Endoscope insertion tube auxiliary insertion system and method, and endoscope system
阅读说明:本技术 内窥镜插管辅助***系统及方法、内窥镜系统 (Endoscope insertion tube auxiliary insertion system and method, and endoscope system ) 是由 汪丰 王琛 叶欢 徐俊 张弛 徐成 于 2019-11-01 设计创作,主要内容包括:本发明提供一种内窥镜插管辅助插入系统及方法、内窥镜系统。所述辅助插入系统包括影像识别处理单元和角度传感器,所述影像识别处理单元用于与内窥镜的影像输出端相连,所述角度传感器与所述影像识别处理单元相连;其中,所述影像识别处理单元,用于接收所述插管插入过程中的输出影像,并根据所述输出影像识别所述插管是否遇到预设的定位基准部位;响应于所述插管遇到所述定位基准部位,所述角度传感器实时获取所述插管相对所述声门裂部位的偏移角度,并将所述偏移角度反馈至所述影像识别处理单元;所述影像识别处理单元,还用于根据所述偏移角度调整所述插管的插入方向。可以大大降低插管插入目标部位的难度,提高插管插入的准确率。(The invention provides an endoscope intubation tube auxiliary insertion system and method and an endoscope system. The auxiliary insertion system comprises an image recognition processing unit and an angle sensor, wherein the image recognition processing unit is used for being connected with an image output end of the endoscope, and the angle sensor is connected with the image recognition processing unit; the image identification processing unit is used for receiving an output image in the insertion process of the insertion tube and identifying whether the insertion tube meets a preset positioning reference part according to the output image; responding to the situation that the intubation tube meets the positioning reference part, the angle sensor acquires the offset angle of the intubation tube relative to the glottic crack part in real time and feeds the offset angle back to the image recognition processing unit; the image recognition processing unit is further used for adjusting the insertion direction of the insertion tube according to the offset angle. The difficulty of inserting the cannula into the target part can be greatly reduced, and the accuracy of inserting the cannula is improved.)
1. An auxiliary insertion system for an endoscope intubation tube is characterized by comprising an image recognition processing unit and an angle sensor, wherein the image recognition processing unit is used for being connected with an image output end of an endoscope, and the angle sensor is connected with the image recognition processing unit; wherein the content of the first and second substances,
the image identification processing unit is used for receiving an output image in the insertion process of the insertion tube and identifying whether the insertion tube meets a preset positioning reference part or not according to the output image;
in response to the situation that the intubation tube meets the positioning reference part, the angle sensor acquires the offset angle of the intubation tube relative to the positioning reference part in real time and feeds the offset angle back to the image recognition processing unit;
the image recognition processing unit is further used for determining the insertion direction of the insertion tube according to the offset angle.
2. The system of claim 1, wherein the location reference site is a glottic site, and the image recognition processing unit comprises a storage subunit and a processing subunit connected to the storage subunit, wherein,
the storage subunit stores an image database of the glottic fissure part in advance;
the processing subunit is configured to compare the output image with the image database to determine whether the intubation tube encounters the glottic fracture site.
3. The system of claim 1, wherein the location reference site is a glottic site, and the image recognition processing unit comprises a training subunit and a processing subunit connected to the training subunit; wherein the content of the first and second substances,
the training subunit is used for performing machine learning on the image data of a large number of glottic fissure parts according to a preset image recognition algorithm to obtain an image database;
the processing subunit is configured to compare the output image with the image database to determine whether the cannula meets the glottic fracture site.
4. The endoscopic cannula assisted insertion system according to any of claims 1-3, wherein the angle sensor acquires the offset angle of the cannula in real time after calibration zeroing in response to the cannula encountering the positioning reference site.
5. The system of any of claims 1-3, further comprising a display unit, wherein the display unit is connected to the image recognition processing unit; wherein the content of the first and second substances,
the image recognition processing unit is also used for distinguishing a direction identifier according to the shape of the positioning reference part;
and the display unit is used for displaying the direction identifier.
6. The endoscope cannula assisted insertion system of claim 5, wherein the display unit employs an LCD display or an OLED display.
7. The endoscopic cannula assisted insertion system of any of claims 1-3, wherein the angle sensor is fixedly disposed on the endoscope handle.
8. The endoscope cannula assisted insertion system of claim 7, wherein the angle sensor comprises a housing, an angle sensor board and a USB transmission interface;
the case comprises an insertion part and an installation part which is connected with the insertion part and is provided with an accommodating space, the insertion part is inserted on the endoscope handle, and the accommodating space of the installation part accommodates the angle sensor board card and the USB transmission interface;
the USB transmission interface is located on one side, away from the inserting part, of the angle sensor board card, the first end of the USB transmission interface is connected with the angle sensor board card, and the second end of the USB transmission interface is connected with the image recognition processing unit.
9. An endoscope cannula assisted insertion method, comprising:
receiving an output image in the insertion process of the insertion tube, and identifying whether the insertion tube meets a preset positioning reference part according to the output image;
responding to the situation that the intubation tube meets the positioning reference part, acquiring a deviation angle of the intubation tube relative to the positioning reference part in real time, and feeding the deviation angle back to the image recognition processing unit;
determining an insertion direction of the cannula according to the offset angle.
10. An endoscopic system comprising an endoscope, a cannula and a handle, wherein the endoscopic system further comprises an auxiliary insertion system according to any of claims 1-8.
Technical Field
The invention relates to the technical field of medical equipment, in particular to an auxiliary insertion system and an auxiliary insertion method for an endoscope insertion tube and an endoscope system.
Background
Disclosure of Invention
The present invention is directed to at least one of the problems of the prior art, and provides an insertion assisting system for an endoscope insertion tube, an insertion assisting method for an endoscope insertion tube, and an endoscope system.
In one aspect of the invention, an auxiliary insertion system for an insertion tube of an endoscope is provided, which comprises an image recognition processing unit and an angle sensor, wherein the image recognition processing unit is used for being connected with an image output end of the endoscope, and the angle sensor is connected with the image recognition processing unit; wherein the content of the first and second substances,
the image identification processing unit is used for receiving an output image in the insertion process of the insertion tube and identifying whether the insertion tube meets a preset positioning reference part or not according to the output image;
in response to the situation that the intubation tube meets the positioning reference part, the angle sensor acquires the offset angle of the intubation tube relative to the positioning reference part in real time and feeds the offset angle back to the image recognition processing unit;
the image recognition processing unit is further used for determining the insertion direction of the insertion tube according to the offset angle.
Optionally, the positioning reference portion is a glottic fracture portion, and the image recognition processing unit includes a storage subunit and a processing subunit connected to the storage subunit, wherein,
the storage subunit stores an image database of the glottic fissure part in advance;
the processing subunit is configured to compare the output image with the image database to determine whether the intubation tube encounters the glottic fracture site.
Optionally, the positioning reference part is a glottic part, and the image recognition processing unit includes a training subunit and a processing subunit connected to the training subunit; wherein the content of the first and second substances,
the training subunit is used for performing machine learning on the image data of a large number of glottic fissure parts according to a preset image recognition algorithm to obtain an image database;
the processing subunit is configured to compare the output image with the image database to determine whether the cannula meets the glottic fracture site.
Optionally, the angle sensor acquires the offset angle of the cannula in real time after calibration zeroing in response to the cannula encountering the positioning reference site.
Optionally, the auxiliary insertion system further comprises a display unit, and the display unit is connected with the image recognition processing unit; wherein the content of the first and second substances,
the image recognition processing unit is also used for distinguishing a direction identifier according to the shape of the positioning reference part;
and the display unit is used for displaying the direction identifier.
Optionally, the display unit employs an LCD display or an OLED display.
Optionally, the angle sensor is fixedly arranged on the endoscope handle.
Optionally, the angle sensor comprises a case, an angle sensor board card and a USB transmission interface;
the case comprises an insertion part and an installation part which is connected with the insertion part and is provided with an accommodating space, the insertion part is inserted on the endoscope handle, and the accommodating space of the installation part accommodates the angle sensor board card and the USB transmission interface;
the USB transmission interface is located on one side, away from the inserting part, of the angle sensor board card, the first end of the USB transmission interface is connected with the angle sensor board card, and the second end of the USB transmission interface is connected with the image recognition processing unit.
In another aspect of the present invention, there is provided an insertion assisting method for an insertion tube of an endoscope, including:
receiving an output image in the insertion process of the insertion tube, and identifying whether the insertion tube meets a preset positioning reference part according to the output image;
responding to the situation that the intubation tube meets the positioning reference part, acquiring a deviation angle of the intubation tube relative to the positioning reference part in real time, and feeding the deviation angle back to the image recognition processing unit;
determining an insertion direction of the cannula according to the offset angle.
In another aspect of the present invention, there is provided an endoscope system comprising an endoscope, a cannula and a handle, the endoscope system further comprising the auxiliary insertion system as described above.
The auxiliary insertion system, the auxiliary insertion method and the endoscope system can realize that a positioning reference part (such as a glottic part and the like) of a patient can be identified in the intubation process by virtue of the image identification processing unit, the left and right directions can be identified according to the shape presented by the part, meanwhile, the offset angle in the intubation insertion process can be fed back by virtue of the angle sensor and fed back to the image identification processing unit, and the image identification processing unit adjusts the insertion direction of the intubation according to the offset angle, so that the intubation can be inserted into an expected part (such as a left lung or a right lung and the like). The difficulty of inserting the cannula into the target part can be greatly reduced, and the accuracy of inserting the cannula is improved.
Drawings
FIG. 1 is a schematic structural diagram of an auxiliary insertion system according to a first embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an image recognition processing unit according to a second embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an image recognition processing unit according to a third embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an angle sensor according to a fourth embodiment of the present invention;
FIG. 5 is an exploded view of the angle sensor shown in FIG. 4;
FIG. 6 is a schematic view of an angle sensor assembled with an endoscope handle according to a fifth embodiment of the present invention;
FIG. 7 is an exploded view of the angle sensor of FIG. 6 assembled with the endoscope handle;
fig. 8 is a flowchart of an auxiliary insertion method according to a sixth embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, an
Specifically, as shown in fig. 1, when the
The image
It will be appreciated that during the insertion of the cannula, the medical staff may adjust the insertion direction of the cannula according to the offset angle, for example, the medical staff originally intends to insert the cannula into the left lung of the patient, but the current offset angle feeds back the deviation to the right lung of the patient, at this time, the medical staff may manually adjust the insertion direction of the cannula, so that the cannula may be finally inserted into the left lung of the patient. In addition to adjusting the insertion direction of the cannula according to the offset angle, the medical staff can also adjust itself according to the program setting, for example, inputting the target insertion site of the cannula (for example, the left lung of the patient), comparing the real-time offset angle of the cannula insertion process with the offset angle required by the target insertion site, and adjusting in real time to ensure that the cannula can be finally inserted into the target insertion site.
The auxiliary insertion system of the embodiment can realize that the positioning reference part of the patient is identified in the intubation process by virtue of the image identification processing unit, the left and right directions are identified according to the shape presented by the part, meanwhile, the offset angle in the intubation insertion process can be fed back by virtue of the angle sensor, the offset angle is fed back to the image identification processing unit, and the image identification processing unit determines the insertion direction of the intubation according to the offset angle. Obviously, the auxiliary insertion system of the embodiment can greatly reduce the difficulty of inserting the cannula into the target part and improve the accuracy of inserting the cannula.
It should be noted that, besides the location reference part may be a glottic fissure part, a person skilled in the art may also select some other location reference parts according to actual needs, for example, the location reference part may be manually added by an operating physician, and the like, which is not limited in this embodiment.
In some optional embodiments, as shown in fig. 2, the image
In the auxiliary insertion system of the embodiment, the processing subunit compares the output image of the endoscope with the pre-stored image database, so that the accurate time of inserting the intubation tube into the glottis part can be accurately identified, and the identification accuracy is improved.
In some optional embodiments, as shown in fig. 3, the image
In the auxiliary insertion system of this embodiment, the training subunit performs machine learning according to a preset image recognition algorithm, for example, images of glottic parts of different patients may be collected in advance, and machine learning is performed on the images of the glottic parts, so that an image database may be obtained. Therefore, when the processing subunit compares the output image of the endoscope with the image database, the accuracy of identifying the glottic fissure part can be effectively improved.
In some optional embodiments, the image recognition processing unit may be implemented by raspberry pi 4B as a processor, and the reserved interface of the chassis thereof has USB or micro HDMI. Most of the output of the front-end endoscope system is DVI format video, the video format is converted into MIPI CSI after being converted, the MIPI CSI can be directly connected with a raspberry group, an autonomously developed image recognition algorithm runs on a hardware platform, machine learning training is carried out on a large amount of sample data through the algorithm, and the glottic crack shape can be quickly and accurately recognized.
In some alternative embodiments, the angle sensor acquires the offset angle of the cannula in real time after calibration zeroing in response to the cannula encountering a glottic site. That is to say, at the moment of confirming that the intubate inserts glottic fissure position, calibrate zero-return to angle sensor, medical personnel can initiatively calibrate zero-return to angle sensor, perhaps, also can adopt program control, calibrate zero-return to angle sensor automatically. The angle acquired by the angle sensor after the zero resetting is the offset angle of the intubation tube, so that the insertion direction of the intubation tube can be conveniently determined.
In some optional embodiments, as shown in fig. 1, the
It should be noted that, the specific structure of the
In some optional embodiments, the
In some optional embodiments, as shown in fig. 4 to 7, the
In another aspect of the present invention, as shown in fig. 8, an auxiliary insertion method S100 for an endoscope insertion tube is provided, in which the auxiliary insertion system described above can be used, and the detailed structure of the auxiliary insertion system can refer to the related description above, which is not repeated herein. The auxiliary insertion method S100 includes:
s110, receiving an output image in the insertion process of the cannula, and identifying whether the cannula meets a preset positioning reference part according to the output image.
And S120, responding to the situation that the intubation tube meets the positioning reference part, acquiring the offset angle of the intubation tube relative to the positioning reference part in real time, and feeding the offset angle back to the image recognition processing unit.
And S130, determining the insertion direction of the cannula according to the offset angle.
The auxiliary insertion method of the embodiment can realize that the positioning reference part of the patient is identified in the intubation process by identifying the output image, the left and right directions are distinguished according to the shape presented by the part, and the insertion direction of the intubation is determined according to the deviation angle of the intubation relative to the glottic fracture part. Obviously, the auxiliary insertion method of the embodiment can greatly reduce the difficulty of inserting the cannula into the target part and improve the accuracy of inserting the cannula.
In another aspect of the invention, an endoscope system is provided, comprising an endoscope, a cannula and a handle, the endoscope system further comprising the auxiliary insertion system as described above.
The endoscope system of the present embodiment includes the above-mentioned auxiliary insertion system, and can recognize the positioning reference portion of the patient during the intubation procedure by the image recognition processing unit, recognize the left and right directions according to the shape of the portion, and feed back the offset angle during the intubation insertion procedure by the angle sensor, and feed back the offset angle to the image recognition processing unit, and the image recognition processing unit determines the insertion direction of the intubation according to the offset angle. Obviously, the auxiliary insertion system of the embodiment can greatly reduce the difficulty of inserting the cannula into the target part and improve the accuracy of inserting the cannula.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
- 上一篇:一种医用注射器针头装配设备
- 下一篇:经肛门向回盲瓣的深度植入器械及其植入方法