Method and system for automatically guiding intubation in airway three-dimensional reconstruction based on CT (Computed Tomography) image

Through three-dimensional reconstruction technology based on CT imaging and real-time monitoring system, the tracheal catheter path is dynamically adjusted, which solves the problems of low success rate and complex operation of traditional tracheal intubation technology, and achieves a more efficient and safe intubation process.

CN119015554APending Publication Date: 2024-11-26SHANGHAI NINTH PEOPLES HOSPITAL SHANGHAI JIAO TONG UNIV SCHOOL OF MEDICINE

Patent Information

Application Number
CN202411012817.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Traditional tracheal intubation technology relies on the experience and skills of the operator, resulting in a low success rate of intubation, which can easily cause serious complications such as asphyxiation, hypoxemia, and lung inhalation, especially in patients with difficult airways for a long time and Strong discomfort.

Method used

Using the three-dimensional reconstruction of airways based on CT images, the three-dimensional airway model is generated by obtaining the patient's CT image data, and the preset path of the tracheal catheter is planned using the Bezier curve algorithm, combining sensors and cameras to monitor the air in real time In the environment of the channel, dynamically adjust the intubation path to ensure that the tracheal catheter advances along the airway path.

Benefits of technology

It significantly improves the success rate and safety of intubation, simplifies the operation process, reduces the burden on medical staff, improves surgical efficiency, and can adapt to the changing airway structure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119015554A_ABST
    Figure CN119015554A_ABST
Patent Text Reader

Abstract

The invention discloses an airway three-dimensional reconstruction automatic guide intubation method and system based on a CT image, and the method comprises the following steps: S1, at least obtaining the CT image data of a patient, and transmitting the CT image data to a three-dimensional reconstruction unit module; s2, a three-dimensional reconstruction unit module processes the CT image data, a three-dimensional airway model is generated, the position and shape of an airway are preset, and a preset path of the tracheal catheter is planned; s3, a sensor and a camera in the guide unit module capture airway images and tissue position information in real time in the advancing process of the intubation tube, and real-time image data are obtained; s4, receiving the real-time image data by the image processing module, comparing the real-time image data with the three-dimensional airway model, and positioning the tracheal catheter in the three-dimensional airway model; and S5, the control module controls the tracheal catheter to enter the airway according to the real-time image data. The system can greatly improve the success rate and safety of tracheal catheter intubation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical equipment and biomedical engineering technology, and in particular to a method and system for automatically guiding intubation of an airway based on three-dimensional reconstruction of CT images. Background Art

[0002] Tracheal intubation is a method of inserting a special endotracheal tube into the trachea or bronchus through the oral or nasal cavity and the glottis, which provides the best conditions for airway patency, ventilation and oxygen supply, and airway suction, and is an important measure to rescue patients with respiratory dysfunction. However, traditional endotracheal intubation technology relies on the experience and skills of the operator. When facing patients with difficult airways, the success rate of intubation is low, and it is easy to cause serious complications such as suffocation, hypoxemia, and pulmonary aspiration. In addition, intubation when the patient is awake requires constant adjustment and trial of the intubation position, which leads to long operation time and strong patient discomfort. Summary of the invention

[0003] The purpose of the present invention is to overcome the defects of low accuracy and success rate of existing tracheal intubation and to provide an intubation method and intubation system that can significantly improve the success rate and safety of intubation.

[0004] In order to achieve the above object, the present invention provides a method for automatically guiding intubation of airway based on CT image three-dimensional reconstruction, comprising the following steps:

[0005] Step S1, at least obtaining CT image data of a patient, and transmitting the CT image data to a three-dimensional reconstruction unit module;

[0006] Step S2, the three-dimensional reconstruction unit module processes the CT image data to generate a three-dimensional airway model, presets the position and shape of the airway, and plans a preset path of the tracheal tube using a Bezier curve algorithm;

[0007] Step S3, the sensor and camera in the guiding unit module capture the airway image and tissue position information of the intubation in real time to obtain real-time image data;

[0008] Step S4, the image processing module receives the real-time image data in step S3, compares the real-time image data with the three-dimensional airway model in step S2, and locates the position of the endotracheal tube in the three-dimensional airway model;

[0009] Step S5: the control module controls the endotracheal tube to enter the airway according to the real-time image data.

[0010] Preferably, step S1 further includes acquiring an image using an endoscope, and transmitting the image acquired by the endoscope and the CT image data together to a three-dimensional reconstruction unit module.

[0011] Preferably, in step S2, the processing is to process the CT image data by using a diffusion model and an image point cloud method.

[0012] Preferably, in step S3, the sensor includes an ultrasonic sensor or a pressure sensor for acquiring real-time image data of multiple dimensions.

[0013] Preferably, the guiding unit module further comprises a guiding device, and the guiding device moves along the preset path until it passes through the glottis and enters the airway.

[0014] Preferably, steps S3 to S5 are looped to continuously receive real-time image data and dynamically adjust the actual path of the endotracheal tube.

[0015] Preferably, in step S5, the airway includes a normal airway or an airway with anatomical variation.

[0016] Preferably, in step S5, the control module identifies that the preset path is different from the actual path, and the sensor and camera in the guidance unit module further identify the airway tissue environment, update the three-dimensional airway model, and re-plan the preset path of the tracheal tube.

[0017] The present invention also provides a system for automatically guiding intubation based on three-dimensional reconstruction of the airway by CT images, wherein the system at least comprises the three-dimensional reconstruction unit module, the guiding unit module, the image processing module, and the control module as described above.

[0018] Preferably, the system further comprises a display module, wherein the display module comprises a visualization interface, and the display module is used to display real-time image data and a three-dimensional airway model.

[0019] Compared with the prior art, the beneficial effects of the present invention include at least:

[0020] (1) The present invention provides a system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway, which comprises a three-dimensional reconstruction unit module, a guidance unit module, an image processing module, a control module, and a display module. The system first acquires the patient's CT image data, and then processes the CT image data using the three-dimensional reconstruction unit module to generate a three-dimensional airway model, thereby providing detailed airway structure information and significantly improving the accuracy and success rate of subsequent intubation.

[0021] (2) The present invention adopts advanced image processing modules and recognition technologies, uses sensors and cameras to monitor the airway environment in real time, and automatically guides the tracheal tube along the airway path, further ensuring the accuracy of intubation. The fully automatic intubation solution simplifies the operation process, reduces the burden on medical staff, and greatly improves the efficiency of the operation.

[0022] (3) When the control module of the present invention recognizes that the preset path of the tracheal tube is different from the actual path, the sensor and camera in the guidance unit module further identify the airway tissue environment and re-plan the preset path of the tracheal tube, thereby dynamically adjusting the intubation path, so that the system can adapt to the changing airway structure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a flow chart of a method for automatically guiding intubation based on three-dimensional reconstruction of airways using CT images.

[0024] Figure 2 Schematic diagram of the airway three-dimensional reconstruction process of the present invention.

[0025] Figure 3 The schematic diagram of generating a tracheal intubation path of the present invention.

[0026] Figure 4 It is a schematic diagram of the working of the guiding device of the present invention when the patient is in a normal supine position.

[0027] Figure 5 It is a schematic diagram of the operation of the guiding device of the present invention when the patient's body position changes and thus the airway morphology changes.

[0028] Figure 6 The present invention is a system block diagram of automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway. DETAILED DESCRIPTION

[0029] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0030] Traditional tracheal intubation relies on the experience and skills of the operator, and there is a high risk of failure and complication rate, especially when facing patients with difficult airways. Traditional methods lack a detailed understanding of the airway structure and real-time navigation, which can easily lead to misinsertion and intubation failure. In order to provide a certain visual aid, some image-guided technologies are usually used in conjunction with tracheal intubation. Although image-guided technologies (such as video laryngoscopes and optical coherence tomography) are also used in the prior art to provide a certain visual aid, the effect of automatic intubation guidance when the patient is awake is limited, and it is impossible to fully utilize preoperative images for real-time navigation. In addition, these technologies still need to rely heavily on manual intervention in actual operation, with a low degree of automation, and it is difficult to cope with complex airway structures. These intubation guidance devices also lack an efficient real-time feedback mechanism, and cannot make accurate real-time adjustments during the intubation process, which can easily lead to misinsertion or intubation failure, and the intubation guidance devices have limited ability to identify and track airway structures and cannot cope with dynamically changing airway environments.

[0031] Based on the above defects, the present invention provides a method for automatically guiding intubation of airway based on CT images using three-dimensional reconstruction, comprising the following steps:

[0032] Step S1, such as Figure 1 , 2 As shown, CT image data of several difficult airway patients (from the Ninth People's Hospital affiliated to Shanghai Jiao Tong University School of Medicine) are obtained, and the CT image data are transmitted to the three-dimensional reconstruction unit module.

[0033] In some embodiments, in addition to CT image data, MRI, ultrasound or other medical imaging equipment can be used to obtain airway structure data, and multiple image data can be integrated to generate more comprehensive data for subsequent three-dimensional reconstruction. During surgery, real-time endoscopic images can be combined to provide more comprehensive airway information, thereby enhancing navigation accuracy.

[0034] Step S2, the three-dimensional reconstruction unit module processes the CT image data, for example, using a diffusion model and an image point cloud method to generate a three-dimensional airway model (i.e., the 3D structure of the head and neck). The diffusion model accurately simulates the tissue structure at different positions of the head and neck through data of multi-angle and displacement changes, presets the position and shape of the airway, and uses a Bezier curve algorithm to plan a preset path for the tracheal tube to ensure the smoothness and controllability of the intubation path. The Bezier curve algorithm generates an optimal one based on the intubation starting point, end point, and control point, and fits the preset path to the patient's CT image data, such as Figure 3 As shown, based on the fitted three-dimensional shape, the actual guidance path is established to guide the subsequent cannulation.

[0035] Among them, 1. Diffusion model and image point cloud method are as follows:

[0036] 1.1 Point cloud data generation

[0037] -**CT image preprocessing**:

[0038] -Input: Patient's head and neck CT image data.

[0039] -Processing: Use image processing algorithms to denoise and enhance CT images to improve image quality.

[0040] -Output: enhanced CT image data.

[0041] -**Image Segmentation**:

[0042] -Use convolutional neural networks (such as U-Net) to segment CT images and extract the main tissues and structures of the head and neck, such as bones, airways, blood vessels, etc.

[0043] -Generate binary mask images of different tissues.

[0044] -**Point cloud data generation**:

[0045] -Use Marching Cubes algorithm to convert segmented 3D volume data into surface mesh.

[0046] -Sample points from the surface mesh to generate point cloud data.

[0047] -Output: Point cloud data containing various tissue structures of the head and neck.

[0048] 1.2 Diffusion model training

[0049] - **Data preprocessing**:

[0050] -Input: point cloud data.

[0051] -Processing: Standardize the point cloud data so that its center is at the origin and its scale is within a uniform range.

[0052] -**Diffusion model architecture**:

[0053] -Choose an appropriate diffusion model framework, such as Denoising Diffusion Probabilistic Models (DDPM).

[0054] -Model input: standardized point cloud data.

[0055] -Model output: 3D structure of head and neck generated under multiple angles and displacements.

[0056] -**Model training**:

[0057] -Loss function: Use mean square error (MSE) or Chamfer distance to measure the similarity between the generated point cloud and the real point cloud.

[0058] -Optimizer: Use Adam optimizer.

[0059] - Training process: The model parameters are continuously optimized through back propagation so that the generated head and neck 3D structure is close to the real structure.

[0060] -**Model Reasoning**:

[0061] -Input: Point cloud data generated from new CT images.

[0062] -Output: Generated 3D structure of head and neck, including changes under different angles and displacements.

[0063] 2. Bezier curve path planning algorithm

[0064] 2.1 Basic principles of Bezier curve

[0065] -**Bezier curve definition**:

[0066] -A Bezier curve is defined by a set of control points, and its shape is determined by the arrangement of these control points.

[0067] The expression of the n-order Bezier curve is:

[0068]

[0069] Among them, t ranges from [0, 1], P i is the control point, is the binomial coefficient.

[0070] 2.2 Intubation Path Planning

[0071] -**Control Point Settings**:

[0072] - Starting point: The starting position of the cannula.

[0073] - End point: the target location for intubation (eg, trachea).

[0074] -Intermediate control points: Several control points are set between the starting point and the end point to ensure that the path avoids important organizations and obstacles.

[0075] - **Path Generation**:

[0076] - Generate the cannula path based on the control points using the Bezier curve formula described above.

[0077] -The smoothness and controllability of the path is determined by the number and location of control points.

[0078] 2.3 Path Optimization Based on Point Cloud

[0079] - **Path evaluation**:

[0080] - Input: Generated Bezier path.

[0081] -Processing: Evaluate the path in the 3D point cloud model of the head and neck, and determine the safety of the path by calculating the distance between the path and important tissues in the point cloud (such as blood vessels, bones, etc.).

[0082] -Output: Path evaluation results, including the safety score of the path.

[0083] -**Path Optimization**:

[0084] -If the path is unsafe, adjust the position of the middle control point and regenerate the Bezier curve.

[0085] -Iterative process: Continuously adjust the control point positions until a safe intubation path is generated.

[0086] - Further optimize the path using genetic algorithms or simulated annealing algorithms to find the global optimal solution.

[0087] Through the above detailed diffusion model and Bezier curve path planning algorithm, this system can generate a high-precision virtual head and neck 3D structure based on the patient's CT images, plan the optimal intubation path, and ensure the safety and accuracy of the intubation process.

[0088] Step S3, the system for automatic airway intubation based on CT image three-dimensional reconstruction of the present invention includes a guiding unit module, and the guiding unit module includes a guiding mechanical arm, a camera, an ultrasonic sensor, a pressure sensor, and a guiding device. The camera, ultrasonic sensor, and pressure sensor are located at the top of the guiding mechanical arm, and are used to capture images and data in real time, and transmit the images and data to the image processing module. Specifically, the ultrasonic sensor, pressure sensor, and camera in the guiding unit module capture the airway image and tissue position information during the intubation in real time, the sensor is used to collect the position distance information around the tracheal tube in real time, the camera collects environmental information and the insertion depth of the tracheal tube, and the data fusion algorithm combines the data of the sensor and the camera to obtain real-time image data, which is used to locate the position of the tracheal tube in real time. Figure 4 and Figure 5 As shown, the guiding device moves along the preset path when the patient's body position and anatomical structure do not change.

[0089] Step S4, the system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway of the present invention comprises an image processing module, wherein the image processing module is used to receive the real-time image data in step S3, compare the real-time image data with the three-dimensional airway model in step S2, and locate the position of the tracheal tube in the three-dimensional airway model.

[0090] Step S5, the system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway of the present invention includes a control module, which controls the tracheal tube to enter the airway along a preset path according to real-time image data. If the control module recognizes that the preset path is different from the actual path, the ultrasonic sensor, pressure sensor and camera in the guidance unit module further identify the airway tissue environment, and based on the artificial intelligence algorithm, update the three-dimensional airway model to re-plan the preset path of the tracheal tube. If a small error occurs, the path and position are fine-tuned by the information obtained to correct the position of the tracheal tube; if a large error occurs, the system will analyze the current position of the tracheal tube and the structural difference with the suspected position in the three-dimensional airway model, update the virtual head and neck 3D structure, and re-plan the path of the tracheal tube.

[0091] In some embodiments, steps S3 to S5 may be performed in a loop, and the image processing module continuously receives real-time image data, thereby dynamically adjusting the actual path of the tracheal tube. The actual path may be continuously adjusted during the intubation process until the tracheal tube reaches a predetermined position, thereby improving the accuracy of the intubation.

[0092] It can be understood that in order to facilitate the operator to realize the above-mentioned dynamic adjustment of the actual path of the tracheal tube, the system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway also includes a display module, and the display module presents the above-mentioned real-time image data and three-dimensional airway model in real time, so as to facilitate the operator to monitor in real time and assist in decision-making during the intubation process.

[0093] like Figure 6 As shown, based on the above, the present invention also provides a system for automatic airway intubation guidance based on CT images and three-dimensional reconstruction, the system comprising a three-dimensional reconstruction unit module, a guidance unit module, an image processing module, a control module, and a display module.

[0094] Example 1

[0095] The method and system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway of the present invention were verified on a simulated human model:

[0096] Step S1, obtaining CT image data of a simulated person, and transmitting the CT image data to a three-dimensional reconstruction unit module;

[0097] Step S2, placing the system for automatic intubation guidance based on CT images of the airway three-dimensional reconstruction of the present invention into the oral cavity of the simulated human model; the three-dimensional reconstruction unit module processes the CT image data to generate a three-dimensional airway model;

[0098] Step S3, the camera and sensor in the guidance unit module in the system detect the airway environment in real time to obtain real-time image data;

[0099] Step S4, the control module in the system adjusts the actual path according to the real-time image data to ensure that the endotracheal tube passes through the glottis accurately.

[0100] Example 2

[0101] The method and system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway of the present invention were verified on real patients:

[0102] Step S1, obtaining CT image data of the patient before surgery and performing three-dimensional reconstruction;

[0103] Step S2, performing routine preparation and anesthesia while the patient is awake;

[0104] Step S3, placing the system for automatic intubation guidance based on CT image-based three-dimensional reconstruction of the airway into the patient's oral cavity to start the intubation operation;

[0105] Step S4, the camera and sensor in the guidance unit module in the system detect the airway environment in real time to obtain real-time image data;

[0106] Step S5, the control module in the system controls the endotracheal tube to enter the airway through the glottis to complete the intubation operation.

[0107] Example 3

[0108] A comparative experiment was conducted using the method and system for automatically guiding intubation based on CT image-based three-dimensional reconstruction of the airway and the traditional manual intubation method, and intubation operations were performed on the same batch of patients.

[0109] In summary, the method of automatic intubation guidance based on CT image-based three-dimensional reconstruction of airway of the present invention comprises the following steps: step S1, at least obtaining the CT image data of the patient, and transmitting the CT image data to the three-dimensional reconstruction unit module; step S2, the three-dimensional reconstruction unit module processes the CT image data, generates a three-dimensional airway model, presets the position and shape of the airway, and plans the preset path of the tracheal tube; step S3, the sensor and camera in the guidance unit module capture the airway image and tissue position information in the process of intubation in real time, and obtain real-time image data; step S4, the image processing module receives the real-time image data, compares the real-time image data with the three-dimensional airway model, and locates the position of the tracheal tube in the three-dimensional airway model; step S5, the control module controls the tracheal tube to enter the airway according to the real-time image data. The system can greatly improve the success rate and safety of tracheal tube intubation.

[0110] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be appreciated that the above description should not be considered as a limitation of the present invention. After reading the above content, it will be apparent to those skilled in the art that various modifications and substitutions of the present invention will occur. Therefore, the protection scope of the present invention should be limited by the appended claims.

Claims

1. A method for automatically guiding intubation based on three-dimensional reconstruction of airway by CT images, characterized in that: The following steps are involved: Step S1, at least obtaining CT image data of a patient, and transmitting the CT image data to a three-dimensional reconstruction unit module; Step S2, the three-dimensional reconstruction unit module processes the CT image data to generate a three-dimensional airway model, presets the position and shape of the airway, and plans a preset path of the tracheal tube using a Bezier curve algorithm; Step S3, the sensor and camera in the guiding unit module capture the airway image and tissue position information of the intubation in real time to obtain real-time image data; Step S4, the image processing module receives the real-time image data in step S3, compares the real-time image data with the three-dimensional airway model in step S2, and locates the position of the endotracheal tube in the three-dimensional airway model; Step S5: the control module controls the endotracheal tube to enter the airway according to the real-time image data.

2. The method according to claim 1, characterized in that Step S1 also includes acquiring an image using an endoscope, and transmitting the image acquired by the endoscope and the CT image data to a three-dimensional reconstruction unit module.

3. The method according to claim 1, characterized in that In step S2, the processing is to process the CT image data by using a diffusion model and an image point cloud method.

4. The method according to claim 1, characterized in that In step S3, the sensor includes an ultrasonic sensor or a pressure sensor, which is used to obtain real-time image data of multiple dimensions.

5. The method according to claim 1, characterized in that The guiding unit module further comprises a guiding device, which moves along the preset path until it passes through the glottis and enters the airway.

6. The method according to claim 1, characterized in that Steps S3 to S5 are looped to continuously receive real-time image data and dynamically adjust the actual path of the endotracheal tube.

7. The method according to claim 1, characterized in that In step S5, the airway includes a normal airway or an airway with anatomical variation.

8. The method according to claim 1, characterized in that In step S5, the control module identifies that the preset path is different from the actual path, and the sensor and camera in the guidance unit module further identify the airway tissue environment, update the three-dimensional airway model, and re-plan the preset path of the tracheal tube.

9. A system for automatically guiding intubation based on three-dimensional reconstruction of airway using CT images, characterized in that: The system at least comprises the three-dimensional reconstruction unit module, the guiding unit module, the image processing module, and the control module as described in any one of claims 1 to 8.

10. The system according to claim 9, characterized in that The system further comprises a display module, wherein the display module comprises a visualization interface, and the display module is used for displaying real-time image data and a three-dimensional airway model.

Citation Information

Patent Citations

  • Airway management navigation method and system

    CN113456225A

  • Trachea cannula navigation system based on structured light stereoscopic vision

    CN115252992A

  • Insertion support system

    JP2005013358A

  • Imaging system for automated intubation

    WO2023167668A1

Cited By

  • Visual guide type tracheal intubation endoscope system

    CN119498763A

  • A visual guidance endotracheal intubation scope system

    CN119498763B

  • Airway intubation guiding method and airway intubation device

    CN119524280A

  • Airway intubation guiding method and airway intubation device

    CN119524280B

  • Visual trachea cannula and control method thereof

    CN120094061A