Secretion detection method and device and visible and adjustable temperature measurement trachea cannula

Abstract

The invention provides a secretion detection method and device and a visible and adjustable temperature measurement trachea cannula, and belongs to the technical field of tracheal catheters.The secretion detection method comprises the steps that a target image frame set in a target time period where a moment is located and target breathing phase data in the target time period are obtained on the basis of a to-be-detected image; and performing secretion boundary segmentation on the obtained to-be-detected image again to obtain a target secretion image, and determining the amount of the secretion in the trachea, so that under the condition that the secretion boundary segmentation on the to-be-detected image is not accurate, the image frame set and the respiration phase data in the target time period are comprehensively considered, the influence of respiration on the image form is identified, and the accuracy of the detection is improved. The to-be-detected image is re-segmented to obtain a more accurate target secretion image, so that the problem of inaccurate secretion form detection of the image in the prior art is effectively solved, and the effect of obtaining the accurate secretion amount according to the image is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of tracheal catheters, and particularly to a method and device for detecting secretions and a visible and adjustable temperature measuring tracheal intubation tube. Background Art

[0002] Tracheal catheters, such as bronchial catheters, are mainly used to ensure that at least a part of the patient's respiratory system or lungs can be effectively ventilated. These catheters can be safely inserted into the patient's respiratory tract through non-invasive methods, such as through the mouth or nose.

[0003] However, during the operation, the patient may produce various secretions due to various reasons such as inflammatory reactions, allergic reactions or surgical traumas. These secretions are particularly difficult to detect when the patient is unconscious. If the secretions accumulate too much, they may stimulate the patient's respiratory system, triggering reflexive reactions such as coughing and sneezing, which will not only interfere with the smooth progress of related operations, but may also have an adverse impact on the patient's recovery. Therefore, the operator needs to closely monitor the patient's respiratory tract condition and remove the secretions in time. Due to the harsh imaging environment and the influence of respiratory airflow, the detection result of the secretions image obtained by single imaging has a certain contingency, and it is difficult to determine the actual amount of secretions. Summary of the Invention

[0004] The present invention provides a method and device for detecting secretions and a visible and adjustable temperature measuring tracheal intubation tube, which are used to solve the defect that the detection of the morphology of secretions in the image in the prior art is inaccurate, and achieve the effect of obtaining the accurate amount of secretions according to the image.

[0005] The present invention provides a method for detecting secretions, including:

[0006] Performing segmentation on the boundaries of secretions in the obtained image to be detected to obtain an initial secretions image;

[0007] Based on the initial secretions image, determining an initial cumulative amount level of the secretions; the cumulative amount level is used to evaluate the amount of secretions in the trachea;

[0008] In the case where the initial cumulative amount level indicates that the amount of secretions in the trachea is less than a preset cumulative amount, determining the segmentation confidence of the initial secretions image; the segmentation confidence is used to measure the accuracy of the segmentation of the boundaries of secretions in the image to be detected;

[0009] In the case where the segmentation confidence is less than a preset confidence, based on the set of target image frames in the target time period where the time of obtaining the image to be detected is located and the target respiratory phase data in the target time period, re-performing segmentation on the boundaries of secretions in the obtained image to be detected to obtain a target secretions image;

[0010] Determine the amount of secretions in the trachea based on the target secretion image.

[0011] According to a secretion detection method provided by the present invention, the method for re-segmenting the secretion boundary of the acquired image to be detected to obtain a target secretion image based on the target image frame set within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period includes:

[0012] Based on the target image frame set within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period, determine the target motion characteristics of the secretions in the trachea within the target time period; the motion characteristics are used to represent the morphological changes of the tracheal secretions under the action of the respiratory airflow.

[0013] Based on the target motion characteristics and the reference image in the target image frame set, determine the target image corresponding to the image to be detected; the reference image is determined based on the segmentation confidence of the secretion image obtained by segmenting the images in the target image frame set.

[0014] Perform secretion boundary segmentation on the target image to obtain the target secretion image.

[0015] According to a secretion detection method provided by the present invention, the method for determining the target motion characteristics of the tracheal secretions within the target time period based on the target image frame set within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period includes:

[0016] Based on the first image set acquired before the acquisition time of the image to be detected in the target image frame set and the first respiratory phase data corresponding to the first image set in the target respiratory phase data, obtain the first motion characteristics of the secretions.

[0017] Based on the second image set acquired after the acquisition time of the image to be detected in the target image frame set and the second respiratory phase data corresponding to the second image set in the target respiratory phase data, obtain the second motion characteristics of the secretions.

[0018] Based on the first motion characteristics and the second motion characteristics, determine the target motion characteristics.

[0019] According to a secretion detection method provided by the present invention, the motion characteristics of the secretions are determined by the following method:

[0020] Based on the target respiratory phase data, align each frame image in the target image frame set with the respiratory phase to determine the pixel displacement field of the images in the target image frame set; the pixel displacement field is calculated according to different smoothing coefficients corresponding to different respiratory phases;

[0021] Based on the pixel displacement field and the target respiratory phase data, statistically analyze the main direction of secretion movement and the regional activity; the regional activity is used to evaluate the movement intensity and distribution of secretions in the trachea;

[0022] Based on the empirical data of the movement characteristics of secretions, adjust the main direction of secretion movement and the regional activity, and determine the adjusted main direction of secretion movement and the regional activity as the movement characteristics of the secretions.

[0023] According to a secretion detection method provided by the present invention, determining the target image corresponding to the image to be detected based on the target movement characteristics and the reference image in the target image frame set includes:

[0024] Perform morphological enhancement on the image to be detected based on the target movement characteristics and the reference image in the target image frame set to obtain the target image.

[0025] According to a secretion detection method provided by the present invention, the segmentation confidence is determined based on the pixel value variance and / or boundary sharpness of the secretion region.

[0026] The present invention also provides a secretion detection device, including:

[0027] A first segmentation module for performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image;

[0028] A first processing module for determining the initial cumulative amount level of the secretion based on the initial secretion image; the cumulative amount level is used to evaluate the amount of secretions in the trachea;

[0029] A second processing module for determining the segmentation confidence of the initial secretion image when the initial cumulative amount level indicates that the amount of secretions in the trachea is less than a preset cumulative amount; the segmentation confidence is used to measure the accuracy of performing secretion boundary segmentation on the image to be detected;

[0030] A second segmentation module for, when the segmentation confidence is less than a preset confidence, re-performing secretion boundary segmentation on the acquired image to be detected based on the target image frame set within the target time period where the time when the image to be detected is acquired is located and the target respiratory phase data within the target time period to obtain a target secretion image;

[0031] A third processing module, configured to determine the amount of tracheal secretions based on the target secretion image.

[0032] The present invention also provides a visual adjustable temperature measuring endotracheal tube, including: a main body tube, a temperature sensing module, an image acquisition module, and a cuff; an adjusting member for adjusting the bending angle of the main body tube is further provided inside the main body tube; one end of the main body tube is provided with a first opening, the image acquisition module is disposed inside the first opening, and the temperature sensing module is disposed inside the cuff;

[0033] The temperature sensing module is configured to acquire temperature, and the image acquisition module is configured to collect a to-be-detected image in the trachea to obtain the amount of tracheal secretions; the amount of tracheal secretions is determined based on the to-be-detected image processed by any one of the above-mentioned secretion detection methods.

[0034] According to the visual adjustable temperature measuring endotracheal tube provided by the present invention, a sputum suction ring is sleeved on the main body tube, the sputum suction ring is located at one end of the cuff away from the image acquisition module, the sputum suction ring includes a chamber structure with a hollow interior for absorbing sputum, and a plurality of sputum suction holes are formed on the sputum suction ring.

[0035] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, any one of the above-mentioned secretion detection methods is implemented.

[0036] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, any one of the above-mentioned secretion detection methods is implemented.

[0037] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, any one of the above-mentioned secretion detection methods is implemented.

[0038] The secretion detection method, device, and visual adjustable temperature measuring endotracheal tube provided by the present invention, by comprehensively considering the set of image frames and respiratory phase data in a target period and combining the target image frame set and target respiratory phase data in the target period to identify the influence of respiration on the image morphology when the secretion boundary segmentation of the to-be-detected image is inaccurate, and re-segmenting the to-be-detected image to obtain a more accurate target secretion image, can effectively solve the problem of inaccurate detection of the secretion morphology of the image in the prior art, and achieve the effect of obtaining an accurate amount of secretions according to the image. Description of the Drawings

[0039] To more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0040] Figure 1 is one of the schematic flowcharts of the secretion detection method provided by the present invention;

[0041] Figure 2 is the second schematic flowchart of the secretion detection method provided by the present invention;

[0042] Figure 3 is the schematic structural diagram of the secretion detection device provided by the present invention;

[0043] Figure 4 is one of the schematic structural diagrams of the visible adjustable temperature measuring endotracheal intubation provided by the present invention;

[0044] Figure 5 is the second schematic structural diagram of the visible adjustable temperature measuring endotracheal intubation provided by the present invention;

[0045] Figure 6 is the schematic structural diagram of the electronic device provided by the present invention. Detailed Embodiments

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0047] The following combines Figures 1 - 5 to describe the secretion detection method, device, and visible adjustable temperature measuring endotracheal intubation of the present invention.

[0048] As Figure 1 , the secretion detection method of the embodiments of the present invention mainly includes step 110, step 120, step 130, step 140, and step 150.

[0049] In step 110, the boundary of the secretion in the acquired image to be detected is segmented to obtain an initial secretion image.

[0050] The boundary of the tracheal secretion can be segmented from the acquired image to be detected, so as to obtain an initial secretion image that only contains the secretion.

[0051] A variety of image segmentation techniques can be adopted, such as threshold-based methods, edge detection algorithms, region growing algorithms, etc. Considering the characteristics of tracheal secretions and the complexity of the images, more advanced segmentation algorithms need to be selected, such as semantic segmentation methods based on deep learning, to improve the accuracy and robustness of segmentation.

[0052] It can be understood that through accurate boundary segmentation, interference information in subsequent analysis can be reduced, the accuracy of extracting secretion characteristics can be improved, and a reliable basis can be provided for subsequent cumulative amount evaluation and motion feature analysis.

[0053] Step 120: Based on the initial secretion image, determine the initial cumulative amount level of the secretion; the cumulative amount level is used to evaluate the amount of tracheal secretions.

[0054] After obtaining the initial secretion image, it is necessary to evaluate the amount of secretions it represents and classify it into a specific cumulative amount level. This cumulative amount level is used to preliminarily judge whether the amount of tracheal secretions has reached the level that requires attention or treatment.

[0055] The initial cumulative amount level can be determined by calculating quantitative indicators such as the area and volume occupied by the secretions in the initial secretion image and comparing them with preset thresholds. For example, according to the proportion of the secretions in the image, it can be divided into three levels: low, medium, and high.

[0056] Determining the initial cumulative amount level helps to quickly screen out situations that require further attention, avoid complex subsequent processing for all samples, and improve the efficiency of the entire detection process.

[0057] Step 130: In the case where the initial cumulative amount level indicates that the amount of tracheal secretions is less than the preset cumulative amount, determine the segmentation confidence of the initial secretion image.

[0058] The segmentation confidence is used to measure the accuracy of the secretion boundary segmentation for the image to be detected.

[0059] When the initially evaluated amount of secretions does not reach the preset processing threshold, it is necessary to further determine the accuracy of the previous secretion boundary segmentation, that is, the segmentation confidence. This step is to ensure the reliability of the detection even when the detected amount of secretions is small.

[0060] The segmentation confidence can be determined in various ways, such as calculating indicators such as the stability and consistency of the segmentation results. For example, by comparing the segmentation results of different segmentation algorithms for the same image and evaluating their consistency to determine the confidence.

[0061] In some embodiments, the segmentation confidence is determined based on the pixel value variance and / or boundary clarity of the secretion region.

[0062] In some implementations, the method evaluates the accuracy of segmentation by calculating the consistency of pixel values inside the segmented secretion region. The higher the consistency of pixel values, the more accurate the segmentation and the higher the confidence. For the segmented secretion region, calculate the variance of all pixel values within the region. The smaller the variance, the more consistent the pixel values and the more accurate the segmentation. Determine a variance threshold based on experience or experiments. When the calculated variance is less than the threshold, the segmentation is considered accurate and the confidence is high; otherwise, the confidence is low.

[0063] In some implementations, the boundary sharpness can also be evaluated by analyzing the smoothness of the segmented boundary. A smooth boundary usually conforms more to the actual shape of the secretion, so it can be used as a basis for evaluating the segmentation accuracy.

[0064] Indicators such as the curvature of the curve and the continuity of boundary pixels can be used to measure the smoothness of the segmented boundary. For example, calculate the change in the curvature of boundary pixels. The smaller the change in curvature, the smoother the boundary. Set a threshold based on the smoothness indicator. When the smoothness is higher than the threshold, the segmentation is considered accurate and the confidence is high, which can effectively identify the rationality of the segmented boundary, avoid problems such as boundary serration or mutation caused by the segmentation algorithm, and improve the accuracy and stability of the entire detection system.

[0065] It can be understood that by evaluating the segmentation confidence, quality control can be performed on the segmentation results, avoiding misjudgments caused by inaccurate segmentation, and improving the reliability and accuracy of the entire detection system.

[0066] The accuracy of segmentation can also be comprehensively evaluated by combining the two aspects of regional consistency and boundary sharpness, so as to obtain a more comprehensive and reliable segmentation confidence. Perform weighted fusion on the evaluation results of regional consistency and boundary sharpness to obtain a comprehensive segmentation confidence index.

[0067] Step 140, in the case where the segmentation confidence is less than the preset confidence, based on the set of target image frames in the target time period where the time of the image to be detected is located and the target respiratory phase data in the target time period, re-segment the boundary of the secretion of the obtained image to be detected to obtain a target secretion image.

[0068] If it is found that the previous segmentation confidence is low, that is, there may be a large error in the segmentation result, then more relevant information needs to be used, such as multiple image frames in the target time period and the corresponding respiratory phase data, to re-segment the secretion boundary to obtain a more accurate target secretion image.

[0069] The information of multiple frames of images can be combined, such as through time series analysis and image analysis algorithms, to better capture the movement characteristics and boundary changes of secretions. At the same time, using respiratory phase data, more accurate segmentation can be performed at specific respiratory stages to reduce errors caused by respiratory movement.

[0070] It can be understood that re-segmentation can improve the accuracy of the secretion boundary. Especially in complex or dynamic scenarios, it ensures that a reliable secretion image can be obtained even when the initial segmentation is not ideal, providing a guarantee for subsequent accurate detection.

[0071] Step 150: Determine the amount of tracheal secretions based on the target secretion image.

[0072] According to the target secretion image obtained through re-segmentation, accurately determine the actual amount of tracheal secretions. Similar to step 120, the amount of secretions can be accurately calculated by quantifying relevant features in the target secretion image, such as area, volume, etc. At the same time, other historical experience data and parameters can be combined to correct and verify the results, improving the accuracy of detection.

[0073] In this embodiment, by automatically and accurately detecting the amount of tracheal secretions at the intubation position, the amount of tracheal secretions can be automatically identified and determined, so that automatic suction can be performed when necessary, avoiding unnecessary high-frequency suction operations from irritating the trachea. Especially during the intraoperative process, there is no need to set up dedicated operators to continuously check the secretion situation, saving personnel allocation and improving operation efficiency.

[0074] According to the secretion detection method provided by the embodiments of the present invention, when the secretion boundary segmentation of the image to be detected is inaccurate, by comprehensively considering the set of image frames and respiratory phase data within the target time period, and combining the set of target image frames and target respiratory phase data within the target time period, identifying the influence of respiration on the image morphology, and re-segmenting the image to be detected to obtain a more accurate target secretion image, it can effectively solve the problem of inaccurate detection of the secretion morphology in the existing technology and achieve the effect of obtaining an accurate amount of secretions according to the image.

[0075] In some embodiments, as Figure 2 shown, based on the set of target image frames and target respiratory phase data within the target time period where the moment of obtaining the image to be detected is located, re-segment the obtained image to be detected for the secretion boundary to obtain a target secretion image, including step 210, step 220, and step 230.

[0076] Step 210: Based on the set of target image frames within the target time period when the image to be detected is acquired and the target respiratory phase data within the target time period, determine the target motion characteristics of the tracheal secretions within the target time period.

[0077] The motion characteristics are used to represent the morphological changes of the tracheal secretions under the action of the respiratory airflow.

[0078] Within the target time period when the image to be detected is acquired, collect multiple target image frames, and in combination with the target respiratory phase data within this time period, analyze the morphological changes of the tracheal secretions under the action of the respiratory airflow, so as to determine its target motion characteristics.

[0079] Within the target time period, multiple image frames can be acquired at a certain frame rate to form a set of target image frames. Obtain the target respiratory phase data within the target time period through devices such as respiratory signal acquisition sensors, including information on the inhalation phase and exhalation phase. For example, the respiratory waveform can be obtained in real time through a thoracic impedance sensor (sampling rate 200Hz), and the time stamp can be recorded.

[0080] On this basis, algorithms such as optical flow algorithm and frame difference method can be used in combination with the respiratory phase data to analyze the motion changes of the secretions between different frames, and extract features such as the main direction of motion and motion energy.

[0081] In this embodiment, by comprehensively considering multiple image frames and respiratory phase data, the motion information of the secretions can be captured more comprehensively, improving the accuracy and reliability of the motion characteristics.

[0082] Step 220: Based on the target motion characteristics and the reference image in the set of target image frames, determine the target image corresponding to the image to be detected.

[0083] The reference image is determined based on the segmentation confidence of the secretion image obtained after segmenting the secretions in the images in the set of target image frames.

[0084] The previously extracted target motion characteristics and the reference image in the set of target image frames can be used to determine the target image that best matches or is most representative of the image to be detected.

[0085] Select several images from the set of target image frames for secretion segmentation, calculate the segmentation confidence of each image, and select the image with a higher confidence as the reference image.

[0086] According to the target motion characteristics, match the image to be detected with the reference image, and the target image that is most similar to the image to be detected in terms of motion characteristics can be found.

[0087] Alternatively, based on the motion features in the reference image, the morphological features of the secretions in the image to be detected can be simulated, and then the image to be detected can be modified to obtain the target image.

[0088] It can be understood that with the assistance of the reference image, the position and state of the image to be detected in the target time period can be better determined, providing a more accurate basis for subsequent re-segmentation.

[0089] Determining the target image corresponding to the image to be detected based on the target motion features and the reference images in the target image frame set includes: performing morphological enhancement on the image to be detected based on the target motion features and the reference images in the target image frame set to obtain the target image.

[0090] The previously determined motion features of the secretion target (such as the main motion direction, regional activity, etc.) and the reference images in the target image frame set can be used to perform morphological enhancement processing on the current image to be detected. The purpose is to highlight the features of the secretion, making it more obvious and easier to segment in the image, so as to obtain a more accurate target image.

[0091] Specifically, the information of the target motion features and the reference image can be combined to perform morphological operations on the current image to be detected, such as dilation, erosion, opening and closing operations, etc., to enhance the morphological features of the secretion, making its boundary clearer and the region more complete.

[0092] Step 230: Perform segmentation on the boundaries of the secretions in the target image to obtain the target secretion image.

[0093] Perform segmentation on the boundaries of the secretions in the determined target image to obtain a more accurate target secretion image. Post-processing operations such as morphological processing and smoothing processing are performed on the segmentation result to further optimize the accuracy of the secretion boundary.

[0094] In this embodiment, by using multiple image frames and respiratory phase data in the target time period, combining motion feature analysis and reference image matching, the target image corresponding to the image to be detected can be determined more accurately, and precise segmentation of the secretion boundary can be performed on it, thereby improving the accuracy of the entire detection process. Especially in complex or dynamic scenarios, it can effectively cope with the morphological changes of secretions and the interference of respiratory motion, improving the robustness and stability of the system.

[0095] In some embodiments, the motion features of the secretion are determined in the following manner.

[0096] First, based on the target respiratory phase data, each frame image in the target image frame set can be aligned with the respiratory phase to determine the pixel displacement field of the images in the target image frame set; the pixel displacement field is calculated according to different smoothing coefficients corresponding to different respiratory phases.

[0097] Further, based on the pixel displacement field and the target respiratory phase data, the main direction of secretion movement and the regional activity are statistically analyzed; the regional activity is used to evaluate the movement intensity and distribution of secretions in the trachea.

[0098] Finally, based on the empirical data of the movement characteristics of secretions, the main direction of secretion movement and the regional activity are adjusted, and the adjusted main direction of secretion movement and regional activity are determined as the movement characteristics of secretions.

[0099] Using the target respiratory phase data, each frame of the target image frame set is aligned with the corresponding respiratory phase, and then the pixel displacement field of each frame of the image is calculated. The pixel displacement field describes the displacement of pixel points in the image under different respiratory phases and reflects the movement information of secretions.

[0100] According to the target respiratory phase data, each frame of the target image frame set is corresponding to a specific respiratory stage, such as the inspiratory phase, the expiratory phase, etc.

[0101] The optical flow algorithm can be used to calculate the pixel displacement field of each frame of the image in combination with different smoothing coefficients corresponding to different respiratory phases. The selection of the smoothing coefficient may be adjusted according to the characteristics of the respiratory phase. For example, a smaller smoothing coefficient is used in the inspiratory phase to retain details, and a larger smoothing coefficient is used in the expiratory phase to suppress noise.

[0102] It can be understood that by aligning the image with the respiratory phase and calculating the pixel displacement field, the movement changes of secretions in different respiratory stages can be accurately captured, providing a basis for subsequent movement feature analysis.

[0103] The main direction of movement represents the main movement trend of secretions under the action of respiratory airflow, while the regional activity is used to evaluate the movement intensity and distribution of secretions in different regions.

[0104] The direction distribution of displacement vectors in the pixel displacement field can be analyzed, and the main movement direction can be determined by methods such as clustering. For example, the average direction or dominant direction of the displacement vectors can be calculated.

[0105] According to the size and distribution of the pixel displacement field, the movement energy or activity of different regions can be calculated. For example, the average value or sum of the magnitudes of the displacement vectors of each region can be calculated to reflect the movement intensity of the region.

[0106] In this case, by statistically analyzing the main direction of movement and the regional activity, the movement characteristics of secretions in the trachea can be comprehensively understood, providing an important basis for subsequent secretion volume evaluation and clinical diagnosis.

[0107] Based on the empirical data of the movement characteristics of the secretion, the statistically obtained main movement direction and regional activity can be adjusted to more accurately reflect the actual movement characteristics of the secretion. In this case, an empirical model or reference standard is established using the accumulated secretion movement characteristic data from the past.

[0108] Compare the statistically obtained main movement direction and regional activity with the empirical data and make adjustments according to the differences. For example, if there is a deviation between the statistical result and the empirical data, adjustments can be made through methods such as weighted average and deviation correction.

[0109] It can be understood that by making adjustments in combination with empirical data, statistical errors and uncertainties can be reduced, the accuracy and reliability of the movement characteristics can be improved, and thus the accurate assessment of the secretion volume can be better supported.

[0110] In this embodiment, by calculating the pixel displacement field based on the target respiratory phase data and making adjustments in combination with empirical data, the movement characteristics of the secretion can be determined more accurately, errors and uncertainties can be reduced, it can adapt to different respiratory stages and changes in the movement of the secretion, and the adaptability and robustness of the entire detection system can be improved. On this basis, the accurate movement characteristics provide a reliable basis for the subsequent assessment of the secretion volume, which helps to achieve more accurate detection of the secretion volume and provides strong support for clinical diagnosis and treatment.

[0111] In some embodiments, based on the set of target image frames within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period, determining the target movement characteristics of the tracheal secretion within the target time period includes the following process.

[0112] First, the first movement characteristics of the secretion can be obtained based on the first image set acquired before the acquisition time of the image to be detected in the set of target image frames and the corresponding first respiratory phase data of the first image set in the target respiratory phase data.

[0113] Using the first image set acquired before the acquisition time of the image to be detected and the corresponding first respiratory phase data of these images, analyze the movement characteristics of the secretion during this time period.

[0114] Specifically, a series of images before the acquisition time of the image to be detected can be screened out from the target time period to form the first image set. Obtain the corresponding first respiratory phase data of the first image set and clarify the respiratory stage of each image, such as the inspiratory phase or the expiratory phase.

[0115] Techniques such as optical flow algorithms can be applied, combined with the first respiratory phase data, to calculate the motion information of the secretions in the first image set, obtain the first motion feature, and provide important preliminary data support for the subsequent comprehensive evaluation of the motion characteristics of the secretions by analyzing the motion of the secretions before the image to be detected.

[0116] Furthermore, based on the second image set obtained after the acquisition time of the image to be detected in the target image frame set and the second respiratory phase data corresponding to the second image set in the target respiratory phase data, the second motion feature of the secretions is obtained.

[0117] The second image set collected after the acquisition time of the image to be detected and the corresponding second respiratory phase data of these images can be used to analyze the motion characteristics of the secretions during this time period.

[0118] A series of images after the acquisition time of the image to be detected are selected from the target time period to form the second image set. The second respiratory phase data corresponding to the second image set is obtained to clarify the respiratory phase of each image. Similarly, techniques such as optical flow algorithms are applied, combined with the second respiratory phase data, to calculate the motion information of the secretions in the second image set, and the second motion feature is obtained.

[0119] In this embodiment, by analyzing the motion of the secretions after the image to be detected, the motion data in the subsequent time period is supplemented, and combined with the previous data, the motion changes of the secretions can be captured more comprehensively.

[0120] Finally, based on the first motion feature and the second motion feature, the target motion feature is determined.

[0121] Taking into account the first motion feature and the second motion feature comprehensively, the overall target motion feature of the tracheal secretions in the target time period is determined.

[0122] In some implementation manners, the first motion feature and the second motion feature can be fused, and a comprehensive motion feature description can be obtained through methods such as weighted average and feature splicing.

[0123] In other implementation manners, the change trends between the first motion feature and the second motion feature can also be analyzed, such as the change of the motion direction, the increase or decrease of the speed, etc., to further refine the target motion feature.

[0124] It can be understood that by integrating the motion features of the two time periods before and after, the motion law of the secretions in the target time period can be described more accurately, providing a more reliable basis for the subsequent segmentation of the secretion boundary and the evaluation of the quantity.

[0125] In some implementations, by integrating high-speed imaging, respiratory dynamics parameters, and optical flow algorithms, the motion characteristics of tracheal secretions can be extracted. First, a high-speed camera can start collecting dynamic image sequences when the thoracic cavity expansion speed is relatively large during the inhalation phase. The respiratory phase can be accurately aligned by synchronously combining a thoracic impedance sensor and ventilator airflow data (error < 2 ms). Subsequently, an improved optical flow method is used to calculate the pixel displacement field through multi-scale pyramid decomposition, where the smoothing coefficient α is dynamically adjusted according to the respiratory phase. For example, α = 0.01 during the inhalation phase to retain details, and α = 0.1 during the exhalation phase to suppress noise. Morphological enhancement is performed on the secretion area. For example, different feature regions can be segmented using the HSV color space and the CLAHE algorithm is combined to enhance the contrast of the flow region. Finally, the main motion direction is statistically analyzed through the displacement vector field, the regional activity is quantified using the motion energy map, and the motion characteristics are calibrated to obtain the target segmentation image with high segmentation and extraction accuracy, which can effectively distinguish real secretion motion from respiratory artifacts.

[0126] The secretion detection device provided by the present invention will be described below. The secretion detection device described below can be correspondingly referred to the secretion detection method described above.

[0127] As Figure 3 shown, the secretion detection device according to an embodiment of the present invention mainly includes a first segmentation module 310, a first processing module 320, a second processing module 330, a second segmentation module 340, and a third processing module 350.

[0128] The first segmentation module 310 is configured to perform secretion boundary segmentation on the acquired image to be detected, and obtain an initial secretion image;

[0129] The first processing module 320 is configured to determine an initial cumulative amount level of the secretion based on the initial secretion image; the cumulative amount level is used to evaluate the amount of tracheal secretions;

[0130] The second processing module 330 is configured to determine a segmentation confidence level of the initial secretion image when the amount of tracheal secretions indicated by the initial cumulative amount level is less than a preset cumulative amount; the segmentation confidence level is used to measure the accuracy of the secretion boundary segmentation of the image to be detected.

[0131] The second segmentation module 340 is configured to, when the segmentation confidence level is less than a preset confidence level, obtain a set of target image frames within a target time period where the time of the image to be detected is located and target respiratory phase data within the target time period, and re-perform secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image;

[0132] The third processing module 350 is configured to determine the amount of tracheal secretions based on the target secretion image.

[0133] According to the secretion detection device provided by the embodiments of the present invention, when the segmentation of the secretion boundary of the image to be detected is inaccurate, by comprehensively considering the set of image frames and respiratory phase data within the target time period, and combining the set of target image frames and target respiratory phase data within the target time period, the influence of respiration on the image morphology is identified, and the image to be detected is re-segmented to obtain a more accurate target secretion image, which can effectively solve the problem of inaccurate detection of the secretion morphology of the image in the prior art and achieve the effect of obtaining an accurate secretion volume according to the image.

[0134] Figure 4 The structural schematic diagram of a visible and adjustable temperature measuring endotracheal tube is exemplified. The visible and adjustable temperature measuring endotracheal tube may include: a main body tube 410, a temperature sensing module 420, and an image acquisition module 430; an adjusting member 440 for adjusting the bending angle of the main body tube is further provided inside the main body tube 410; one end of the main body tube 410 is provided with a first opening, and the image acquisition module 430 is disposed inside the first opening. The visible and adjustable temperature measuring endotracheal tube is equipped with an inflatable cuff 450 for easy fixation. The temperature sensing module 420 may be disposed on the inner wall of the cuff 450, and more accurate temperature signs can be obtained when the cuff 450 is attached to the tracheal tissue. The other end is provided with a wire harness integration terminal 460, which is convenient for integrating various electronic wire harnesses, and the electronic wire harnesses may be connected to the temperature sensing module 420 and the image acquisition module 430.

[0135] The temperature sensing module is used to obtain the temperature to achieve body temperature monitoring. The image acquisition module is used to collect the image to be detected in the trachea to obtain the amount of secretions in the trachea, can obtain the tracheal image in real time and transmit it to the display, and can also assist in observing and adjusting the catheter position. The amount of secretions in the trachea is determined after processing the image to be detected based on the above-mentioned secretion detection method.

[0136] The bending angle can be adjusted by the adjusting member, which is convenient for quickly and accurately inserting into the target lobar bronchus, shortening the intubation time and reducing airway injury. The adjusting member may adopt the form of a threaded metal wire, which has a certain stiffness and good bending performance.

[0137] During the insertion and use process, real-time imaging and angle adjustment can also be combined to quickly enter the target bronchus, reducing the intubation time and airway injury. When displaced, the isolation effect can be restored by adjustment to avoid secondary intubation.

[0138] The main body tube, as the intubation main body, may be made of silicone or PVC material, which has the characteristics of flexibility and low damage. The built-in adjusting member provides controllable bending performance, and precise intubation positioning is achieved by adjusting the bending angle.

[0139] The temperature sensing module is positioned near the inflatable cuff at the distal end of the intubation tube, and it monitors the airway temperature (close to the core body temperature) in real time. The data is transmitted to the monitor through the wire harness integrated terminal, which is applicable to intraoperative body temperature management and postoperative complication warning.

[0140] The image acquisition module integrates a micro camera and can also be equipped with a ring-shaped LED light source to capture the images inside the trachea in real time. It combines algorithms to analyze the amount of secretions, mucosal status, etc., and assists in judging whether the position of the catheter is deviated or there is a risk of secretion blockage.

[0141] Specifically, the adjusting part is made of a spiral steel wire or a shape memory alloy wire embedded in the tube wall, and the distal bending angle can also be controlled by rotating or pulling the guide wire through setting a proximal handle, reducing the intubation blind area.

[0142] As Figure 4 shown, an adjusting handle 461 for adjustment can be set on the wire harness integrated terminal 460. By rotating the adjusting handle 461, the adjusting part 440 is driven to move, thereby realizing the adjustment of the main pipeline 410.

[0143] On this basis, based on image recognition technology, the covered area of the secretions is quantified to trigger an automatic warning. For example, when the amount of secretions is relatively large, it reminds to suction sputum. In one example, manual sputum suction can be adopted according to the prompt to clean the secretions. In another example, automatic sputum suction can also be adopted. An automatic sputum suction device can be integrated to start automatic sputum suction when the amount of secretions is large, improving the automation degree and monitoring efficiency.

[0144] Sputum suction can be based on the negative pressure principle. The negative pressure suction system creates an environment with a pressure lower than the atmospheric pressure in the sputum suction pipeline, and uses the pressure difference of the atmospheric pressure to suck the secretions (such as sputum) in the airway into the sputum suction device.

[0145] Specifically, when sputum suction is turned on, the negative pressure device will reduce the air pressure inside the sputum suction pipeline (generate negative pressure). Since the pressure in the airway is relatively high, the secretions will be sucked into the sputum suction pipeline. With this negative pressure difference, the secretions are guided through the pipeline to the collection container. However, in the prior art, the negative pressure of the sputum suction hole has a large impact on the airway surface, which is likely to cause airway reactions, and then cause discomfort or reactions of the patient.

[0146] In some implementation manners, as Figure 5 shown, a sputum suction ring 470 is sleeved on the main pipeline. The sputum suction ring 470 is located at one end of the cuff 450 away from the image acquisition module 430. The sputum suction ring 470 includes a chamber structure with a hollow interior to absorb sputum, and a plurality of sputum suction holes 480 are formed on the sputum suction ring 470.

[0147] In this implementation, in addition to monitoring and suctioning in the trachea at the first opening, suctioning can also be performed in the trachea at the rear end of the cuff. The design purpose of the suction ring is to reduce the direct impact force of negative pressure on the airway surface near the cuff by optimizing the layout of the suction holes and the distribution of suction force, thereby reducing the risk of airway reaction and patient discomfort.

[0148] Specifically, the suction ring is located at the rear end of the cuff, and a hollow chamber structure is provided inside it. This structure can effectively absorb and collect secretions (such as sputum) in the airway, avoiding directly concentrating the negative pressure on the airway surface, thereby reducing the impact force on the airway. The design of the suction ring makes the suction force more evenly distributed, helping to reduce airway discomfort caused by excessive or concentrated negative pressure.

[0149] The multiple suction holes on the suction ring are distributed, and the size and number of the suction holes can be set according to the actual situation such as the thickness of the main pipe. The suction force of each hole is small, reducing the excessive negative pressure impact of a single hole on the airway, which can effectively reduce the damage to airway tissues. At the same time, these suction holes can cover the secretions in the airway more comprehensively, ensuring more efficient and uniform sputum clearance. The hollow chamber design of the suction ring also helps to buffer the suction force, making the suction process more gentle and precise, avoiding damage to the airway caused by excessive negative pressure, improving the patient's comfort, especially during long-term use, and reducing the risk of discomfort and potential airway damage.

[0150] A communication wire harness for external connection can be set through the wire harness integrated terminal to send the detection results to an external device. The visible adjustable temperature tracheal intubation can be connected to an external display device, and the acquired real-time image and real-time body temperature data can be displayed through the display device for intuitive personnel monitoring. In addition, an alarm prompt can be set to give a voice prompt or display specific prompt text when the amount of secretions exceeds the control requirements and the body temperature is too high. Specifically, the display device can be a separately set terminal device such as a tablet computer or a monitor, and there is no limitation here.

[0151] In addition, the temperature monitoring function can give a prompt when the temperature is abnormal. Body temperature monitoring is an important part of perioperative management, and patients are highly likely to have perioperative hypothermia. Because the head is covered with a sterile cloth during the patient's surgery, the traditional nasopharyngeal temperature monitoring has problems such as catheter displacement, and the body temperature data is inaccurate due to unstable position. Monitoring the body temperature through the cuff of the tracheal catheter has a relatively fixed and firm monitoring site and is closer to the core body temperature. Therefore, when wrapping materials such as sterile cloth, directly using the tracheal catheter to obtain images and body temperature is more convenient without the need to separately set additional detection instruments to intervene in the body.

[0152] This tracheal catheter integrates intelligent sensing and mechanical optimization, marking the upgrade of the tracheal catheter from a passive tool to an active monitoring device. It can also achieve remote data sharing and AI decision support through the Internet of Things.

[0153] Figure 6 An example of the physical structure diagram of an electronic device is shown as Figure 6 shown. The electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call the logical instructions in the memory 630 to execute the secretion detection method, which includes: performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image; determining an initial cumulative amount level of the secretion based on the initial secretion image; the cumulative amount level is used to evaluate the amount of secretions in the trachea; when the initial cumulative amount level indicates that the amount of secretions in the trachea is less than a preset cumulative amount, determining the segmentation confidence of the initial secretion image; the segmentation confidence is used to measure the accuracy of the secretion boundary segmentation of the image to be detected; when the segmentation confidence is less than a preset confidence, based on the set of target image frames within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period, re-performing secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image; determining the amount of secretions in the trachea based on the target secretion image.

[0154] In addition, when the logical instructions in the above-mentioned memory 630 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0155] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the secretion detection method provided by the above-mentioned various methods. The method includes: performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image; determining an initial cumulative amount level of the secretion based on the initial secretion image; the cumulative amount level is used to evaluate the amount of secretion in the trachea; when the initial cumulative amount level indicates that the amount of secretion in the trachea is less than a preset cumulative amount, determining the segmentation confidence of the initial secretion image; the segmentation confidence is used to measure the accuracy of the secretion boundary segmentation of the image to be detected; when the segmentation confidence is less than a preset confidence, based on the target image frame set within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period, re-performing secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image; determining the amount of secretion in the trachea based on the target secretion image.

[0156] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the secretion detection method provided by the above-mentioned various methods. The method includes: performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image; determining an initial cumulative amount level of the secretion based on the initial secretion image; the cumulative amount level is used to evaluate the amount of secretion in the trachea; when the initial cumulative amount level indicates that the amount of secretion in the trachea is less than a preset cumulative amount, determining the segmentation confidence of the initial secretion image; the segmentation confidence is used to measure the accuracy of the secretion boundary segmentation of the image to be detected; when the segmentation confidence is less than a preset confidence, based on the target image frame set within the target time period where the acquisition time of the image to be detected is located and the target respiratory phase data within the target time period, re-performing secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image; determining the amount of secretion in the trachea based on the target secretion image.

[0157] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0158] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting a secretion, characterized in that, Including: Performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image; Based on the initial secretion image, determining an initial cumulative amount level of the secretion; The cumulative amount level is used to evaluate the amount of tracheal secretions; When the initial cumulative amount level indicates that the amount of tracheal secretions is less than a preset cumulative amount, determining a segmentation confidence of the initial secretion image; the segmentation confidence is used to measure the accuracy of performing secretion boundary segmentation on the image to be detected; When the segmentation confidence is less than a preset confidence, based on a set of target image frames within a target time period where the acquisition time of the image to be detected is located and target respiratory phase data within the target time period, re-performing secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image; Based on the target secretion image, determining the amount of tracheal secretions.

2. The secretion detection method according to claim 1, characterized in that The re-performing secretion boundary segmentation on the acquired image to be detected based on a set of target image frames within a target time period where the acquisition time of the image to be detected is located and target respiratory phase data within the target time period to obtain a target secretion image includes: Based on a set of target image frames within a target time period where the acquisition time of the image to be detected is located and target respiratory phase data within the target time period, determining a target motion feature of tracheal secretions within the target time period; the motion feature is used to represent the morphological change of tracheal secretions under the action of respiratory airflow; Based on the target motion feature and a reference image in the set of target image frames, determining a target image corresponding to the image to be detected; the reference image is determined based on the segmentation confidence of a secretion image obtained by performing secretion segmentation on the images in the set of target image frames; Performing secretion boundary segmentation on the target image to obtain the target secretion image.

3. The secretion detection method according to claim 2, wherein The determining a target motion feature of tracheal secretions within the target time period based on a set of target image frames within a target time period where the acquisition time of the image to be detected is located and target respiratory phase data within the target time period includes: Based on a first set of images acquired before the acquisition time of the image to be detected in the set of target image frames and first respiratory phase data corresponding to the first set of images in the target respiratory phase data, obtaining a first motion feature of the secretion; Based on a second set of images acquired after the acquisition time of the image to be detected in the set of target image frames and second respiratory phase data corresponding to the second set of images in the target respiratory phase data, obtaining a second motion feature of the secretion; Based on the first motion feature and the second motion feature, determining the target motion feature.

4. The secretion detection method according to any one of claims 1 to 3, characterized in that The motion feature of the secretion is determined by the following method: Based on the target respiratory phase data, aligning each frame of the images in the set of target image frames with the respiratory phase to determine a pixel displacement field of the images in the set of target image frames; The pixel displacement field is calculated according to different smoothing coefficients corresponding to different respiratory phases; Based on the pixel displacement field and the target respiratory phase data, statistically determine the main direction of secretion movement and the regional activity; The regional activity is used to evaluate the movement intensity and distribution of secretions in the trachea; Based on the empirical data of the movement characteristics of secretions, adjust the main direction of secretion movement and the regional activity, and determine the adjusted main direction of secretion movement and the regional activity as the movement characteristics of the secretions.

5. The secretion detection method according to claim 3, characterized in that, The determining of the target image corresponding to the image to be detected based on the target movement characteristics and the reference image in the target image frame set includes: Based on the target movement characteristics and the reference image in the target image frame set, perform morphological enhancement on the image to be detected to obtain the target image.

6. The secretion detection method according to claim 1, wherein, The segmentation confidence is determined based on the pixel value variance and / or boundary clarity of the secretion region.

7. A secretion detection device, characterized in that, It includes: A first segmentation module for performing secretion boundary segmentation on the acquired image to be detected to obtain an initial secretion image; A first processing module for determining the initial cumulative amount level of the secretions based on the initial secretion image; The cumulative amount level is used to evaluate the amount of secretions in the trachea; A second processing module for determining the segmentation confidence of the initial secretion image when the initial cumulative amount level indicates that the amount of secretions in the trachea is less than a preset cumulative amount; the segmentation confidence is used to measure the accuracy of performing secretion boundary segmentation on the image to be detected; A second segmentation module for, when the segmentation confidence is less than a preset confidence, based on the target image frame set within the target time period where the time when the image to be detected is acquired is located and the target respiratory phase data within the target time period, re-perform secretion boundary segmentation on the acquired image to be detected to obtain a target secretion image; A third processing module for determining the amount of secretions in the trachea based on the target secretion image.

8. A visible adjustable temperature measuring endotracheal tube, characterized in that, It includes: A main pipeline, a temperature sensing module, an image acquisition module, and a cuff; an adjusting member for adjusting the bending angle of the main pipeline is further provided inside the main pipeline; one end of the main pipeline is provided with a first opening, the image acquisition module is disposed inside the first opening, and the temperature sensing module is disposed inside the cuff; The temperature sensing module is used to acquire temperature, and the image acquisition module is used to collect the image to be detected in the trachea to obtain the amount of secretions in the trachea; the amount of secretions in the trachea is determined after processing the image to be detected by the secretion detection method according to any one of claims 1 to 6.

9. The visible adjustable temperature measuring endotracheal tube according to claim 8, characterized in that, A sputum suction ring is sleeved on the main pipeline, the sputum suction ring is located at one end of the cuff away from the image acquisition module, the sputum suction ring includes a chamber structure with a hollow interior for absorbing sputum, and a plurality of sputum suction holes are provided on the sputum suction ring.

Images