A gastric tube trajectory monitoring system and apparatus

Abstract

The application relates to a gastric tube trajectory monitoring system and device, wherein the gastric tube trajectory monitoring system comprises a trajectory acquisition module for acquiring a running trajectory of a gastric tube insertion end based on a real-time position; a trajectory analysis module for obtaining a real-time deviation of a target trajectory from the running trajectory, determining a deviation position and an adjustment angle according to the real-time deviation; a trajectory topology graph construction module for constructing a trajectory point cloud set according to the running trajectory, determining an abnormal point set according to a trajectory topology graph group based on a persistent homology mechanism; a real-time risk early warning module for synchronizing the running trajectory to a pre-obtained trachea-esophagus model, and performing real-time risk early warning according to the position of the abnormal point set in the trachea-esophagus model; and a potential risk early warning module for inputting the real-time position into a preset trajectory prediction model, obtaining a predicted trajectory at a next moment, and performing potential risk early warning according to the deviation of the predicted trajectory from the target trajectory.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical auxiliary, in particular to a gastric tube trajectory monitoring system and device. BACKGROUND

[0002] Gastric tube insertion is a common medical operation in clinical medicine, which realizes the patient's nutritional support, drug infusion and gastrointestinal function recovery by inserting a gastric tube into the patient's stomach. The traditional gastric tube insertion technique relies on the experience of medical personnel for operation, and the gastric tube position is determined by observing the insertion depth of the gastric tube and the patient's reaction to intubation. In complex cases, there is a risk of gastric tube misplacement into the trachea or other incorrect sites.

[0003] The existing gastric tube positioning technology mainly relies on imaging assistance or physical positioning devices. The imaging method includes providing real-time position images of the gastric tube through X-ray or CT scan technology, but there are problems of radiation exposure and complex operation. The physical positioning method provides a trajectory path to medical personnel through a pressure sensor or an electromagnetic positioning system, but the accuracy and real-time performance are limited. Therefore, how to realize real-time and accurate trajectory monitoring during gastric tube insertion has become a problem to be solved. SUMMARY

[0004] The embodiments of the present application provide a gastric tube trajectory monitoring system and device to at least solve the problem of inability to realize real-time and accurate trajectory monitoring during gastric tube insertion in the related art.

[0005] In a first aspect, the embodiments of the present application provide a gastric tube trajectory monitoring system, characterized in that the monitoring system is applied to a gastric tube trajectory monitoring device, the device includes a gastric tube, and a metal ring is arranged at the insertion end of the gastric tube. The monitoring system includes:

[0006] a trajectory acquisition module, configured to acquire a real-time position of the metal ring during the insertion process of the gastric tube, and acquire a running trajectory of the insertion end of the gastric tube based on the real-time position;

[0007] a trajectory analysis module, configured to determine a target trajectory based on a pre-obtained gastric tube insertion path model, acquire a real-time deviation between the target trajectory and the running trajectory, and determine a deviation position and an adjustment angle according to the real-time deviation;

[0008] a trajectory topology graph construction module, configured to construct a trajectory point cloud set according to the running trajectory, construct a trajectory topology graph group based on a preset radius sequence and the trajectory point cloud set, and determine an abnormal point set according to the trajectory topology graph group based on a persistent homology mechanism;

[0009] A real-time risk warning module is configured to synchronize the running track to a pre-obtained trachea-esophagus model, and perform real-time risk warning according to positions of the abnormal point set in the trachea-esophagus model.

[0010] A potential risk warning module is configured to input the real-time position into a preset track prediction model, obtain a predicted track at a next time, and perform potential risk warning according to a deviation between the predicted track and the target track.

[0011] In an embodiment, the constructing a track topology graph group based on the preset radius sequence and the track point cloud set comprises:

[0012] If a distance between any two points in the track point cloud set is less than or equal to the preset radius, an edge of the two points is obtained.

[0013] If distances between any point pairs in a point set in the track point cloud set are all less than or equal to the preset radius, a simplex is obtained, and a number of points in the point set is greater than or equal to 3.

[0014] A track topology graph of the preset radius is constructed according to the points, the edges and the simplex.

[0015] A track topology graph group is obtained according to the preset radius sequence, and the track topology graphs under each preset radius are constructed.

[0016] In an embodiment, the determining an abnormal point set based on a persistent homology mechanism according to the track topology graph group comprises:

[0017] A topology feature of each track topology graph is determined based on the track topology graph group.

[0018] A coordinate system is constructed with the preset radius sequence as an abscissa and the topology feature as an ordinate, a life cycle of the topology feature is determined according to a preset radius at which the topology feature first appears as a starting point and a preset radius at which the topology feature disappears as an ending point, and a persistent bar chart is obtained by displaying the life cycle of each topology feature in the coordinate system.

[0019] An abnormal topology feature is determined by analyzing a life cycle distribution in the persistent bar chart, the abnormal topology feature is mapped to the track point cloud set, and an abnormal point set is determined.

[0020] In an embodiment, the topology feature comprises a connected component, and the determining an abnormal topology feature by analyzing a life cycle distribution in the persistent bar chart, mapping the abnormal topology feature to the track point cloud set, and determining an abnormal point set comprises:

[0021] If a life cycle of any one connected component is less than a preset life cycle threshold, the connected component is an unstable connected component;

[0022] In the persistent bar graph, breaking points of all the unstable connected components are determined, and the breaking points are clustered to obtain a plurality of cluster clusters;

[0023] In each of the cluster clusters, a target breaking point is determined, a first topological graph is reconstructed based on the target breaking point and the trajectory point cloud set, vertex coordinates of the unstable connected component are determined in the first topological graph, and a first abnormal point set in the trajectory point cloud set is determined according to the vertex coordinates.

[0024] In an embodiment, the topological feature includes a ring, the analysis of the life cycle distribution in the persistent bar graph determines an abnormal topological feature, the abnormal topological feature is mapped to the trajectory point cloud set, and the determination of the abnormal point set includes:

[0025] If a life cycle of any one ring is greater than or equal to a preset life cycle threshold, the ring is a stable ring;

[0026] All the stable rings in the persistent bar graph are determined, a generator of the stable ring is extracted, coordinates constituting the stable ring are determined according to the generator, and a second abnormal point set in the trajectory point cloud set is determined according to the coordinates of the ring.

[0027] In an embodiment, the real-time risk warning according to the position of the abnormal point set in the trachea-esophageal model includes:

[0028] If a center of the first abnormal point set is located in an esophageal stenosis region in the trachea-esophageal model, a real-time insertion path deviation risk warning is performed;

[0029] If a geometric center of the second abnormal point set is located in a tracheal region in the trachea-gastric tube model, a real-time tracheal entry risk warning is performed.

[0030] In an embodiment, the synchronization of the running trajectory to the pre-obtained trachea-esophageal model includes:

[0031] An esophageal inlet position and a cardia position in the trachea-esophageal model are determined, and a first point set of the esophageal inlet position and a second point set of the cardia position are obtained, wherein the esophageal inlet position is a position where the esophagus and the pharynx connect, and the cardia position is a position where the esophagus and the stomach connect;

[0032] In the trajectory point cloud set, a third point set of esophageal positions and a fourth point set of cardia positions in the running trajectory are determined, alignment and matching of the first point set and the third point set and the second point set and the fourth point set are realized by an ICP algorithm to obtain a rotation matrix and a translation vector;

[0033] The running trajectory is adjusted based on the rotation matrix and the translation vector to complete synchronization of the running trajectory to the pre-obtained trachea-esophagus model.

[0034] In an embodiment, the potential risk warning is performed according to a deviation of the predicted trajectory from the target trajectory, including:

[0035] A target coordinate is obtained based on the target trajectory, a predicted coordinate is obtained based on the predicted trajectory, and a distance deviation between the target coordinate and the predicted coordinate is obtained.

[0036] A target turning angle is determined based on a plurality of target coordinates, a predicted turning angle is determined based on a plurality of predicted coordinates, and an angle deviation of the target turning angle and the predicted turning angle is obtained.

[0037] When the distance deviation is greater than a preset distance threshold, a potential risk warning is performed; and / or

[0038] When the angle deviation is greater than a preset angle threshold, a potential risk warning is performed.

[0039] In an embodiment, the gastric tube trajectory monitoring device further includes a processing terminal, and the method further includes:

[0040] The running trajectory and the predicted trajectory are displayed in the processing terminal.

[0041] In a second aspect, the embodiments of the present application provide a gastric tube trajectory monitoring device, which includes a processing terminal, a camera device, a metal ring, and a metal detection head.

[0042] The metal ring is located outside the gastric tube insertion end and surrounds the gastric tube insertion end.

[0043] The metal detection head includes a receiving coil and a transmitting coil, the transmitting coil is used to generate a variable magnetic field, the receiving coil is used to receive a changing magnetic flux of the metal ring when the metal ring moves in the variable magnetic field and obtain an induced voltage, the metal detection head is electrically connected with the processing terminal, and the induced voltage of the receiving coil is transmitted to the processing terminal.

[0044] The processing terminal is used to receive the induced voltage to obtain a trajectory of the movement of the metal ring and display the trajectory of the movement of the metal ring.

[0045] The gastric tube trajectory monitoring system and device provided in this application embodiment have at least the following technical effects.

[0046] The real-time position of the metal ring is acquired by the trajectory acquisition module to obtain the trajectory of the gastric tube insertion end. The real-time deviation between the running trajectory and the target trajectory is used to determine the offset position and adjustment angle, allowing operators to make timely adjustments. A trajectory topology map is constructed from the trajectory point cloud set of the running trajectory, and an anomaly set in the trajectory point cloud set is identified using a continuous synchronization mechanism. This continuous synchronization mechanism, by processing complex geometric and topological information, can identify point sets that do not follow the overall pattern or disrupt the expected structural consistency, thereby improving the accuracy of obtaining anomaly set information. Based on the position of the anomaly set in the tracheoesophageal model, real-time risk warnings are issued to monitor the trajectory of the gastric tube insertion end in real time, enabling timely detection of abnormal risks during the insertion process. Based on the deviation between the running trajectory and the predicted trajectory, potential risk warnings are issued to detect the risk of deviation during insertion in advance. Potential risk warnings can prevent errors during insertion, such as inserting the gastric tube into other organs. The continuous synchronization mechanism can accurately identify abnormal trajectories in the movement trajectory, thus enabling precise control of the gastric tube insertion operation. Real-time risk warnings and potential risk warnings achieve risk prevention during the insertion process.

[0047] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0048] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0049] Figure 1 This is a schematic diagram of a gastric tube trajectory monitoring system according to an exemplary embodiment;

[0050] Figure 2 This is a schematic diagram illustrating a gastric tube trajectory monitoring device according to an exemplary embodiment.

[0051] In the above figures, the meanings of the reference numerals are as follows:

[0052] 1. Metal ring, 2. Metal detector head, 3. Fixing strap, 4. Processing terminal, 5. Gastric tube, 6. Camera equipment. Detailed Implementation

[0053] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. Based on the embodiments provided by the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort belong to the scope of the present application.

[0054] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application, and for those of ordinary skill in the art, the present application can also be applied to other similar scenarios without creative effort based on the accompanying drawings. In addition, it can be understood that although the efforts made in the development process can be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacture or production changes based on the technical content disclosed in the present application are only routine technical means and should not be understood as insufficient disclosure of the content disclosed in the present application.

[0055] In the present application, the phrase "embodiments" means that the specific features, structures or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the present application. The appearance of this phrase at various places in the specification does not necessarily mean the same embodiment, nor is it an independent or alternative embodiment to other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in the present application can be combined with other embodiments without conflict.

[0056] Unless otherwise defined, technical terms and scientific terms used in the present application shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. Unless otherwise defined, the terms "one" and "a" or "an" shall not be construed to mean "at least one" or "one or more". Unless otherwise defined, the terms "including," "comprising," "attached," "coupled," "connected," and the like are not limited to direct connections, but can include indirect connections unless otherwise contextually implied. The term "plurality" means two or more. The term "and / or" describes associated objects in association relationships, which means that there are three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. The character " / " generally represents an "or" relationship between the front and rear associated objects. The terms "first", "second", "third", and the like are merely to distinguish similar objects, and do not represent a specific order of the objects.

[0057] In a first aspect, the embodiments of the present application provide a gastric tube trajectory monitoring system, Figure 1 is a schematic diagram of a gastric tube trajectory monitoring system according to an exemplary embodiment. As shown in the figure, the monitoring system is applied to a gastric tube trajectory monitoring device, the device includes a gastric tube, and a metal ring is arranged at the insertion end of the gastric tube. The gastric tube trajectory monitoring system includes: Figure 1

[0058] The trajectory acquisition module is configured to acquire a real-time position of the metal ring during the insertion process of the gastric tube, and acquire a running trajectory of the insertion end of the gastric tube based on the real-time position.

[0059] The gastric tube trajectory monitoring device includes a gastric tube provided with a metal ring and a metal detection probe. The metal detection probe is provided with an array of inductance coils. The insertion process of the gastric tube is equivalent to the movement process of the metal ring. The array of inductance coils in the metal detection probe generates a magnetic field, and each inductance coil covers a different range. When the metal ring moves in the magnetic field, the metal ring changes the magnetic flux of the magnetic field, thereby changing the induced voltage of the inductance coils in the array. According to the change of the induced voltage of different coils, the real-time position of the metal ring is determined. The real-time position of the metal ring is the spatial position at each time, and different real-time positions constitute the running trajectory of the insertion end of the gastric tube. The real-time position of the metal ring is the three-dimensional coordinates of the geometric center of the metal ring, and the running trajectory is composed of a plurality of three-dimensional coordinates. ​

[0060] In addition, the insertion speed and the turning angle of the gastric tube can be obtained based on the running trajectory. For the turning angle, any three consecutive points in the running trajectory are selected, a vector is determined by the adjacent two points, and the turning angle formed by the three points is obtained based on the vector. Alternatively, curve fitting is performed on a plurality of consecutive points, the curvature is obtained according to the fitted curve, and the turning angle is obtained by calculating the curvature.

[0061] For the insertion speed, the real-time insertion speed of the gastric tube is obtained by using the position difference based on the time and coordinates of the real-time position of the metal ring. All the obtained real-time insertion speeds are subjected to sliding window or Kalman filtering for smoothing processing to reduce the influence of noise and obtain more accurate insertion speed.

[0062] For the insertion depth, the scale exposed by the gastric tube is photographed by the camera of the gastric tube trajectory monitoring device to obtain the current insertion depth of the gastric tube.

[0063] The trajectory acquisition module acquires the running trajectory of the gastric tube in the body in real time, and obtains the turning angle and the insertion speed to provide a comprehensive and accurate basis for subsequent analysis of the running trajectory.

[0064] The trajectory analysis module is used to determine a target trajectory based on a pre-obtained gastric tube insertion path model, obtain a real-time deviation of the target trajectory and the running trajectory, and determine a deviation position and an adjustment angle according to the real-time deviation.

[0065] The gastric tube insertion path model is obtained by scanning the region to be passed through by the gastric tube in the patient's body by using a three-dimensional computer tomography technique. The target trajectory is determined in the gastric tube insertion path model, wherein the target trajectory avoids all potential risk points of the patient, such as blood vessels, weak regions of organs, scar tissues left by previous surgeries, and has feasibility.

[0066] The real-time coordinates of the target trajectory and the running trajectory are obtained, the real-time coordinates are subjected to smoothing processing to remove noise when the real-time coordinates are collected, and the time of each coordinate in the target trajectory and the running trajectory is synchronized to ensure that the coordinates in the target trajectory and the running trajectory correspond one by one.

[0067] The real-time path deviation between the real-time coordinates of the target trajectory and the running trajectory is calculated by using the Euclidean distance, and the real-time turning angle deviation between the target trajectory and the running trajectory is calculated by using the turning angle difference formula. The real-time path deviation specifically satisfies the following formula:

[0068]

[0069] wherein D(t) represents the path deviation distance at time t, x r (t) represents the x-axis coordinate of the running trajectory at time t, y r(t) represents the y-axis coordinate of the running trajectory at time t, z r (t) represents the z-axis coordinate of the running trajectory at time t; x i (t) represents the x-axis coordinate of the target trajectory at time t, y i (t) represents the y-axis coordinate of the target trajectory at time t, z i (t) represents the z-axis coordinate of the target trajectory at time t

[0070] The real-time turning angle deviation specifically satisfies the following formula:

[0071]

[0072] Wherein, Δ θ ( t ) represents the turning angle deviation of the running trajectory and the target trajectory at time t, θ r ( t ) represents the turning angle of the running trajectory at time t, θ i ( t ) represents the turning angle of the target trajectory at time t, represents the direction vector of the first section of the running path in the running trajectory, represents the direction vector of the second section of the running path in the running trajectory, represents the direction vector of the first section of the running path in the target trajectory, represents the direction vector of the second section of the running path in the target trajectory.

[0073] The real-time path deviation is compared with the preset path deviation threshold, the real-time turning angle deviation is compared with the preset turning angle threshold, and if any one of the real-time path deviation greater than the preset path deviation threshold and the real-time turning angle deviation greater than the preset turning angle threshold is satisfied, the deviation coordinate of the real-time deviation in the running trajectory is determined, and the adjustment angle is determined according to the real-time turning angle deviation. And prompt "slight trajectory deviation" and display deviation coordinate and adjustment angle on the processing terminal, so as to assist the operator to adjust the inserted gastric tube according to the offset in the cannula, prevent the running trajectory of the gastric tube from deviating from the target trajectory, and also avoid the gastric tube from entering the trachea or mispenetrating the tissue.

[0074] The trajectory topology graph module is used to construct a trajectory point cloud set according to the running trajectory, construct a trajectory topology graph group based on the preset radius sequence and the trajectory point cloud set, and determine an abnormal point set based on the persistent homology mechanism according to the trajectory topology graph group.

[0075] The trajectory data of each position in the running trajectory is subjected to noise filtering and normalization processing, the position coordinates of each time in the processed running trajectory are taken as a trajectory point, and a trajectory point cloud set is constructed based on all the trajectory points in the running trajectory.

[0076] The distance between any two trajectory points is calculated, and the maximum distance is determined among all distances. Half of the maximum distance is taken as the maximum radius threshold of the preset radius sequence to prevent the topology graph formed from being too dense. The preset radius sequence is an increasing radius sequence. Each preset radius satisfies the following formula: wherein, is the kth preset radius, is a fixed step size, is a natural number, , T.

[0077] Under any preset radius, if the distance between any two trajectory points in the trajectory point cloud set is less than or equal to the preset radius, the two points are connected to form an edge in the trajectory topology graph.

[0078] If the distance between any point pair in any point set in the trajectory point cloud set is less than or equal to the preset radius, a simplex is obtained, and the number of points in the point set is greater than or equal to 3. For example, there are three trajectory points in a point set in the trajectory point cloud set, and the distance between any two of the three trajectory points is less than or equal to the preset radius, a two-dimensional simplex, i.e., a triangular face formed by connecting the three trajectory points, is obtained.

[0079] The trajectory topology graph under the current preset radius is constructed by taking all trajectory points as vertices of the trajectory topology graph and determining edges and simplices based on the distances between the trajectory points. According to the preset radius sequence, the trajectory topology graph under each preset radius is obtained, and these trajectory topology graphs form a trajectory topology graph group.

[0080] Continuing to refer to the step of determining the abnormal point set based on the persistent homology mechanism according to the trajectory topology graph group in the step of constructing the trajectory topology graph, the step specifically includes:

[0081] Based on the trajectory topology graph group, the topological features of each trajectory topology graph are determined. The homology groups are obtained according to the trajectory topology graph, and the homology groups include zero-order homology groups and first-order homology groups. The topological features of the trajectory topology graph are determined based on the homology groups. The topological feature determined by the zero-order homology group is a connected component, which represents the connection between any two trajectory points in the topology graph. The topological feature determined by the first-order homology group is a loop, which represents that the trajectory points form a closed path in the topology graph.

[0082] A coordinate system is constructed with a preset radius sequence as the horizontal coordinate and the topological features as the vertical coordinate. The life cycle of the topological features is determined according to the preset radius at which the topological features first appear as the starting point and the preset radius at which the topological features disappear as the ending point. The life cycle of each topological feature is displayed in the coordinate system to obtain a persistent bar chart. In the trajectory topological graph, there are several quantities of the same topological feature, and there are also multiple life cycles of the topological features in the persistent bar chart. Therefore, in the persistent bar chart, the same topological feature will display different life cycles, and the life cycles are displayed as horizontal line segments parallel to the horizontal coordinate.

[0083] The distribution of the life cycles in the persistent bar chart is analyzed to determine an abnormal topological feature. The abnormal topological feature is mapped to the trajectory point cloud set to determine an abnormal point set.

[0084] For the life cycle of a connected component, if the life cycle of any connected component is less than a preset life cycle threshold, the connected component is an unstable connected component. If there is a break or deviation in the running trajectory, more connected components will be formed in the trajectory graph, and the connected components generated due to the deviation or break are usually unstable connected components. Therefore, determining the unstable connected components can exist abnormal trajectories.

[0085] In the persistent bar chart, the end point of the unstable connected component is determined according to the life cycle, and the preset radius corresponding to the end point is taken as a break point. The density clustering is performed on all the break points to obtain a plurality of clustering clusters. A target break point, i.e., a target radius, is determined in each clustering cluster. The trajectory topological graph is reconstructed based on the target radius and the trajectory point cloud set, the vertex coordinates of the unstable connected component are determined in the trajectory topological graph, and a first abnormal point set in the trajectory point cloud set is determined according to the vertex coordinates.

[0086] For the life cycle of a ring, if the life cycle of any ring is greater than or equal to a preset life cycle threshold, the ring is a stable ring. The stable ring represents that the trajectory forms a closed path in the running trajectory. All the stable rings are determined in the persistent bar chart, and the generators of the stable rings are extracted. The coordinates constituting the stable rings are determined according to the generators, and a second abnormal point set in the trajectory point cloud set is determined according to the coordinates of the rings. Each point in the second abnormal point set constitutes a closed ring in distribution.

[0087] Continuing to refer to the trajectory topological graph construction module, the topological features of the running trajectory are obtained by the persistent homology mechanism, and then the abnormal point sets in the trajectory point cloud set can be determined. These abnormal point sets reflect the risks existing in the running trajectory. Timely warning of these risks can help the operator to make timely adjustments, and the specific warning method is described in the real-time risk warning module.

[0088] The real-time risk warning module is used for synchronizing the running track to the pre-obtained trachea-esophagus model, and performing real-time risk warning according to the positions of the abnormal point set in the trachea-esophagus model.

[0089] The pre-obtained trachea-esophagus model is obtained by scanning the patient by a computer tomography technique. Synchronizing the running track to the pre-obtained trachea-esophagus model specifically includes:

[0090] The esophageal inlet position and the cardia position in the trachea-esophagus model are determined, and a first point set of the esophageal inlet position and a second point set of the cardia position are obtained, wherein the esophageal inlet position is the position where the esophagus and the pharynx connect, and the cardia position is the position where the esophagus and the stomach connect.

[0091] The three-dimensional coordinates of the esophageal inlet position and the cardia position in the trachea-esophagus model are determined by labeling or automatic segmentation, and the first point set of the esophageal inlet position and the second point set of the cardia position are obtained. In the process of synchronizing the running track to the trachea-esophagus model, the esophageal inlet and the cardia position are used as alignment points to ensure that the abnormal track in the running track can be accurately positioned in the trachea-esophagus model.

[0092] In the track point cloud set, a third point set of the esophageal position and a fourth point set of the cardia position in the running track are determined, and the alignment and matching of the first point set and the third point set and the second point set and the fourth point set are realized by an ICP algorithm to obtain a rotation matrix and a translation vector.

[0093] In the running track, the esophageal inlet position and the cardia position are determined by pre-marking or estimation, and a third point set of the esophageal inlet position region and a fourth point set of the cardia position region are obtained. The points in the first point set and the third point set are one-to-one corresponding to obtain a plurality of first point pairs, and the points in the second point set and the fourth point set are one-to-one corresponding to obtain a plurality of second point pairs. The distance between the two points in each first point pair and the distance between the two points in each second point pair are minimized by the ICP algorithm, the alignment and matching of the first point set and the third point set and the second point set and the fourth point set are realized, and a rotation matrix and a translation vector are obtained.

[0094] The running track is adjusted based on the rotation matrix and the translation vector, and the synchronization of the running track to the pre-obtained trachea-esophagus model is completed.

[0095] The running track is adjusted by the rotation matrix and the translation vector, so that the running track fits the trachea-esophagus model, and the abnormal warning can be performed according to the position of the running track in the trachea-esophagus model. The abnormal warning includes real-time path deviation risk warning and real-time trachea entry risk warning.

[0096] If the center of the first abnormal point set is located in the esophageal stenosis region in the trachea-esophagus model, a real-time risk warning of deviation of the insertion path is performed.

[0097] The first abnormal point set is determined by a topological feature of a connected component, and represents that the points in the first abnormal point set deviate from the trajectory. If the trajectory in the esophageal stenosis region deviates, the gastric tube cannot reach the stomach, and the deviated gastric tube will stay in the stomach tube, which is easy to cause damage to the esophageal wall and cause pain to the patient. Therefore, when the center of the first abnormal point set is located in the esophageal stenosis region in the trachea-esophagus model, a real-time risk warning of deviation of the insertion path is performed to prompt the operator that the insertion of the gastric tube deviates and to make timely adjustment.

[0098] If the geometric center of the second abnormal point set is located in the trachea region in the trachea-gastric tube model, a real-time risk warning of entering the trachea is performed.

[0099] The second abnormal point set is determined by a topological feature of a stable ring, and represents that the points in the second abnormal point set form a closed path. In the process of inserting the gastric tube, when the gastric tube is wound into a ring, it is easy to deviate from the original path, thereby there is a risk of entering the trachea. The mean value of the coordinates of all points constituting the ring in the second abnormal point set is taken as the geometric center coordinate. When the geometric center coordinate is located in the trachea region, it indicates that the gastric tube enters the trachea after being wound into a ring, and then a real-time risk warning of entering the trachea is performed to remind the operator that the gastric tube has entered the trachea and the insertion direction of the gastric tube needs to be adjusted.

[0100] Continuing to refer to the real-time risk warning module, a real-time risk warning is performed based on the positions of the first abnormal point set and the second abnormal point set to remind the operator that there is a fault in the gastric tube path, so as to make adjustment.

[0101] The potential risk warning module is configured to input the real-time position into a preset trajectory prediction model to obtain a predicted trajectory at the next moment, and perform a potential risk warning according to the deviation of the predicted trajectory from the target trajectory.

[0102] The preset trajectory prediction model is obtained by:

[0103] The historical data of the running trajectory of the gastric tube is collected, and the historical data is preprocessed and feature extracted. The features include the change rate of the coordinates of each point in the trajectory, the change of the turning angle, the change of the insertion depth of the gastric tube, and the change of the insertion speed of the gastric tube.

[0104] The obtained features are taken as training data to train a long short-term memory network model, and the trained long short-term memory network model is the preset trajectory prediction model.

[0105] The real-time position of the gastric tube is input to a preset trajectory prediction model to obtain a predicted trajectory at a next time. Deviation is obtained by comparing the predicted trajectory and the target trajectory, and potential risk warning is performed, specifically including:

[0106] A target coordinate is obtained based on the target trajectory, a predicted coordinate is obtained based on the predicted trajectory, and a distance deviation between the target coordinate and the predicted coordinate is obtained. The distance deviation represents the distance between the predicted trajectory and the target trajectory. When the distance deviation is greater than a preset distance threshold, it indicates that if the insertion of the gastric tube continues under the current running trajectory, the gastric tube will be inserted into other parts.

[0107] A target turning angle is determined based on a plurality of target coordinates, a predicted turning angle is determined based on a plurality of predicted coordinates, and an angle deviation between the target turning angle and the predicted turning angle is obtained. The angle deviation represents that the turning of the predicted trajectory and the turning of the target trajectory are significantly inconsistent. The predicted turning angle may be significantly greater than the target turning angle, or the threshold turning angle may be significantly less than the target turning angle. The significant difference in the predicted turning angle will cause the gastric tube insertion path to be seriously deviated and inserted into different positions.

[0108] According to the deviation between the predicted trajectory and the target trajectory, potential risk warning is performed, indicating that the current trajectory of the gastric tube does not constitute a direct threat to the human body, but has a tendency to evolve into a threat to the human body. To avoid the tendency of danger to the human body, potential risk warning is needed. Therefore, when the deviation between the predicted trajectory and the target trajectory satisfies any of the following conditions, potential risk warning is performed:

[0109] When the distance deviation is greater than the preset distance threshold, potential risk warning is performed.

[0110] When the angle deviation is greater than the preset angle threshold, potential risk warning is performed.

[0111] Continuing to refer to the potential risk warning module, the trajectory prediction model is used to predict the direction of the running trajectory, to identify potential risks in the gastric tube insertion process in advance and to perform warning, thereby preventing errors or risks in the gastric tube insertion process.

[0112] It should be noted that the real-time risk warning is an abnormal warning based on topological features, and the potential risk warning is an abnormal warning that the running trajectory has not yet constituted a direct threat, but has a tendency to evolve into an abnormal running trajectory. For the two kinds of risk warning, a multi-factor scoring mechanism is used to quantify the risk state of the current running trajectory, specifically including:

[0113] When the second abnormal point set is located in the trachea region, the risk score is 5 points, and the risk level is high risk.

[0114] When the first abnormal point set is located in the esophageal stenosis region and near the aortic arch impression, the risk score is 4 points, and the risk level is higher risk.

[0115] When the distance deviation is greater than the preset distance threshold, the risk score is 3 points, and the risk level is medium risk.

[0116] When the turning angle deviation is greater than the preset turning angle threshold, the risk score is 4 points, and the risk level is high risk.

[0117] When the running trajectory is too close to the esophageal boundary, the risk score is 1 point, and the risk level is low risk.

[0118] For the above risk score, a warning prompt is given:

[0119] If it is low risk, it is a green prompt, indicating that the running trajectory is stable and there is no abnormal deviation, and the system background continues to monitor.

[0120] If it is medium risk, it is a yellow warning prompt, indicating that a slight trajectory deviation is detected, suggesting slowing down the insertion speed, and displaying the coordinates of the deviation and the adjustment angle.

[0121] If it is high risk, it is an orange alert, showing the deviation distance of the position of the three consecutive points, marking the risk area on the trajectory graph, and prompting the potential penetration risk.

[0122] If it is high risk, it is a red alert, which emits sound and light alarm to prevent the gastric tube from continuing to insert, and displays the second point set in the tracheal area, suggesting a high possibility of entering the trachea, suggesting stopping insertion immediately, and confirming the current gastric tube insertion position.

[0123] The gastric tube trajectory monitoring device further comprises a processing terminal for displaying the running trajectory and the predicted trajectory and displaying the warning. Through the processing terminal, the running trajectory and the predicted trajectory of the gastric tube in the three-dimensional space are displayed, and the wind direction area is highlighted and the predicted trajectory is guided, providing an intuitive trajectory view for the operator, providing visualization during gastric tube insertion and ensuring gastric tube insertion safety.

[0124] In summary, the gastric tube trajectory monitoring system provided by the embodiments of the present application adjusts the deviation between the running trajectory and the target trajectory in real time through the trajectory analysis module to ensure that the gastric tube can be accurately inserted into the stomach. The coherent mechanism is used to process complex geometric and topological information, the position of the abnormal point cloud is accurately identified through topological characteristics, real-time risk warning is performed according to the position of the abnormal point set, and potential risk warning is performed according to the deviation between the running trajectory and the predicted trajectory, avoiding errors during gastric tube insertion, thereby realizing risk prevention and control. And through the processing terminal, the running trajectory and the predicted trajectory of the gastric tube are displayed, providing an visualized insertion path for the operator, thereby ensuring the safety and accuracy of the gastric tube insertion process, and improving the safety and precision of the gastric tube insertion operation, reducing the human operation error in the traditional surgical method.

[0125] Secondly, embodiments of this application provide a gastric tube trajectory detection device. Figure 2 This is a schematic diagram illustrating a gastric tube trajectory monitoring device according to an exemplary embodiment. Figure 2 As shown, the gastric tube trajectory monitoring device includes: gastric tube 5, processing terminal 4, camera device 6, metal ring 1, and metal detector head 2.

[0126] A metal ring 1 is located outside the insertion end of the gastric tube 5 and surrounds the outside of the insertion end. When the gastric tube 5 is inserted, the metal ring 1 can characterize the trajectory of the gastric tube 5. The metal detector head 2 is internally equipped with a receiving coil and a transmitting coil, the transmitting coil being used to generate a variable magnetic field. The metal ring 1 is located in the variable magnetic field; when the metal ring 1 moves, it causes a change in the magnetic flux of the variable magnetic field. The receiving coil receives the induced voltage generated by the movement of the metal ring 1. The metal detector head 2 is electrically connected to the processing terminal, transmitting the induced voltage to the processing terminal 4. The processing terminal 4 receives and processes the induced voltage, obtains the trajectory generated by the movement of the metal ring 1, and displays the trajectory. The processing terminal 4 is internally equipped with a gastric tube trajectory detection system to monitor the trajectory. The processing terminal 4 and the metal detector head 2 are fixed together by a fixing strap 3. A camera device 6 is located on the metal detector head 2 and is used to obtain the depth of the gastric tube 5 entering the stomach. The camera device 6 is pluggable; its use can be determined based on the specific application scenario.

[0127] In summary, the gastric tube trajectory detection device provided in this application determines the running trajectory of the gastric tube 5 through the metal ring 1 and the metal probe 2. The gastric tube trajectory detection system set in the processing terminal 4 can monitor the running trajectory and display the running trajectory on the processing terminal in a visual way to provide the operator with a visual running path, improve the accuracy of gastric tube insertion and reduce the risk of gastric tube insertion operation.

[0128] It should be noted that the gastric tube trajectory detection device provided in this embodiment is used to implement the above-described embodiments, and details already described will not be repeated. As used above, the terms "module," "unit," and "subunit" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the above embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0129] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0130] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific and detailed manner, but should not be construed as limiting the scope of the patent. It should be noted that for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A gastric tube trajectory monitoring system, characterized by, The monitoring system is applied to a gastric tube trajectory monitoring device, the device comprising a gastric tube, an insertion end of the gastric tube being provided with a metal ring, and the monitoring system comprising: a trajectory acquisition module, configured to acquire a real-time position of the metal ring in an insertion process of the gastric tube, and acquire a running trajectory of the insertion end of the gastric tube based on the real-time position; a trajectory analysis module, configured to determine a target trajectory based on a pre-obtained gastric tube insertion path model, acquire a real-time deviation between the target trajectory and the running trajectory, and determine a deviation position and an adjustment angle according to the real-time deviation; a trajectory topology graph construction module, configured to construct a trajectory point cloud set according to the running trajectory, construct a trajectory topology graph group based on a preset radius sequence and the trajectory point cloud set, and determine an abnormal point set according to the trajectory topology graph group based on a persistent homology mechanism, the construction of the trajectory topology graph group based on the preset radius sequence and the trajectory point cloud set comprising: if a distance between any two points in the trajectory point cloud set is less than or equal to the preset radius at any one of the preset radii, an edge of the two points is obtained; if a distance between any point pair in any point set in the trajectory point cloud set is less than or equal to the preset radius, a simplex is obtained, and a number of points in the point set is greater than or equal to 3; the trajectory topology graph of the current preset radius is constructed by taking all trajectory points as vertices of the trajectory topology graph and determining edges and simplices according to distances between the trajectory points; and the trajectory topology graph of each preset radius is obtained according to the preset radius sequence, so as to obtain the trajectory topology graph group; in a persistent bar chart obtained based on the trajectory topology graph group, an end point of an unstable connected component is determined according to a life cycle, and a preset radius corresponding to the end point is taken as a breaking point; a plurality of clustering clusters are obtained by density clustering on all breaking points; a target breaking point, i.e., a target radius, is determined in each clustering cluster, a trajectory topology graph is reconstructed based on the target radius and the trajectory point cloud set, vertex coordinates of the unstable connected component are determined in the trajectory topology graph, and a first abnormal point set in the trajectory point cloud set is determined according to the vertex coordinates; a real-time risk early warning module, configured to synchronize the running trajectory to a pre-obtained trachea-esophagus model, and perform real-time risk early warning according to positions of the abnormal point set in the trachea-esophagus model; a potential risk early warning module, configured to input the real-time position into a preset trajectory prediction model, obtain a predicted trajectory at a next moment, and perform potential risk early warning according to a deviation between the predicted trajectory and the target trajectory.

2. The gastric tube trajectory monitoring system of claim 1, wherein, the determination of the abnormal point set based on the trajectory topology graph group and the persistent homology mechanism comprising: topological features of each trajectory topology graph are determined based on the trajectory topology graph group; a coordinate system is constructed by taking the preset radius sequence as an abscissa and taking the topological features as an ordinate, a life cycle of the topological features is determined according to a preset radius at which the topological features first appear as a starting point and a preset radius at which the topological features disappear as an ending point, and a persistent bar chart is obtained by displaying the life cycle of each topological feature through the coordinate system. The abnormal topological features are determined by analyzing the life cycle distribution in the persistent bar graph, and the abnormal point set is determined by mapping the abnormal topological features to the trajectory point cloud set.

3. The gastric tube trajectory monitoring system of claim 2, wherein, The topological features include connected components, and the determination of the abnormal topological features by analyzing the life cycle distribution in the persistent bar graph and the mapping of the abnormal topological features to the trajectory point cloud set to determine the abnormal point set include: If the life cycle of any connected component is less than a preset life cycle threshold, the connected component is an unstable connected component; In the persistent bar graph, the breaking points of all the unstable connected components are determined, and the breaking points are clustered to obtain a plurality of clustering clusters; In each of the clustering clusters, a target breaking point is determined, a first topological graph is reconstructed based on the target breaking point and the trajectory point cloud set, the vertex coordinates of the unstable connected component are determined in the first topological graph, and a first abnormal point set in the trajectory point cloud set is determined according to the vertex coordinates.

4. The gastric tube trajectory monitoring system of claim 3, wherein, The topological features include rings, and the determination of the abnormal topological features by analyzing the life cycle distribution in the persistent bar graph and the mapping of the abnormal topological features to the trajectory point cloud set to determine the abnormal point set include: If the life cycle of any ring is greater than or equal to a preset life cycle threshold, the ring is a stable ring; In the persistent bar graph, all the stable rings are determined, the generators of the stable rings are extracted, the coordinates constituting the stable rings are determined according to the generators, and a second abnormal point set in the trajectory point cloud set is determined according to the coordinates of the rings.

5. The gastric tube trajectory monitoring system of claim 4, wherein, The real-time risk warning according to the positions of the abnormal point set in the trachea-esophagus model includes: If the center of the first abnormal point set is located in an esophageal stenosis region in the trachea-esophagus model, a real-time insertion path deviation risk warning is performed; If the geometric center of the second abnormal point set is located in a tracheal region in the trachea-gastric tube model, a real-time tracheal entry risk warning is performed.

6. The gastric tube trajectory monitoring system of claim 1, wherein, The synchronization of the running trajectory to the pre-obtained trachea-esophagus model includes: The esophageal inlet position and the cardia position in the trachea-esophagus model are determined, and a first point set of the esophageal inlet position and a second point set of the cardia position are obtained, wherein the esophageal inlet position is the position where the esophagus and the pharynx connect, and the cardia position is the position where the esophagus and the stomach connect; In the trajectory point cloud set, a third point set of the esophageal position and a fourth point set of the cardia position in the running trajectory are determined, and the alignment and matching of the first point set and the third point set and the second point set and the fourth point set are realized through an ICP algorithm to obtain a rotation matrix and a translation vector; The running trajectory is adjusted based on the rotation matrix and the translation vector to complete the synchronization of the running trajectory to the pre-obtained trachea-esophagus model.

7. The gastric tube trajectory monitoring system of claim 1, wherein, The potential risk warning according to the deviation of the predicted trajectory from the target trajectory includes: Target coordinates are obtained based on the target trajectory, predicted coordinates are obtained based on the predicted trajectory, and a distance deviation between the target coordinates and the predicted coordinates is obtained. determining a target turning angle based on the target coordinates, determining a predicted turning angle based on the predicted coordinates, and obtaining an angle deviation between the target turning angle and the predicted turning angle; when the distance deviation is greater than a preset distance threshold, a potential risk warning is performed; and / or when the angle deviation is greater than a preset angle threshold, a potential risk warning is performed.

8. The gastric tube trajectory monitoring system of claim 1, wherein, The gastric tube trajectory monitoring device further comprises a processing terminal for displaying the running trajectory and the predicted trajectory.

9. A gastric tube trajectory monitoring device, characterized by The device comprises a processing terminal, a camera device, a metal ring, and a metal detection head. The metal ring is located outside the gastric tube insertion end and surrounds the gastric tube insertion end. The metal detection head comprises a receiving coil and a transmitting coil. The transmitting coil is used to generate a variable magnetic field. The receiving coil is used to receive the changing magnetic flux of the metal ring when the metal ring moves in the variable magnetic field and obtain an induced voltage. The metal detection head is electrically connected to the processing terminal and transmits the induced voltage of the receiving coil to the processing terminal. The processing terminal is used to receive the induced voltage to obtain the trajectory of the metal ring movement and display the trajectory of the metal ring movement. The camera device is electrically connected to the processing terminal and is used to take pictures of the scale of the gastric tube.

Images