Method and device for judging left and right lumens of bronchus of endoscope and endoscope system
阅读说明：本技术 内窥镜支气管左右管腔的判断方法及装置、内窥镜系统 (Method and device for judging left and right lumens of bronchus of endoscope 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支气管左右管腔的判断方法及装置、内窥镜系统,方法包括：接收内窥镜插管插入支气管过程中的成像视频；根据所述成像视频识别出躯体左右侧标识；获取所述插管的转动角度,并根据所述转动角度和所述左右侧标识,判断所述插管所插入的支气管腔。从接收到的成像视频中识别出躯体左右侧标识,并结合插管的转动角度和该左右侧标识来判断插管的插入位置,从而可以辅助操作医师准确判别支气管腔,提高插管准确度,减少重复插管对患者支气管的损伤。(The invention provides a method and a device for judging left and right lumens of a bronchus of an endoscope and an endoscope system, wherein the method comprises the following steps: receiving an imaging video of an endoscope insertion tube in a bronchial process; identifying left and right side marks of a body according to the imaging video; and acquiring the rotation angle of the cannula, and judging the bronchial cavity into which the cannula is inserted according to the rotation angle and the left and right side marks. The identification of the left side and the right side of the body is identified from the received imaging video, and the insertion position of the intubation tube is judged by combining the rotation angle of the intubation tube and the identification of the left side and the right side, so that a doctor can accurately judge the bronchial cavity, the intubation accuracy is improved, and the damage of repeated intubation to the bronchus of a patient is reduced.)
1. A method for judging the left and right lumens of a bronchus of an endoscope is characterized by comprising the following steps:
receiving an imaging video of an endoscope insertion tube in a bronchial process;
identifying left and right side marks of a body according to the imaging video;
and acquiring the rotation angle of the cannula, and judging the bronchial cavity into which the cannula is inserted according to the rotation angle and the left and right side marks.
2. The method of claim 1, wherein the identifying the left and right side of the body from the imaging video comprises:
identifying a glottic fissure part according to the imaging video;
and identifying the left and right side marks according to the shape of the glottic fissure part.
3. The method of claim 2, wherein identifying a glottic crack site from the imaging video comprises:
extracting a plurality of frame pictures from the imaging video based on a preset frame number;
preprocessing the plurality of frames of pictures;
and continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until the glottic fracture part is identified.
4. The method of claim 3, wherein the pre-processing comprises noise reduction processing; and/or the presence of a gas in the gas,
the image recognition algorithm adopts an image recognition algorithm based on OpenCV.
5. The method according to any one of claims 1 to 4, wherein the determining the bronchial lumen into which the cannula is inserted according to the rotation angle and the left and right side identifiers comprises:
displaying an imaging video of the endoscope on a display screen, and marking the left and right side marks in the imaging video displayed on the display screen in real time;
calculating the rotation angle of the imaging video displayed by the display screen in real time according to the rotation angle so as to realize the synchronous rotation of the left and right side marks;
and judging the bronchial cavity into which the cannula is inserted according to the bronchial cavity in the imaging video displayed by the display screen and the left and right side marks tracked in real time.
6. A device for judging the left and right lumens of a bronchus of an endoscope, comprising:
the receiving unit is used for receiving an imaging video in the process of inserting the endoscope insertion tube into the bronchus;
the identification unit is used for identifying left and right side marks of the body according to the imaging video;
the angle acquisition unit is used for acquiring the rotation angle of the insertion tube;
and the processing unit is used for judging the bronchial cavity into which the cannula is inserted according to the rotating angle and the left and right side marks.
7. The apparatus according to claim 6, wherein the identification unit is further configured to:
identifying a glottic fissure part according to the imaging video;
and identifying the left and right side marks according to the shape of the glottic fissure part.
8. The apparatus according to claim 7, wherein the identification unit is further configured to:
extracting a plurality of frame pictures from the imaging video based on a preset frame number;
preprocessing the plurality of frames of pictures;
and continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until the glottic fracture part is identified.
9. The apparatus according to any one of claims 6-8, further comprising a display unit, wherein the processing unit comprises a marking subunit, a calculating subunit, and a judging subunit; wherein the content of the first and second substances,
the display unit is used for displaying the imaging video of the endoscope in real time;
the marking subunit is used for marking the left and right side marks in real time in the imaging video displayed by the display unit;
the calculating subunit is configured to calculate, in real time, a rotation angle of the imaging video displayed by the display unit according to the rotation angle, so as to implement synchronous rotation of the left and right side identifiers;
and the judging subunit is used for judging the bronchial cavity into which the cannula is inserted according to the bronchial cavity in the imaging video displayed by the display unit and the real-time tracked left and right side marks.
10. An endoscopic system comprising an endoscope, a cannula and a handle, characterized in that the endoscopic system further comprises a device according to any of claims 6-9.
Technical Field
The invention relates to the technical field of medical instruments, in particular to a method for judging left and right bronchus cavities of an endoscope, a device for judging left and right bronchus cavities of the endoscope and an endoscope system.
Background
With advances in computer vision technology, medical endoscopes have evolved rapidly over the last several decades. However, the endoscope still has a space for improvement in the aspect of trachea intubation, and particularly, clinically, because the endoscope rotates at an unknown angle in an intubation process, the problem that a left bronchial lumen and a right bronchial lumen cannot be distinguished quickly exists, an operating physician often needs to perform repeated intubation to judge the left bronchial lumen and the right bronchial lumen, so that the operation time is increased, and the trachea of a patient can be damaged to a certain extent.
Disclosure of Invention
The present invention is directed to solve at least one of the problems of the prior art, and provides a method for determining a left and right lumen of a bronchus of an endoscope, a device for determining a left and right lumen of a bronchus of an endoscope, and an endoscope system.
In one aspect of the present invention, a method for determining a left and right bronchus lumen of an endoscope is provided, including:
receiving an imaging video of an endoscope insertion tube in a bronchial process;
identifying left and right side marks of a body according to the imaging video;
and acquiring the rotation angle of the cannula, and judging the bronchial cavity into which the cannula is inserted according to the rotation angle and the left and right side marks.
Optionally, the recognizing, according to the imaging video, the left and right side identifiers of the body includes:
identifying a glottic fissure part according to the imaging video;
and identifying the left and right side marks according to the shape of the glottic fissure part.
Optionally, the identifying a glottic crack portion according to the imaging video includes:
extracting a plurality of frame pictures from the imaging video based on a preset frame number;
preprocessing the plurality of frames of pictures;
and continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until the glottic fracture part is identified.
Optionally, the pre-processing comprises noise reduction processing; and/or the presence of a gas in the gas,
the image recognition algorithm adopts an image recognition algorithm based on OpenCV.
Optionally, the determining, according to the rotation angle and the left and right side identifiers, the bronchial lumen into which the cannula is inserted includes:
displaying an imaging video of the endoscope on a display screen, and marking the left and right side marks in the imaging video displayed on the display screen in real time;
calculating the rotation angle of the imaging video displayed by the display screen in real time according to the rotation angle so as to realize the synchronous rotation of the left and right side marks;
and judging the bronchial cavity into which the cannula is inserted according to the bronchial cavity in the imaging video displayed by the display screen and the left and right side marks tracked in real time.
In another aspect of the present invention, there is provided an apparatus for determining a left and right bronchus lumen of an endoscope, comprising:
the receiving unit is used for receiving an imaging video in the process of inserting the endoscope insertion tube into the bronchus;
the identification unit is used for identifying left and right side marks of the body according to the imaging video;
the angle acquisition unit is used for acquiring the rotation angle of the insertion tube;
and the processing unit is used for judging the bronchial cavity into which the cannula is inserted according to the rotating angle and the left and right side marks.
Optionally, the identification unit is specifically further configured to:
identifying a glottic fissure part according to the imaging video;
and identifying the left and right side marks according to the shape of the glottic fissure part.
Optionally, the identification unit is specifically further configured to:
extracting a plurality of frame pictures from the imaging video based on a preset frame number;
preprocessing the plurality of frames of pictures;
and continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until the glottic fracture part is identified.
Optionally, the system further comprises a display unit, and the processing unit comprises a marking subunit, a calculating subunit and a judging subunit; wherein the content of the first and second substances,
the display unit is used for displaying the imaging video of the endoscope in real time;
the marking subunit is used for marking the left and right side marks in real time in the imaging video displayed by the display unit;
the calculating subunit is configured to calculate, in real time, a rotation angle of the imaging video displayed by the display unit according to the rotation angle, so as to implement synchronous rotation of the left and right side identifiers;
and the judging subunit is used for judging the bronchial cavity into which the cannula is inserted according to the bronchial cavity in the imaging video displayed by the display unit and the real-time tracked left and right side marks.
In another aspect of the invention, there is provided an endoscope system comprising an endoscope, a cannula and a handle, the endoscope system further comprising the device as described above.
According to the judging method and the judging device for the left and right bronchus cavities of the endoscope and the endoscope system, the left and right side marks of the body are identified from the received imaging video, and the inserting position of the inserting tube is judged by combining the rotating angle of the inserting tube and the left and right side marks, so that an operation doctor can be assisted to accurately judge the bronchus cavities, the accuracy of inserting the tube is improved, and the damage of repeated inserting tubes to the bronchus of a patient is reduced.
Drawings
FIG. 1 is a flowchart of a method for determining the lumen of a left or right bronchus of an endoscope according to a first embodiment of the present invention;
FIG. 2 is a schematic structural view of a device for judging the left and right lumens of a bronchus of an endoscope in a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a processing unit according to a third 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, one aspect of the present invention relates to a method S100 for determining a left and right bronchus lumen of an endoscope, including:
and S110, receiving an imaging video in the process of inserting the endoscope insertion tube into the bronchus.
Specifically, in this step, a video acquisition device may be used to acquire an imaging video output by the lens of the endoscope during insertion of the insertion tube of the endoscope into the bronchus, for example, an FPGA image processing board may be used, and the FPGA image processing board is connected to the endoscope, so as to receive the imaging video output by the endoscope. Of course, besides, those skilled in the art may select other video capturing devices according to actual needs, and this embodiment does not specifically limit this.
And S120, identifying the left and right side marks of the body according to the imaging video.
Specifically, in the step, medical personnel can automatically recognize the left and right side marks of the body according to the imaging video; alternatively, in this step, the imaging video may also be recognized by image recognition, so as to recognize a portion that can represent the left and right side marks, for example, a glottic crack portion. Of course, besides these two ways, those skilled in the art may also use other identification ways to determine the left and right side marks, which is not specifically limited in this embodiment.
S130, obtaining the rotation angle of the cannula, and judging the bronchial cavity into which the cannula is inserted according to the rotation angle and the left and right side marks.
Specifically, in this step, the rotation angle of the cannula may be obtained by providing an angle sensor (such as a gyroscope) on the cannula, and then the rotation angle may be input to a main control board connected to the FPGA image processing board, and the main control board may determine whether the position where the cannula is inserted is the left bronchial lumen or the right bronchial lumen according to the rotation angle and the left and right side identifiers.
According to the judgment method, the left and right side marks of the body are recognized from the received imaging video, and the insertion position of the intubation tube is judged by combining the rotation angle of the intubation tube and the left and right side marks, so that an operation doctor can be assisted to accurately judge the bronchial cavity, the intubation accuracy is improved, and the damage of repeated intubation to the bronchus of a patient is reduced.
In some optional embodiments, identifying the left and right side signs of the body from the imaging video includes:
identifying a glottic fissure part according to the imaging video;
and identifying left and right side marks according to the shape of the glottic fissure part.
In this embodiment, the glottic region is the narrowest part of the laryngeal cavity, between the bilateral vocal folds, the medial border of the arytenoid fundus, and the vocal cord prominence, and is triangular, narrow in the anterior aspect and wide in the posterior aspect. Therefore, by means of the unique shape of the glottic fissure part, the marks on the left side and the right side of the body can be effectively identified, and the intubation accuracy can be further improved.
In some optional embodiments, identifying a glottic crack site from the imaged video comprises:
a plurality of frame pictures are extracted from the imaging video based on a preset frame number.
Specifically, in this step, the number of frames may be 25 or other frames, and a 25-frame picture is extracted from the imaging video based on the set number of frames.
Preprocessing is carried out on 25 frames of pictures, and noise reduction processing is mainly carried out on each picture.
And continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until a glottic crack part is identified, wherein the image identification algorithm based on OpenCV is preferably adopted. Of course, besides, those skilled in the art may select other image recognition algorithms according to actual needs, and this embodiment does not specifically limit this.
In some optional embodiments, determining the bronchial lumen into which the cannula is inserted based on the rotation angle and the left and right side identifiers comprises:
displaying an imaging video of the endoscope on a display screen, and marking left and right side marks in the imaging video displayed on the display screen in real time;
according to the rotation angle, the rotation angle of the imaging video displayed by the display screen is calculated in real time so as to realize synchronous rotation of the left and right side marks;
and judging the bronchial cavity into which the cannula is inserted according to the bronchial cavity in the imaging video displayed by the display screen and the left and right side marks tracked in real time.
Specifically, in the embodiment, in the embedded main control board, the intubation tube rotation angle data transmitted by the angle sensors such as the gyroscope is acquired, the intubation tube rotation angle data is converted into the video rotation angle in the display screen through calculation, and the synchronous rotation of the left and right side mark information is realized, so that the left and right sides of the patient are tracked in real time in the endoscope intubation process until the patient reaches the carina, and at the moment, two bronchial cavities of the carina and the left and right side marks tracked in real time appear in the display screen picture, so that an operator is assisted to distinguish the left and right bronchi, and errors are effectively prevented.
In another aspect of the present invention, as shown in fig. 2, there is provided a
a receiving unit 110, configured to receive an imaging video during insertion of an endoscope insertion tube into a bronchus;
the recognition unit 120 is configured to recognize left and right side identifiers of a body according to the imaging video;
an angle acquisition unit 130 for acquiring a rotation angle of the cannula;
and the processing unit 140 is used for judging the bronchial cavity into which the cannula is inserted according to the rotation angle and the left and right side marks.
The judgment device of this embodiment discerns the left and right side sign of body from the imaging video that receives to the inserted position of intubate is judged to the turned angle who combines the intubate with should controlling the side sign, thereby can the accurate bronchial chamber of differentiating of auxiliary operation doctor, improves the intubate degree of accuracy, reduces the damage of repeated intubate to patient's bronchus.
In some optional embodiments, the identifying unit 120 is further specifically configured to:
identifying a glottic fissure part according to the imaging video;
and identifying left and right side marks according to the shape of the glottic fissure part.
In some optional embodiments, the identifying unit 120 is further specifically configured to:
extracting a plurality of frame pictures from an imaging video based on a preset frame number;
preprocessing a plurality of frame pictures;
and continuously carrying out region identification on each preprocessed picture by adopting a preset image identification algorithm until a glottic crack part is identified.
In some optional embodiments, the determining
It should be noted that, the specific structure of the display unit 150 is not limited, for example, the display unit 150 may adopt an LCD display, or the display unit 150 may also adopt an OLED display. Of course, besides the above, those skilled in the art may select other display devices according to actual needs, and the present invention is not limited to this.
In another aspect of the present invention, an endoscope system is provided, which includes an endoscope, a cannula and a handle, and the endoscope system further includes the devices described above, which may be referred to the related descriptions above, and are not described herein again.
The endoscope system of the embodiment has the judging device described above, and can identify the left and right side marks of the body from the received imaging video, and judge the insertion position of the intubation tube by combining the rotation angle of the intubation tube and the left and right side marks, so that an operating physician can be assisted to accurately judge the bronchial lumen, the intubation accuracy is improved, and the damage of repeated intubation to the bronchus of the patient is reduced.
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：内窥镜插管辅助插入系统及方法、内窥镜系统