporal changes in the acoustic energy distribution in the chest cavity, the continuous time frames are analyzed frame by frame to track the displacement changes of the region with the highest energy density in spatial coordinates and extract the spatial drift trajectory of the region where the acoustic energy in the chest cavity is concentrated. By combining the drift trajectory with the overall structure of the chest cavity acoustic energy distribution map, a temporal and spatial reference for reconstructing the chest cavity acoustic wave propagation path is formed.

3. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 2, characterized in that, When extracting the spatial drift trajectory of the chest cavity acoustic energy concentration area, the energy centroid correlation method is used between adjacent time frames, and the acoustic energy concentration center of the previous time frame is used as the reference point of the current time frame to ensure the continuity of the drift trajectory in the time dimension and the spatial correspondence.

4. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 2, characterized in that, The steps for performing time-series hierarchical analysis of chest cavity acoustic energy data using the spatial drift trajectory obtained from the chest cavity acoustic energy distribution map include: Based on the obtained spatial drift trajectory, the chest cavity acoustic energy distribution map is layered along the time axis so that each time layer corresponds to a stable acoustic field state during the gas layer migration process, and the time layer boundary is determined by the position change of the spatial drift trajectory. Spatial mapping and separation of the sound wave propagation path within each time layer are performed, and the drift trajectory is used as a spatial reference line to distinguish between the main propagation path and the secondary propagation path, thereby obtaining the separation and mapping results of the chest cavity sound wave propagation. Based on the separation mapping results, spatial overlay analysis is performed on the overlapping areas between different paths to extract the energy interference features of chest cavity acoustic energy and identify local energy concentration areas; The obtained energy interference features are fused with the spatial drift trajectory in the chest cavity acoustic energy distribution map to establish the spatiotemporal correlation between the energy interference distribution and the gas layer migration trajectory, forming the positioning basis for the chest cavity acoustic field stabilization processing.

5. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 4, characterized in that, When fusing energy interference features with the spatial drift trajectory in the chest cavity acoustic energy distribution map, the spatial location of the energy interference region is matched with the time node of the drift trajectory, so that each energy interference feature corresponds to the motion state of the gas layer in the chest cavity during a specific time period, forming a spatial positioning result of the dynamic change of the chest cavity acoustic field.

6. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 4, characterized in that, The steps for establishing a chest cavity sound field refraction suppression model based on the obtained energy interference characteristics include: Based on the obtained energy interference characteristics, the regions with abnormally concentrated energy in the chest cavity sound field are identified and spatially calibrated, and the distribution range of the high-energy interference region in the chest cavity coordinate system is determined. Based on the temporal variation law in the energy interference characteristics, the propagation delay characteristics of sound waves in the high-energy interference region are phase-modulated to make the multi-path reflected waves produce a controllable phase shift in time, thereby weakening the formation of local energy peaks. Based on the acoustic energy distribution results after phase modulation, energy attenuation control is applied to the high-energy portion in the interference region to ensure that the local acoustic energy diffuses uniformly in time and space. Based on the chest cavity acoustic energy distribution map after energy attenuation control, the chest cavity sound wave propagation path is reconstructed, so that the chest cavity sound field changes from a multi-path interference state to a stable state with balanced energy distribution and continuous propagation path.

7. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 6, characterized in that, During the energy attenuation control process, based on the continuity of the energy gradient in each region of the chest cavity acoustic energy distribution map, a smooth transition is implemented in the region where the energy density exceeds the normal acoustic energy threshold, so that the energy diffuses evenly from the interference center to the surrounding region. When reconstructing the propagation path of chest cavity acoustic waves, energy redistribution is achieved, so that the propagation of acoustic waves forms a stable channel along the acoustic impedance gradient of chest cavity tissue.

8. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 6, characterized in that, The steps for synchronously rearranging the phase sequence of the chest cavity sound wave signal based on the formed sound wave propagation region include: Based on the chest cavity sound field state after energy return suppression, the sound wave signal is collected in a partition and the phase sequence is extracted in the stable sound wave propagation area. A time-continuous mapping of the chest cavity sound field is established with a unified time reference. Based on the phase difference between different propagation regions, the time reference of the chest cavity sound wave signal is adjusted so that the propagation sequence of sound waves in different regions is consistent with the propagation path of chest cavity sound waves, thus compensating for the local phase shift caused by gas layer migration. By using a time-sliding window to perform continuous registration on adjacent propagation paths, the phase difference between adjacent paths remains stable during the time-sliding process; The amplitude and phase of the chest cavity acoustic wave signal after time-sliding window processing are uniformly rearranged to maintain the continuity and energy balance of the chest cavity acoustic wave signal in time and space.

9. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 8, characterized in that, During the continuous registration process of the time-sliding window, the phase trend and amplitude change trend of the chest cavity sound wave propagation area are used as the registration basis. By dynamically adjusting the time step of the sliding window, the phase connection of adjacent propagation paths on the time axis is kept smooth.

10. The method for detecting the severity of pneumothorax based on voiceprint features according to claim 8, characterized in that, The steps for constructing a chest cavity acoustic energy weighting map based on the synchronously rearranged chest cavity acoustic wave signal include: Based on the chest cavity acoustic wave signal after synchronous rearrangement, the acoustic energy of different propagation paths is extracted and quantified according to the spatial distribution characteristics of the chest cavity acoustic field, and a spatial mapping basis for chest cavity acoustic energy is established. Based on the principle of spatial continuity of the chest cavity sound field, the energy weights of different propagation paths are redistributed so that energy diffuses evenly from high-energy regions to energy-sparse regions, forming a continuous acoustic energy gradient and maintaining the correspondence between energy weights and time series. Based on the updated acoustic energy state, a chest cavity acoustic energy weight mapping is constructed, and the energy weights of each propagation path are marked in the chest cavity acoustic field coordinate space to form an acoustic energy weight mapping diagram. Using thoracic acoustic energy weighting as an energy reference framework, the location and severity of pneumothorax lesions are dynamically assessed.

Images