A method and apparatus for predicting the timing of bronchoscopic catheter entry into a branch

Abstract

The application discloses a bronchoscope catheter branch entry timing prediction method and device, and relates to the technical field of medical instrument navigation, and comprises the following steps: determining a required steering angle according to a branch included angle between a main bronchial tract and a target branch, a catheter posture deviation relative to the target branch, and a target branch entrance azimuth deviation relative to a catheter field of view center; determining a steerable window according to the required steering angle; determining a window matching degree according to a distance from a catheter tip to a bifurcation center to a center of the steerable window; determining a posture matching degree according to the posture deviation and the azimuth deviation; determining a geometric prediction score; if the geometric prediction score reaches a preset geometric trigger threshold value and the remaining distance is located in the steerable window, a candidate entry time is generated and added to a candidate set; and selecting an optimal entry starting time from the candidate set. The method can accurately predict the timing of the bronchoscope catheter entering the branch, avoid early steering and easy wall-hugging deviation, and avoid late steering and missing the target branch entrance.

Description

Technical Field

[0001] This application relates to the field of medical device navigation technology, and in particular to a method and device for predicting the timing of a bronchoscope catheter entering a branch. Background Technology

[0002] When advancing the bronchoscopic catheter through the airway to the target location, the complex tree-like branching structure of the bronchi necessitates accurate judgment of when to begin turning and entering the target branch as the catheter approaches the airway bifurcation. Techniques for entering bronchoscopic branches primarily rely on the operator's experience, path visualization, or local visual impression to determine the timing of entry, which can easily lead to premature turning (causing the catheter to veer off the wall) or delayed turning (missing the target branch entrance).

[0003] Therefore, how to solve the above-mentioned technical defects has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a method and device for predicting the timing of bronchoscope catheter entry into a branch, which can accurately predict the timing of bronchoscope catheter entry into a branch, avoiding premature turning that may cause the catheter to veer against the wall and premature turning that may cause the catheter to miss the target branch entrance.

[0005] To address the aforementioned technical problems, this application provides a method for predicting the timing of bronchoscope catheter entry into a branch, comprising: Obtain the branch angle between the main bronchus and the target branch, the attitude deviation of the catheter relative to the target branch, the orientation deviation of the target branch inlet relative to the center of the catheter's field of view, and the remaining distance from the catheter tip to the center of the bifurcation. The required turning angle is determined based on the branch angle, the attitude deviation, and the orientation deviation; the required turning angle is used to represent the amount of attitude adjustment that the catheter needs to make to enter the target branch from the current position; The steerable window is determined based on the required steering angle; the steerable window is the time or distance interval within which the conduit can complete the alignment of the target branch inlet before reaching the bifurcation without triggering the risk of wall adhesion or missing the inlet. The window matching degree is determined based on the distance from the remaining distance to the center of the rotatable window; The attitude matching degree is determined based on the attitude deviation and the orientation deviation; The geometric prediction score is determined based on the window matching degree and the pose matching degree. If the geometric prediction score reaches the preset geometric trigger threshold and the remaining distance is within the steerable window, then a candidate entry time is generated and the candidate entry time is added to the candidate set; The optimal entry start time is selected from the candidate set.

[0006] In some embodiments, it also includes: Determine the confidence levels for geometric predictions, image observations, and motion states; The timing confidence is determined based on the geometric prediction confidence, the image observation confidence, and the motion state confidence. Based on the timing confidence level and the first and second thresholds, suggestions are output; wherein, if the timing confidence level is greater than or equal to the first threshold and the remaining distance is within the steerable window, an entry suggestion is output; if the timing confidence level is greater than or equal to the second threshold and less than or equal to the first threshold, a deceleration suggestion, a pre-alignment suggestion, or a continue observation suggestion is output.

[0007] In some embodiments, the method further includes the following before outputting a suggestion based on the timing confidence level and the first and second thresholds: The timing confidence is updated over time to output suggestions based on the timing confidence obtained from the time-recursive update and the first and second thresholds.

[0008] In some embodiments, it also includes: When any one of the preset trigger conditions is met, deceleration and maintaining visual stability are prioritized; the preset trigger conditions include: the remaining distance falls into the steerable window and the timing confidence is less than the first threshold; the difference between the geometrically predicted entrance orientation and the image observation entrance orientation is greater than the preset threshold; the attitude deviation exceeds the preset compensable range; the propulsion speed exceeds the preset safety limit; and the optimal entry start time changes for multiple consecutive cycles.

[0009] In some embodiments, determining the steerable window based on the desired steering angle includes: The minimum time to complete the required steering angle is determined based on the required steering angle and the maximum permissible steering angular velocity; The minimum preceding distance is determined based on the minimum time. The steerable window is determined based on the minimum leading distance and the safety margin distance.

[0010] In some embodiments, selecting the optimal entry start time from the candidate set includes: Calculate the product of the risk item and the risk penalty coefficient; The candidate entry time corresponding to the largest difference between the geometric prediction score and the product is taken as the optimal entry start time.

[0011] In some embodiments, determining the geometric prediction confidence level includes: The geometric prediction confidence is determined based on the window matching degree and the pose matching degree.

[0012] In some embodiments, determining the confidence level of an image observation includes: The image observation confidence level is determined based on the consistency between the target branch inlet contour and the predicted inlet shape, image quality or inlet edge sharpness, and the change in inlet area between adjacent frames.

[0013] In some embodiments, determining the confidence level of the motion state includes: The confidence level of the motion state is determined based on the change in propulsion speed between adjacent cycles and the change in attitude deviation between adjacent cycles.

[0014] To address the aforementioned technical problems, this application also provides an electronic device, comprising: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of the method for predicting the timing of bronchoscope catheter entry into a branch as described above.

[0015] To address the aforementioned technical problems, this application also provides a device for predicting the timing of bronchoscope catheter entry into a branch, comprising: The acquisition unit is used to acquire the branch angle between the main bronchus and the target branch, the attitude deviation of the duct relative to the target branch, the orientation deviation of the target branch inlet relative to the center of the duct's field of view, and the remaining distance from the tip of the duct to the center of the bifurcation. The first determining unit is used to determine the required turning angle based on the branch angle, the attitude deviation, and the orientation deviation; the required turning angle is used to represent the amount of attitude adjustment that the duct needs to complete to enter the target branch from the current position; The second determining unit is used to determine the steerable window according to the required steering angle; the steerable window is the time interval or distance interval in which the duct can complete the alignment of the target branch inlet before reaching the bifurcation without triggering the risk of wall adhesion or missing the inlet. The third determining unit is used to determine the window matching degree based on the distance from the remaining distance to the center of the rotatable window; The fourth determining unit is used to determine the attitude matching degree based on the attitude deviation and the orientation deviation; The fifth determining unit is used to determine the geometric prediction score based on the window matching degree and the pose matching degree; The generation unit is used to generate candidate entry times and add the candidate entry times to the candidate set if the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the steerable window; The selection unit is used to select the optimal entry start time from the candidate set.

[0016] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for predicting the timing of bronchoscope catheter entry into a branch as described above.

[0017] The method for predicting the timing of bronchial catheter entry into a branch provided in this application includes: acquiring the branch angle between the main bronchus and the target branch, the attitude deviation of the catheter relative to the target branch, the azimuth deviation of the target branch inlet relative to the center of the catheter's field of view, and the remaining distance from the catheter tip to the center of the bifurcation; determining the required turning angle based on the branch angle, the attitude deviation, and the azimuth deviation; the required turning angle represents the amount of attitude adjustment the catheter needs to complete from its current position to enter the target branch; determining a steerable window based on the required turning angle; the steerable window is a time or distance interval within which the catheter can complete target branch inlet alignment without triggering wall adhesion or missing the inlet risk before reaching the bifurcation; determining the window matching degree based on the distance from the remaining distance to the center of the steerable window; determining the attitude matching degree based on the attitude deviation and the azimuth deviation; determining a geometric prediction score based on the window matching degree and the attitude matching degree; if the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the steerable window, then generating a candidate entry time and adding the candidate entry time to a candidate set; and selecting the optimal entry start time from the candidate set.

[0018] As can be seen, the bronchoscopic catheter entry timing prediction method provided in this application obtains the branch angle between the main bronchus and the target branch, as well as the catheter's attitude deviation relative to the target branch, and calculates the required turning angle by integrating the orientation deviation, thereby determining the turnable window. This transforms fuzzy empirical judgments into precise spatiotemporal boundaries, fundamentally avoiding the physical risks caused by turning too early or too late. Furthermore, this application determines the window matching degree based on the distance from the remaining distance to the center of the turnable window, the attitude matching degree based on the attitude deviation and orientation deviation, and the geometric prediction score based on the window matching degree and attitude matching degree. A candidate entry time is generated only when the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the turnable window. This effectively eliminates the risk of false triggering caused by image noise or single distance estimation errors, greatly improving the reliability of timing determination. Finally, by optimizing the candidate set, this application can converge and output the optimal entry start time. Throughout the entire process of the catheter approaching the bifurcation, it can proactively and continuously provide an optimal entry start time, reducing human trial and error and empirical bias.

[0019] The bronchoscope catheter insertion timing prediction device, equipment, and computer-readable storage medium provided in this application all have the aforementioned technical effects. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the prior art and embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a method for predicting the timing of a bronchoscopic catheter entering a branch, provided in an embodiment of this application. Figure 2 A schematic diagram of a device for predicting the timing of a bronchoscope catheter entering a branch, provided in an embodiment of this application; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0022] The core of this application is to provide a method and device for predicting the timing of bronchoscope catheter entry into a branch, which can accurately predict the timing of bronchoscope catheter entry into a branch, avoiding premature turning that may cause it to veer against the wall and yaw, and premature turning that may cause it to miss the entrance of the target branch.

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a method for predicting the timing of bronchoscope catheter entry into a branch, as provided in an embodiment of this application. (Refer to...) Figure 1 As shown, the method includes: S101: Obtain the branch angle between the main bronchus and the target branch, the attitude deviation of the catheter relative to the target branch, the orientation deviation of the target branch inlet relative to the center of the catheter's field of view, and the remaining distance from the catheter tip to the center of the bifurcation.

[0025] The target branch refers to the branch of the airway into which the bronchoscopic catheter will enter. The main bronchial airway axis vector can be established with the center point of the bifurcation as the origin. Target branch axis vector and the current axial vector of the conduit .

[0026] The branching angle between the main bronchus and the target branch is: .

[0027] The attitude deviation of the catheter relative to the target branch is: .

[0028] S102: Determine the required turning angle based on the branch angle, the attitude deviation, and the orientation deviation; the required turning angle is used to indicate the amount of attitude adjustment that the catheter needs to complete to enter the target branch from the current position.

[0029] The azimuth deviation of the target branch inlet relative to the center of the duct field of view is denoted as According to the branch angle Posture deviation and azimuth deviation The required steering angle is: .

[0030] in, , , The weights are preset or adaptive, and satisfy the following conditions: The required steering angle indicates the amount of attitude adjustment the duct needs to make before entering the target branch from its current position.

[0031] S103: Determine the steerable window based on the required steering angle.

[0032] The steerable window is the time or distance interval within which the duct can align with the target branch inlet before reaching the bifurcation point without triggering the risk of wall adhesion or missing the inlet. The time interval can be represented as: .

[0033] in, This is the earliest time to suggest starting the turn. This is the latest time when it is still safe to enter the target branch.

[0034] The corresponding distance interval is represented as follows: .

[0035] in, This indicates the distance required to begin turning. Indicates the latest turning distance. This represents the current advancement speed of the catheter. For easier control, this can be calculated based on the remaining distance from the catheter tip to the center of the bifurcation. Determine whether output is allowed to proceed to preparation or suggestion. fall into At this time, output can be entered into preparation or into suggestions.

[0036] This embodiment determines the steerable window based on the required steering angle. In some embodiments, determining the steerable window based on the required steering angle includes: The minimum time to complete the required steering angle is determined based on the required steering angle and the maximum permissible steering angular velocity; The minimum preceding distance is determined based on the minimum time. The steerable window is determined based on the minimum leading distance and the safety margin distance.

[0037] The minimum time required to complete the required steering angle is: . To allow the maximum steering angular velocity.

[0038] The minimum preceding distance is: . The minimum allowable turning radius for the conduit. This is the turning radius compensation coefficient.

[0039] The safety margin distance at the target branch entrance is expressed as: The redirectable window can then be represented as: .

[0040] when When, the system maintains a prediction or pre-alignment state; when When this happens, the system enters a suggested redirection state; when If the target branch entry alignment is not completed, the system determines that the optimal window has been missed or that a low-confidence-risk state has been entered.

[0041] S104: Determine the window matching degree based on the distance from the remaining distance to the center of the rotatable window.

[0042] In each control cycle The current remaining distance from the catheter tip to the center of the bifurcation is expressed as The catheter advancement speed is expressed as The window matching degree is: .

[0043] Indicates the current remaining distance To the reversible window The distance from the center, This is the distance tolerance parameter. The value of the distance tolerance parameter can be determined based on experience or experiments.

[0044] S105: Determine the attitude matching degree based on the attitude deviation and the orientation deviation.

[0045] The pose matching degree is: .

[0046] in, This refers to the angle tolerance parameter. The value of the angle tolerance parameter can be determined based on experience or experiments.

[0047] S106: Determine the geometric prediction score based on the window matching degree and the pose matching degree.

[0048] The geometric prediction score is: .

[0049] in, The risk inverse score is calculated based on the turning radius, safety distance, and entrance cross-section dimensions. .

[0050] S107: If the geometric prediction score reaches the preset geometric trigger threshold and the remaining distance is within the steerable window, then a candidate entry time is generated and the candidate entry time is added to the candidate set.

[0051] S108: Select the optimal entry start time from the candidate set.

[0052] when Reaching the preset geometric trigger threshold ,and fall into At that time, candidate entry times are generated. This is then added to the candidate set. Each candidate entry time can be associated with the branch angle, attitude deviation, remaining advance distance, allowable steering range, and risk label.

[0053] Select the optimal entry start time from among the candidate entry times in the candidate set.

[0054] In some embodiments, selecting the optimal entry start time from the candidate set includes: Calculate the product of the risk item and the risk penalty coefficient; The candidate entry time corresponding to the largest difference between the geometric prediction score and the product is taken as the optimal entry start time.

[0055] The optimal entry start time can be determined by the following formula: .

[0056] in, To be at the optimal entry point, For candidate set, Risks include being stuck to the wall, missing the entrance, the target branch entrance being invisible, or the posture not being able to be compensated for. This represents the risk penalty coefficient. The values ​​of the risk item and the risk penalty coefficient can be determined based on experience or experimentation.

[0057] The embodiments of this application do not perform a single-point judgment after the catheter reaches the bifurcation, but rather generate and update candidate entry times continuously before the catheter reaches the bifurcation, and select the best entry start time from them.

[0058] In some embodiments, it also includes: Determine the confidence levels for geometric predictions, image observations, and motion states; The timing confidence is determined based on the geometric prediction confidence, the image observation confidence, and the motion state confidence. Based on the timing confidence level and the first and second thresholds, suggestions are output; wherein, if the timing confidence level is greater than or equal to the first threshold and the remaining distance is within the steerable window, an entry suggestion is output; if the timing confidence level is greater than or equal to the second threshold and less than or equal to the first threshold, a deceleration suggestion, a pre-alignment suggestion, or a continue observation suggestion is output.

[0059] The embodiments of this application map the timing determination result into an executable steering or deceleration action, reducing repeated attempts caused by entering too early or too late.

[0060] Timing confidence Confidence can be predicted by geometry Image observation confidence and confidence level of motion state The result of fusion: .

[0061] in, When the duct is far from the bifurcation and the image entry point is not yet clear, increase the geometric prediction weight. When the duct is near its bifurcation and the target branch inlet profile is stable, the image observation weight can be increased. When there are significant changes in propulsion speed or obvious fluctuations in duct attitude, the penalty effect of motion state constraints on confidence can be increased. For example, the following settings can be configured: .

[0062] .

[0063] in, The distance threshold at which image observation begins to play a dominant role. This indicates that the weights will be limited to a preset upper and lower limit (the upper limit is...). The lower limit is ).

[0064] In some embodiments, determining the geometric prediction confidence level includes: The geometric prediction confidence is determined based on the window matching degree and the pose matching degree.

[0065] The geometric prediction confidence score can be obtained by weighted summing of window matching score and pose matching score. Alternatively, the geometric prediction confidence score can be obtained by weighted summing of window matching score, pose matching score, and risk inverse score, i.e.: .

[0066] In some embodiments, determining the confidence level of an image observation includes: The image observation confidence level is determined based on the consistency between the target branch inlet contour and the predicted inlet shape, image quality or inlet edge sharpness, and the change in inlet area between adjacent frames.

[0067] The change in the inlet area of ​​adjacent frames can refer to whether the change in the inlet area of ​​adjacent frames conforms to the monotonic change trend when the duct approaches the inlet of the target branch.

[0068] The confidence level of an image observation can be calculated in the following ways: .

[0069] in, This indicates the consistency between the target branch inlet profile and the predicted inlet shape. Indicates image quality or the sharpness of the entry edge. This indicates whether the change in the inlet area of ​​adjacent frames conforms to the monotonic change trend when the duct approaches the inlet of the target branch.

[0070] In some embodiments, determining the confidence level of the motion state includes: The confidence level of the motion state is determined based on the change in propulsion speed between adjacent cycles and the change in attitude deviation between adjacent cycles.

[0071] The confidence level of motion state is calculated as follows: .

[0072] in, This represents the change in propulsion speed between adjacent cycles. This represents the change in attitude deviation between adjacent cycles. and This is the tolerance parameter. The value of the tolerance parameter can be determined based on experience or experimentation.

[0073] The initial value for timing confidence can be determined based on preoperative pathway quality, bifurcation geometry recognition reliability, and initial catheter posture, for example: .

[0074] in, For the quality of preoperative pathways or navigation pathways, Extracting mass for bifurcation geometry parameters The initial orientation mass of the catheter. .

[0075] Set the recommended threshold for entry. That is, the first threshold and the hold observation threshold. That is, the second threshold and the low confidence threshold. ,satisfy: .

[0076] when and When, output input suggestions; when When, output deceleration, pre-alignment, or continued observation suggestions are provided; when , or continuous When the confidence level of geometric prediction decreases in a given period, or when the deviation between image observation and geometric prediction exceeds a threshold, a low-confidence strategy is triggered, which may involve rate limiting, shrinking the steerable window, realigning, or abnormal branching.

[0077] Entry suggestions can include the optimal entry start time, suggested advance speed, suggested turning direction, suggested turning range, and target branch entry alignment status.

[0078] In some embodiments, the method further includes the following before outputting a suggestion based on the timing confidence level and the first and second thresholds: The timing confidence is updated over time to output suggestions based on the timing confidence obtained from the time-recursive update and the first and second thresholds.

[0079] To suppress false triggering caused by jitter in a single frame image, the timing confidence can also be updated recursively over time. .

[0080] in, This is the smoothing coefficient. The system uses... As the final timing confidence output, suggestions are output based on this final timing confidence, thereby avoiding directly triggering the entry action based solely on the area difference between adjacent frames or the result of a single image matching.

[0081] In some embodiments, it also includes: When any one of the preset trigger conditions is met, deceleration and maintaining visual stability are prioritized; the preset trigger conditions include: the remaining distance falls into the steerable window and the timing confidence is less than the first threshold; the difference between the geometrically predicted entrance orientation and the image observation entrance orientation is greater than the preset threshold; the attitude deviation exceeds the preset compensable range; the propulsion speed exceeds the preset safety limit; and the optimal entry start time changes for multiple consecutive cycles.

[0082] When either of the following conditions is met: the current remaining distance has entered the steerable window but... The difference between the geometrically predicted inlet azimuth and the image-observed inlet azimuth exceeds a preset threshold; duct attitude deviation. If the compensation range is exceeded; the propulsion speed exceeds the safety limit; or the optimal entry time changes for N consecutive cycles; the system will prioritize deceleration and maintaining visual stability.

[0083] If the timing confidence level recovers to If the above is true, then restore the main chain output; if the confidence level is lower than 1, then restore the main chain output. If the situation continues to deteriorate, a rollback realignment, re-estimation of the bifurcation geometry, or manual intervention should be performed.

[0084] The method for predicting the timing of bronchoscope catheter entry into a branch provided in the above embodiments of this application can be implemented by a prediction system, which includes a geometric modeling module, a timing prediction module, an online correction module, and a conservative passage module.

[0085] The geometric modeling module establishes the main bronchial airway axis vector with the center point of the bifurcation as the origin. Target branch axis vector and the current axial vector of the conduit This yields the branch angle and the attitude deviation of the duct relative to the target branch.

[0086] The branch angle is: .

[0087] The attitude deviation of the catheter relative to the target branch is: .

[0088] The azimuth deviation of the target branch inlet relative to the center of the duct field of view is denoted as The remaining distance from the tip of the catheter to the center of the bifurcation is denoted as The current advance speed of the duct is denoted as The minimum allowable turning radius of the conduit is denoted as The maximum permissible steering angular velocity is denoted as The geometric modeling module will determine the branch angle. Posture deviation and azimuth deviation Convert to the required steering angle: .

[0089] in, , , The weights are preset or adaptive, and satisfy the following conditions: The required steering angle indicates the amount of attitude adjustment the duct needs to make before entering the target branch from its current position.

[0090] The steerable window is a time or distance range within which the conduit can align with the target branch inlet before reaching the bifurcation point without triggering the risk of wall adhesion or missing the inlet.

[0091] The time interval is represented as: .

[0092] in, This is the earliest time to suggest starting the turn. This is the latest time when it is still safe to enter the target branch.

[0093] The corresponding distance window is represented as: .

[0094] in, This indicates the distance required to begin turning. This indicates the latest turning distance. It represents the remaining distance from the tip of the catheter to the center of the bifurcation. fall into At this time, output can be entered into preparation or into suggestions.

[0095] The minimum time required to complete the required steering angle is: .

[0096] The corresponding minimum preceding distance is: .

[0097] in, The turning radius compensation coefficient is denoted as . The safety margin distance at the target branch entrance is denoted as . The redirectable window can then be represented as: .

[0098] when When, the system maintains a prediction or pre-alignment state; when When this happens, the system enters a suggested redirection state; when If the target branch entry alignment is not completed, the system determines that the optimal window has been missed or that a low-confidence-risk state has been entered.

[0099] Each control cycle The timing prediction module is based on the current remaining distance. Speed ​​of advancement and reversible windows Calculate window matching degree: .

[0100] in, Indicates the current remaining distance Distance to the center of the rotatable window This is the distance tolerance parameter.

[0101] The pose matching degree is: .

[0102] in, This is the angle tolerance parameter.

[0103] The geometric prediction score is: .

[0104] in, This is a risk inverse score calculated based on turning radius, safety distance, and entrance cross-section dimensions. .when Reaching the preset geometric trigger threshold ,and fall into At that time, the timing prediction module generates candidate entry times. And add it to the candidate set.

[0105] The optimal entry start time can be determined by the following formula: .

[0106] in, For the set of candidate entry times, Risks include being stuck to the wall, missing the entrance, the target branch entrance being invisible, or the posture not being able to be compensated for. This represents the risk penalty coefficient.

[0107] The online correction module will adjust the geometric prediction confidence level. Image observation confidence and confidence level of motion state Confidence level of timing obtained through fusion : .

[0108] in, When the duct is far from the bifurcation and the image entry point is not yet clear, increase the weight of geometric prediction; when the duct is close to the bifurcation and the target branch entry contour is stable, increase the weight of image observation; when the advancement speed changes significantly or the duct attitude fluctuates significantly, increase the penalty effect of motion state constraints on confidence. For example, it can be set as follows: .

[0109] .

[0110] in, The distance threshold at which image observation begins to play a dominant role. This means that the weights will be limited to the preset upper and lower limits.

[0111] The confidence score for geometric prediction can be calculated from the window matching score and the pose matching score: .

[0112] Image observation confidence can be calculated from the stability of the inlet contour, the change in inlet area between adjacent frames, and the consistency of the target branch inlet orientation: .

[0113] in, This indicates the consistency between the target branch inlet profile and the predicted inlet shape. Indicates image quality or the sharpness of the entry edge. This indicates whether the change in the inlet area of ​​adjacent frames conforms to the monotonic change trend when the duct approaches the inlet of the target branch.

[0114] The confidence level of motion state can be expressed as: .

[0115] in, This represents the change in propulsion speed between adjacent cycles. This represents the change in attitude deviation between adjacent cycles. and This is the tolerance parameter.

[0116] The initial value for timing confidence can be determined based on preoperative pathway quality, bifurcation geometry recognition reliability, and initial catheter posture, for example: .

[0117] in, For the quality of preoperative pathways or navigation pathways, Extracting mass for bifurcation geometry parameters The initial orientation mass of the catheter. .

[0118] In practical control, a suggested threshold can be set. Maintain observation threshold and low confidence threshold For example, satisfying: .

[0119] when and When, output input suggestions; when When, output deceleration, pre-alignment, or continued observation suggestions are provided; when , or continuous When the confidence level drops during a cycle, or when the deviation between image observation and geometric prediction exceeds a threshold, a low-confidence strategy is triggered, which may involve rate limiting, steerable window shrinkage, realignment, or abnormal branching.

[0120] To suppress false triggering caused by jitter in a single frame image, the timing confidence can also be updated recursively over time. .

[0121] in, The smoothing coefficient is used. The system is smoothed... As the final timing confidence output, it avoids directly triggering the entry action based solely on the area difference between adjacent frames or the result of a single image matching.

[0122] The conservative travel module is triggered when either of the following conditions is met: the current remaining distance has entered the steerable window but... The difference between the geometrically predicted inlet azimuth and the image-observed inlet azimuth exceeds a preset threshold; duct attitude deviation. Exceeding the compensable range; propulsion speed exceeding the safety limit; or continuous The optimal entry time for each cycle candidate changes abruptly. After the conservative passage module is triggered, the system prioritizes deceleration and maintaining a stable field of view. If the confidence level recovers to [a certain value], [the system will proceed as planned]. If the above is true, then restore the main chain output; if the confidence level is lower than 1, then restore the main chain output. If the situation continues to deteriorate, a rollback realignment, re-estimation of the bifurcation geometry, or manual intervention should be performed.

[0123] Based on the above embodiments, several specific implementation methods are described below.

[0124] Implementation Method 1: The angle between the target branch and the branch of the main bronchus The angle is 42°, representing the attitude deviation between the current axial direction of the catheter and the axis of the target branch. The azimuth deviation of the target branch entrance relative to the center of the field of view is 18°. The angle is 12°, representing the remaining distance from the catheter tip to the center of the bifurcation. 18mm, propulsion speed The maximum permissible angular velocity of the duct is 2 mm / s. Minimum allowable turning radius: 15° / s It is 10mm. (Setting) , , The required steering angle is: .

[0125] The corresponding minimum turning time is: .

[0126] If a turning radius compensation coefficient is set Safety margin distance The forward steering distance is approximately: .

[0127] The steerable distance window is: .

[0128] Therefore, when the catheter tip advances from 18mm to within approximately 6.63mm of the bifurcation, the system switches from prediction mode to preparation mode; when the distance falls between 3mm and 6.63mm and the timing confidence reaches the entry threshold, the output begins to shift towards a suggestion to enter the target branch; if the distance is less than 3mm but the inlet alignment is still not completed, it is determined that there is a risk of missing the optimal entry window.

[0129] Implementation Method Two: The system sets the recommended threshold. Maintain observation threshold Low confidence threshold When the conduit is far from the bifurcation point, for example... The image entry contour is still unstable; the weights are set to... , , Prediction is primarily based on geometry; when the duct is advanced to And when the entrance outline is visible for 5 consecutive frames, the weight is adjusted to , , When the catheter is advanced to Furthermore, when the edge sharpness of the entrance is greater than 0.75, the weight is further adjusted to... , , .

[0130] For example, in a certain period, the geometric prediction confidence level Image observation confidence Motion state confidence And currently in In the near-entry phase, the timing confidence level is: .

[0131] because ,and Within the steerable distance window, the system outputs entry suggestions. If, at the same distance, obstruction is caused by blood, secretions, or visual field obstruction... If it drops to 0.50, then: .

[0132] at this time The system does not directly output the entry action, but instead outputs suggestions for deceleration, maintaining the field of view, and continuing pre-alignment. If this occurs for three consecutive cycles... If the difference between the image observation entry azimuth and the geometric prediction entry azimuth exceeds 15°, the risk gating weight will be increased and the steerable window will be contracted, limiting the allowable steerable amplitude to within 70% of the original recommended value, while reducing the advance speed from 2 mm / s to 0.5-1 mm / s.

[0133] Implementation Method 3: The system is set to trigger a low-confidence condition with a smoothed confidence level. ; or the change in the optimal entry time over four consecutive cycles is greater than 0.8 s; or the catheter attitude deviation ; or the difference between the predicted entrance azimuth and the image observation is greater than 20°; or the current remaining distance Furthermore, the entry alignment conditions are still not met. Upon triggering, the system first performs a first-level degradation: limiting the propulsion speed to no more than 0.5 mm / s and freezing direct entry to output, allowing only output deceleration, holding, and fine-tuning alignment suggestions.

[0134] If a level 1 downgrade occurs within 5 consecutive cycles If the value recovers to above 0.60 and the duct attitude deviation decreases to within 20°, the system exits Level 1 degradation and resumes steerable window calculation; if If the value remains below 0.45, or the duct is less than 2mm from the bifurcation and the target branch inlet is still not stably identifiable, then proceed to Level 2 downgrade, outputting a suggestion to retract 2-5mm, re-estimate the target branch inlet orientation, and recalculate the steerable window; if two consecutive retractions and re-evaluations still fail to resolve the issue... If the score reaches 0.60 or higher, or if the risk score of catheter apposition to the wall is greater than 0.80, a manual intervention prompt will be triggered.

[0135] For example, in a certain period , , The weight is , , ,but: .

[0136] After time smoothing If the target branch entry orientation deviation is 23°, the system triggers a low-confidence degradation, no longer executes the entry suggestion, and instead outputs a safety policy of rate limiting, rollback realignment, or manual takeover.

[0137] In summary, the bronchoscopic catheter entry timing prediction method provided in this application determines the branch angle between the main bronchus and the target branch, as well as the catheter's attitude deviation relative to the target branch, and calculates the required turning angle by integrating the orientation deviation. This determines the possible turning window, transforming fuzzy empirical judgments into precise spatiotemporal boundaries, fundamentally avoiding the physical risks of turning too early or too late. Furthermore, this application determines the window matching degree based on the distance from the remaining distance to the center of the possible turning window, the attitude matching degree based on the attitude deviation and orientation deviation, and the geometric prediction score based on the window matching degree and attitude matching degree. A candidate entry time is generated only when the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the possible turning window. This effectively eliminates the risk of false triggering caused by image noise or single distance estimation errors, greatly improving the reliability of timing determination. Finally, by optimizing the candidate set, this application can converge and output the optimal entry start time. Throughout the entire process of the catheter approaching the bifurcation, it can proactively and continuously provide an optimal entry start time, reducing human trial and error and empirical bias.

[0138] This application also provides a device for predicting the timing of bronchoscope catheter entry into a branch, which is described below and can be referred to in conjunction with the method described above. Please refer to... Figure 2 , Figure 2 This is a schematic diagram of a device for predicting the timing of a bronchoscope catheter entering a branch, provided in an embodiment of this application. Figure 2 As shown, the device includes: The acquisition unit 10 is used to acquire the branch angle between the main bronchus and the target branch, the attitude deviation of the duct relative to the target branch, the orientation deviation of the target branch inlet relative to the center of the duct's field of view, and the remaining distance from the tip of the duct to the center of the bifurcation. The first determining unit 20 is used to determine the required turning angle based on the branch angle, the attitude deviation, and the orientation deviation; the required turning angle is used to represent the amount of attitude adjustment that the catheter needs to complete to enter the target branch from the current position; The second determining unit 30 is used to determine the steerable window according to the required steering angle; the steerable window is the time interval or distance interval in which the duct can complete the alignment of the target branch inlet before reaching the bifurcation without triggering the risk of wall adhesion or missing the inlet. The third determining unit 40 is used to determine the window matching degree based on the distance from the remaining distance to the center of the rotatable window; The fourth determining unit 50 is used to determine the attitude matching degree based on the attitude deviation and the orientation deviation; The fifth determining unit 60 is used to determine the geometric prediction score based on the window matching degree and the pose matching degree; The generation unit 70 is used to generate candidate entry times and add the candidate entry times to the candidate set if the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the steerable window; Selection unit 80 is used to select the optimal entry start time from the candidate set.

[0139] In some embodiments, it also includes: The sixth determining unit is used to determine the confidence level of geometric prediction, the confidence level of image observation, and the confidence level of motion state; The seventh determining unit is used to determine the timing confidence based on the geometric prediction confidence, the image observation confidence, and the motion state confidence. The output unit is used to output suggestions based on the timing confidence and a first threshold and a second threshold; wherein, if the timing confidence is greater than or equal to the first threshold and the remaining distance is within the steerable window, an entry suggestion is output; if the timing confidence is greater than or equal to the second threshold and less than or equal to the first threshold, a deceleration suggestion, a pre-alignment suggestion, or a continue observation suggestion is output.

[0140] In some embodiments, it also includes: The recursive update unit is used to perform time-recursive updates on the timing confidence, so as to output suggestions based on the timing confidence obtained by the time-recursive update and the first threshold and the second threshold.

[0141] In some embodiments, it also includes: An execution unit is configured to prioritize deceleration and maintaining visual stability when any one of the preset triggering conditions is met. The preset triggering conditions include: the remaining distance falls into the steerable window and the timing confidence is less than the first threshold; the difference between the geometrically predicted entrance orientation and the image observation entrance orientation is greater than a preset threshold; the attitude deviation exceeds a preset compensable range; the propulsion speed exceeds a preset safety limit; and the optimal entry start time changes for multiple consecutive cycles.

[0142] In some embodiments, the second determining unit 30 is used to: The minimum time to complete the required steering angle is determined based on the required steering angle and the maximum permissible steering angular velocity; The minimum preceding distance is determined based on the minimum time. The steerable window is determined based on the minimum leading distance and the safety margin distance.

[0143] In some embodiments, the selection unit 80 is used for: Calculate the product of the risk item and the risk penalty coefficient; The candidate entry time corresponding to the largest difference between the geometric prediction score and the product is taken as the optimal entry start time.

[0144] In some embodiments, the sixth determining unit is used for: The geometric prediction confidence is determined based on the window matching degree and the pose matching degree.

[0145] In some embodiments, the sixth determining unit is used for: The image observation confidence level is determined based on the consistency between the target branch inlet contour and the predicted inlet shape, image quality or inlet edge sharpness, and the change in inlet area between adjacent frames.

[0146] In some embodiments, the sixth determining unit is used for: The confidence level of the motion state is determined based on the change in propulsion speed between adjacent cycles and the change in attitude deviation between adjacent cycles.

[0147] This application also provides an electronic device, referenced... Figure 3 As shown, the device includes a memory 1 and a processor 2.

[0148] Memory 1 is used to store computer programs; Processor 2 is used to execute computer programs to perform the following steps: The following steps are taken: 1. Obtain the branch angle between the main bronchus and the target branch; 2. The attitude deviation of the duct relative to the target branch; 3. The azimuth deviation of the target branch inlet relative to the center of the duct's field of view; 4. The remaining distance from the duct tip to the center of the bifurcation. 5. Determine the required turning angle based on the branch angle, attitude deviation, and azimuth deviation. The required turning angle represents the amount of attitude adjustment required for the duct to enter the target branch from its current position. 6. Determine a steerable window based on the required turning angle. 7. Determine the window matching degree based on the distance from the remaining distance to the center of the steerable window. 8. Determine the attitude matching degree based on the attitude deviation and azimuth deviation. 9. Determine the geometric prediction score based on the window matching degree and attitude matching degree. 10. If the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the steerable window, generate a candidate entry time and add it to the candidate set. 11. Select the optimal entry start time from the candidate set.

[0149] For a description of the equipment provided in this application, please refer to the above method embodiments; further details will not be provided here.

[0150] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the following steps: The following steps are taken: 1. Obtain the branch angle between the main bronchus and the target branch; 2. The attitude deviation of the duct relative to the target branch; 3. The azimuth deviation of the target branch inlet relative to the center of the duct's field of view; 4. The remaining distance from the duct tip to the center of the bifurcation. 5. Determine the required turning angle based on the branch angle, attitude deviation, and azimuth deviation. The required turning angle represents the amount of attitude adjustment required for the duct to enter the target branch from its current position. 6. Determine a steerable window based on the required turning angle. 7. Determine the window matching degree based on the distance from the remaining distance to the center of the steerable window. 8. Determine the attitude matching degree based on the attitude deviation and azimuth deviation. 9. Determine the geometric prediction score based on the window matching degree and attitude matching degree. 10. If the geometric prediction score reaches a preset geometric trigger threshold and the remaining distance is within the steerable window, generate a candidate entry time and add it to the candidate set. 11. Select the optimal entry start time from the candidate set.

[0151] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0152] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0153] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatuses, devices, and computer-readable storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant details can be found in the method section.

[0154] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0155] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0156] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this application. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for predicting the timing of bronchoscope catheter entry into a branch, characterized in that, include: Obtain the branch angle between the main bronchus and the target branch, the attitude deviation of the catheter relative to the target branch, the orientation deviation of the target branch inlet relative to the center of the catheter's field of view, and the remaining distance from the catheter tip to the center of the bifurcation. The required turning angle is determined based on the branch angle, the attitude deviation, and the orientation deviation; the required turning angle is used to represent the amount of attitude adjustment that the catheter needs to make to enter the target branch from the current position; The steerable window is determined based on the required steering angle; the steerable window is the time or distance interval within which the duct can complete the alignment of the target branch inlet before reaching the bifurcation without triggering the risk of wall adhesion or missing the inlet. The window matching degree is determined based on the distance from the remaining distance to the center of the rotatable window; The attitude matching degree is determined based on the attitude deviation and the orientation deviation; The geometric prediction score is determined based on the window matching degree and the pose matching degree. If the geometric prediction score reaches the preset geometric trigger threshold and the remaining distance is within the steerable window, then a candidate entry time is generated and the candidate entry time is added to the candidate set; The optimal entry start time is selected from the candidate set.

2. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 1, characterized in that, Also includes: Determine the confidence levels for geometric predictions, image observations, and motion states; The timing confidence is determined based on the geometric prediction confidence, the image observation confidence, and the motion state confidence. Based on the timing confidence level and the first and second thresholds, a suggestion is output; wherein, if the timing confidence level is greater than or equal to the first threshold and the remaining distance is within the steerable window, an entry suggestion is output. If the timing confidence level is greater than or equal to the second threshold and less than or equal to the first threshold, then a deceleration suggestion, a pre-alignment suggestion, or a continued observation suggestion is output.

3. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 2, characterized in that, Before outputting suggestions based on the aforementioned timing confidence level and the first and second thresholds, the following further steps are included: The timing confidence is updated over time to output suggestions based on the timing confidence obtained from the time-recursive update and the first and second thresholds.

4. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 3, characterized in that, Also includes: If any of the preset trigger conditions are met, deceleration and maintaining a stable field of view will be prioritized. The preset triggering conditions include: the remaining distance falls into the steerable window and the timing confidence is less than the first threshold; the difference between the geometrically predicted entry azimuth and the image observation entry azimuth is greater than a preset threshold; the attitude deviation exceeds a preset compensable range; the propulsion speed exceeds a preset safety limit; and the optimal entry start time changes for multiple consecutive cycles.

5. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 1, characterized in that, The steering window is determined based on the required steering angle, including: The minimum time to complete the required steering angle is determined based on the required steering angle and the maximum permissible steering angular velocity; The minimum preceding distance is determined based on the minimum time. The steerable window is determined based on the minimum leading distance and the safety margin distance.

6. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 1, characterized in that, Selecting the optimal entry start time from the candidate set includes: Calculate the product of the risk item and the risk penalty coefficient; The candidate entry time corresponding to the largest difference between the geometric prediction score and the product is taken as the optimal entry start time.

7. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 2, characterized in that, Determining the confidence level of a geometric prediction includes: The geometric prediction confidence is determined based on the window matching degree and the pose matching degree.

8. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 2, characterized in that, Determining the confidence level of image observations includes: The image observation confidence level is determined based on the consistency between the target branch inlet contour and the predicted inlet shape, image quality or inlet edge sharpness, and the change in inlet area between adjacent frames.

9. The method for predicting the timing of bronchoscope catheter entry into a branch according to claim 2, characterized in that, Determining the confidence level of motion states includes: The confidence level of the motion state is determined based on the change in propulsion speed between adjacent cycles and the change in attitude deviation between adjacent cycles.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the method for predicting the timing of bronchial catheter entry into a branch as described in any one of claims 1 to 9.

Images