An image acquisition system

Abstract

The present invention discloses an image acquisition system, comprising a rotary retraction device, an imaging device, and an image acquisition device. Upon receiving a start command, the rotary retraction device drives the imaging device to perform a rotary retraction motion within a blood vessel, and simultaneously triggers a positioning mechanism to send an image acquisition command to the image acquisition device, causing the image acquisition device to synchronously capture images from within the imaging device. Nurd regions are determined from the images and sent to the rotary retraction device. The rotary retraction device adjusts the motor control signal at the next moment based on the reduction of the nurd region and the current moment motor control signal. The present invention achieves mapping between the image and the motor position by installing a positioning mechanism on the rotating motor and performing image acquisition when the positioning mechanism is triggered. The nurd region is identified from the captured image to adjust the motor control signal at the next moment, effectively reducing the nurd phenomenon.

Description

Technical Field

[0001] The present invention relates to the field of image acquisition technology, and more particularly to an image acquisition system. Background Art

[0002] Intravascular ultrasound (IVUS) is currently widely used for imaging cardiovascular diseases such as atherosclerosis and coronary stents. IVUS uses a catheter to insert a miniature ultrasound probe into the vascular lumen. It then rotates to acquire images and displays cross-sectional images of the vessels, providing in vivo images of the vascular lumen. Determining the severity of lesions relies on image clarity and integrity, and the phenomenon of non-uniform rotation distortion (NURD) can affect image quality during imaging. According to the 2018 Expert Consensus on the Use of Intravascular Ultrasound in Coronary Artery Disease, NURD is caused by image distortion caused by the uneven rotation of the mechanically rotating IVUS catheter. During IVUS catheter imaging, uneven force on the metal spring within the IVUS catheter can lead to uneven torque transmission, resulting in a mismatch between the angular velocity of the transducer end and that of the catheter tip, causing NURD.

[0003] To address the NURD phenomenon, some existing technologies rely on post-processing image processing algorithms (such as image registration, correction, and machine learning algorithm interpolation) to correct the deformed image. However, these image processing algorithms often rely on fixed assumptions, have poor real-time performance, and are limited in effectiveness, making them inapplicable to all NURD situations. Summary of the Invention

[0004] In view of this, the present invention discloses an image acquisition system to solve the image distortion problem from the source and reduce the dependence on post-processing of the image.

[0005] An image acquisition system comprises: a rotating retraction device, an imaging device and an image acquisition device, wherein a positioning mechanism is installed on the rotating motor of the rotating retraction device;

[0006] The rotation and retraction device is used to drive the imaging device to perform a rotation and retraction movement in the blood vessel upon receiving a start instruction, and simultaneously trigger the positioning mechanism to send an image acquisition instruction to the image acquisition device;

[0007] The image acquisition device is used to acquire an image from the imaging device after receiving the image acquisition instruction, and use an image recognition algorithm to determine a NURD area on the image, and send the NURD area to the rotation and retraction device;

[0008] The rotary retraction device is further used to adjust the motor control signal at the next moment according to the current motor control signal collected by the positioning mechanism based on reducing the NURD area, so as to dynamically adjust the motor speed of the rotating motor.

[0009] Optionally, the rotary retraction device comprises: a rotary retraction device body and a retraction guide rail;

[0010] The rotating and retracting device body includes: the positioning mechanism, the driver, the rotating motor and the retracting motor;

[0011] The driver is used to drive the rotating motor and the retracting motor to work;

[0012] The positioning mechanism is mounted on the rotating motor and sends the image acquisition instruction to the image acquisition device when the rotating motor rotates;

[0013] The rotating motor is used to drive the slip ring and the conduit connecting plug to rotate;

[0014] The retraction guide rail is used to install the rotating retraction device body, so that the retraction motor drives the rotating retraction device body to move along the retraction guide rail in the opposite direction of the imaging device.

[0015] Optionally, the rotary retraction device body further comprises: a slip ring and a catheter connecting plug;

[0016] The slip ring is used to transmit signals between the imaging device and the image acquisition device;

[0017] The catheter connecting plug is used to drive the imaging device to perform a rotational retraction movement.

[0018] Optionally, the image acquisition device is further used to:

[0019] After receiving each image acquisition subcommand in the image acquisition instruction, acquiring each image line from the imaging device, wherein each image line corresponds to a position of the positioning mechanism;

[0020] After all the image acquisition subcommands in the image acquisition instruction have acquired corresponding lines of the image, all acquired lines of the image are used to generate the image corresponding to the image acquisition instruction.

[0021] Optionally, the process of the image acquisition device using an image recognition algorithm to determine the NURD area includes:

[0022] Obtaining a grayscale change matrix based on the correlation of each row in the image pixel matrix corresponding to the image;

[0023] The NURD region is obtained based on the correlation of each column in the grayscale change matrix.

[0024] Optionally, obtaining a grayscale change matrix based on the correlation of each row in an image pixel matrix corresponding to the image includes:

[0025] Calculating the row variance of each row of the image pixel matrix of the image, and setting pixel values ​​whose row variance is greater than a first variance threshold to 1, and setting pixel values ​​whose row variance is not greater than the first variance threshold to 0, to obtain a first sequence as a valid image region;

[0026] Performing smoothing on the first sequence using a window function, and determining the starting position and the ending position of the longest subsequence in the smoothed first sequence;

[0027] When the length of the longest subsequence is greater than 0, searching for a corresponding row in the image pixel matrix based on the starting position and the ending position of the longest subsequence;

[0028] The grayscale change matrix is ​​obtained by calculating the grayscale change of the pixels in the previous and next rows of each row of image pixels.

[0029] Optionally, obtaining the NURD region based on the correlation of each column in the grayscale change matrix includes:

[0030] Calculating the column variance of each column of the grayscale change matrix, and setting the columns whose column variance is greater than the second variance threshold to 1, and setting the columns whose column variance is not greater than the second variance threshold to 0, to obtain a second sequence;

[0031] The second sequence is smoothed using the window function, and a target sequence having a length greater than a preset length in the smoothed second sequence is determined as the NURD region.

[0032] Optionally, when the motor control signal is a motor control pulse, the rotation and retraction device is further configured to adjust the motor control signal at the next moment based on the motor control signal at the current moment collected by the positioning mechanism based on reducing the NURD area, and the process includes:

[0033] Get the NURD area;

[0034] Determine a Normal area corresponding to a motor control pulse area based on the NURD area;

[0035] Determining a current moment adjustment factor of the motor control pulse according to preset conditions, wherein the preset conditions at least include: a change trend of the NURD region;

[0036] The motor control pulse at the next moment is adjusted based on the current moment adjustment factor.

[0037] Optionally, the process of the rotation and retraction device determining the current adjustment factor of the motor control pulse according to a preset condition includes:

[0038] The current moment adjustment factor is determined according to the NURD region sequence length at the previous moment, the NURD region sequence length at the current moment, and the previous moment adjustment factor.

[0039] Optionally, the image acquisition system further includes: a PC;

[0040] The image acquisition device is used to acquire images from the imaging device and send the images to the PC for display.

[0041] Optionally, the image acquisition device is further configured to use an image recognition algorithm to determine the NURD area on the image, and send the NURD area to the rotation and retraction device.

[0042] Optionally, the PC is further configured to use an image recognition algorithm to determine the NURD area on the image, and send the NURD area to the rotary retraction device.

[0043] From the above technical solution, it can be seen that the present invention discloses an image acquisition system, including: a rotary retraction device, an imaging device and an image acquisition device, a positioning mechanism is installed on the rotary motor of the rotary retraction device, and when the rotary retraction device receives the start instruction, the rotary motor drives the imaging device to perform a rotary retraction movement in the blood vessel, and at the same time triggers the positioning mechanism to send an image acquisition instruction to the image acquisition device, so that the image acquisition device synchronously acquires images from the imaging device, and uses an image recognition algorithm to determine the NURD area of ​​the image and sends it to the rotary retraction device, and the rotary retraction device adjusts the motor control signal at the next moment based on the reduction of the NURD area, thereby realizing dynamic adjustment of the motor speed. The present invention realizes the mapping of the image and the motor position by installing a positioning mechanism on the rotary motor and performing image acquisition when the positioning mechanism is triggered, and realizes the mapping of the image and the motor position by identifying the NURD area from the acquired image and adjusting the motor control signal at the next moment based on the NURD area, thereby effectively reducing the NURD phenomenon caused by uneven motion of the imaging device during the imaging process at the next moment, solving the image distortion problem from the source, and reducing the dependence on post-processing of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without any creative work.

[0045] Figure 1 A schematic structural diagram of an image acquisition system disclosed in an embodiment of the present invention;

[0046] Figure 2 This is a physical diagram of the internal modules of a rotary retraction device disclosed in an embodiment of the present invention;

[0047] Figure 3 A schematic structural diagram of another image acquisition system disclosed in an embodiment of the present invention;

[0048] Figure 4 This is a flow chart of a method for an image acquisition device to determine a NURD area using an image recognition algorithm according to an embodiment of the present invention;

[0049] Figure 5 A flowchart of a method disclosed in an embodiment of the present invention for adjusting a motor control signal at a next moment based on a current motor control signal collected by a positioning mechanism, by a rotating retraction device based on reducing a NURD area, when the motor control signal is a motor control pulse;

[0050] Figure 6 A schematic structural diagram of an imaging device disclosed in an embodiment of the present invention;

[0051] Figure 7 A schematic diagram of an image captured by an imaging device disclosed in an embodiment of the present invention;

[0052] Figure 8 A schematic diagram of a NURD region identified from an image disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0054] An embodiment of the present invention discloses an image acquisition system, comprising: a rotary retraction device, an imaging device, and an image acquisition device. A positioning mechanism is installed on a rotary motor of the rotary retraction device. When the rotary retraction device receives a start command, the rotary motor drives the imaging device to perform a rotary retraction motion within a blood vessel, and at the same time triggers the positioning mechanism to send an image acquisition command to the image acquisition device, so that the image acquisition device synchronously acquires an image from the imaging device, and uses an image recognition algorithm to determine a NURD region of the image and sends it to the rotary retraction device. The rotary retraction device adjusts the motor control signal at the next moment based on the reduction of the NURD region and the current moment motor control signal to achieve dynamic regulation of the motor speed. The present invention achieves mapping of the image and the motor position by installing a positioning mechanism on the rotary motor and performing image acquisition when the positioning mechanism is triggered. By identifying the NURD region from the acquired image and adjusting the motor control signal at the next moment based on the NURD region, the NURD phenomenon caused by uneven motion of the imaging device during the imaging process is effectively reduced at the next moment, thereby solving the image distortion problem at the source and reducing the dependence on post-processing image processing.

[0055] See also Figure 1 , a structural diagram of an image acquisition system disclosed in an embodiment of the present invention, the system includes: a rotating retraction device 10, an imaging device 20 and an image acquisition device 30, wherein a positioning mechanism is installed on the rotating motor of the rotating retraction device 10.

[0056] The positioning mechanism in this application is a device that characterizes a positional relationship. In practical applications, the positioning mechanism includes but is not limited to: an encoder, a magnetic sensor, an electric sensor, etc.

[0057] The rotation and retraction device 10 is used to drive the imaging device 20 to perform a rotation and retraction movement in the blood vessel based on the rotation motor when receiving a start instruction, and simultaneously trigger the positioning mechanism to send an image acquisition instruction to the image acquisition device 30.

[0058] In practical applications, the start instruction received by the rotating and retracting device 10 can be sent by an external device of the image acquisition system. When a PC is also provided in the image acquisition system, the start instruction received by the rotating and retracting device 10 can be sent by the PC.

[0059] The rotary retraction device 10 is connected to the imaging device 20. After receiving a start command, the rotary retraction device 10 controls the rotation of the rotary motor within the rotary retraction device 10, thereby driving the imaging device 20 to perform a rotary retraction movement within the blood vessel. While driving the imaging device 20 to perform a rotary retraction movement within the blood vessel, the rotary retraction device 10 triggers an internal positioning mechanism to send an image capture command to the image acquisition device 30, causing the image acquisition device 30 to capture images from the imaging device 20. This ensures that the image capture process is synchronized with the rotational position of the rotary motor, ensuring that each captured image frame corresponds to a specific position of the imaging device 20. This eliminates image distortion caused by the asynchrony between the image capture moment and the rotation moment of the rotary motor, further improving image stability and consistency.

[0060] The image acquisition device 30 is used to acquire an image from the imaging device 20 after receiving the image acquisition instruction, determine a NURD area of ​​the image using an image recognition algorithm, and send the NURD area to the rotation and retraction device 10.

[0061] In practical applications, when the image acquisition device 30 receives multiple image acquisition instructions in succession, it can acquire an image from the imaging device 20 for each image acquisition instruction and determine the NURD region corresponding to each image. Based on the size changes of each NURD region, the change trend of the NURD region can be determined. The NURD region is also the area where the NURD phenomenon is located.

[0062] The rotary retraction device 10 is further configured to adjust the motor control signal at the next moment according to the current motor control signal collected by the positioning mechanism based on reducing the NURD area, so as to dynamically adjust the motor speed of the rotary motor.

[0063] The present application utilizes a positioning mechanism to monitor the rotation speed and position of the rotating motor in real time, thereby determining the motor control signal at each moment. The rotating retraction device 10 uses the NURD area as a feedback condition, and based on reducing the NURD area, adjusts the motor control signal at the next moment according to the motor control signal at the current moment, thereby achieving the adjustment of the motor speed of the rotating motor. As a result, the rotational retraction movement of the imaging device 20 driven by the rotating motor in the blood vessel is also adjusted accordingly, ensuring the precise control of the speed of each position of the imaging device 20 during the rotation process in the blood vessel, and reducing the NURD phenomenon caused by uneven rotation of the imaging device 20.

[0064] Preferably, the motor control signal can be a motor control pulse, voltage, current or other signal.

[0065] In summary, the present invention discloses an image acquisition system, comprising: a rotating retraction device 10, an imaging device 20 and an image acquisition device 30. A positioning mechanism is installed on the rotating motor of the rotating retraction device 10. When the rotating retraction device 10 receives a start command, the rotating motor drives the imaging device 20 to perform a rotating retraction movement in the blood vessel, and at the same time triggers the positioning mechanism to send an image acquisition instruction to the image acquisition device 30, so that the image acquisition device 30 synchronously acquires images from the imaging device 20, and uses an image recognition algorithm to determine the NURD area of ​​the image and sends it to the rotating retraction device 10. The rotating retraction device 10 adjusts the motor control signal at the next moment based on reducing the NURD area and combines the motor control signal at the current moment to achieve dynamic adjustment of the motor speed. The present invention achieves mapping of images and motor positions by installing a positioning mechanism on a rotating motor and performing image acquisition when the positioning mechanism is triggered. By identifying the NURD area from the acquired image and adjusting the motor control signal at the next moment based on the NURD area, the NURD phenomenon caused by uneven movement of the imaging device 20 during the imaging process is effectively reduced at the next moment, solving the image distortion problem from the source and reducing dependence on post-image processing.

[0066] In one embodiment, see Figure 2 , a physical view of the internal module of a rotary retraction device disclosed in an embodiment of the present invention, the rotary retraction device includes: a rotary retraction device body 11 and a retraction guide rail 12.

[0067] The rotating and retracting device body 11 includes a positioning mechanism 111 , a driver 112 , a rotating motor 113 and a retracting motor 114 .

[0068] The driver 112 is used to drive the rotating motor 113 and the retracting motor 114 to work.

[0069] The positioning mechanism 111 is installed on the rotating motor 113 and sends the image acquisition instruction to the image acquisition device 30 when the rotating motor 113 rotates.

[0070] The rotary motor 113 is used to drive the slip ring 115 and the catheter connection plug to rotate 116. The catheter connection plug rotates 116 and is connected to the imaging device 20, so that the rotary motor 113 drives the imaging device 20 to perform a rotational retraction movement in the blood vessel.

[0071] The retraction guide rail 12 is used to mount the rotating retraction device body 11 , so that the retraction motor 114 drives the rotating retraction device body 11 to move along the retraction guide rail 12 in a direction opposite to the imaging device 20 .

[0072] During image capture, driver 112 drives rotating motor 113 and retracting motor 114 at a set speed. Driven by rotating motor 113, slip ring 115 and imaging device 20 rotate. Positioning mechanism 111 sends an image capture instruction to image capture device 30 each time it is triggered. Image capture occurs when positioning mechanism 111 is triggered, ensuring that each captured image line has a corresponding positioning mechanism position.

[0073] In one embodiment, the rotary retraction device body 11 may further include a slip ring 115 and a catheter connection plug 116 .

[0074] The slip ring 115 is used to transmit signals between the imaging device 20 and the image acquisition device 30 .

[0075] The catheter connecting plug 116 is connected to the catheter head end of the imaging device 20 , and the catheter connecting plug 116 is used to drive the imaging device 20 to perform a rotational retraction motion.

[0076] It should be noted that the image acquisition instruction sent by the rotational retraction device 10 to the image acquisition device 30 by triggering the internal positioning mechanism 111 includes multiple image acquisition sub-commands. In actual application, the rotational retraction device 10 sends one image acquisition sub-command to the image acquisition device 30 each time the internal positioning mechanism 111 is triggered.

[0077] Therefore, in one embodiment, the image acquisition device 30 may also be used to:

[0078] (1) After receiving each image acquisition subcommand in the image acquisition instruction, each image line is acquired from the imaging device 20 .

[0079] Each line of the image corresponds to a position of the positioning mechanism.

[0080] It should be noted that an image is composed of multiple image lines. Each time the image acquisition device 30 receives an image acquisition sub-command, it acquires an image line from the imaging device 20. When the received image lines can form a complete image, the image acquisition is completed.

[0081] (2) After all the image acquisition subcommands in the image acquisition instruction have acquired corresponding lines of the image, all acquired lines of the image are used to generate the image corresponding to the image acquisition instruction.

[0082] The number of image acquisition subcommands included in the image acquisition instruction in this application is the same as the number of lines of the image that constitutes the image. The specific value depends on actual needs and is not limited by the present invention.

[0083] When the NURD area is identified based on the captured image, since each line of the image corresponds to the position of the positioning mechanism, the NURD area can be mapped to the motor end position of the rotary retraction device 10, that is, the NURD area can be mapped to the position of the rotary motor 113 of the rotary retraction device 10. The present application can control the instantaneous speed of the target area through the intelligent speed control system until the NURD area is reduced to below the area threshold.

[0084] In one embodiment, see Figure 3 The image acquisition device 30 may include: an ultrasound acquisition card 31 and a PC 32 .

[0085] The ultrasound acquisition card 31 is used to acquire images from the imaging device 20 and send the images to the PC 32 for display.

[0086] In practical applications, the image captured by the ultrasound acquisition card 31 can be used to identify the NURD area within the ultrasound acquisition card 31, or after the image is sent to the PC 32, the PC 32 can identify the NURD area of ​​the image.

[0087] Therefore, the ultrasound acquisition card 31 can also be used for:

[0088] The NURD area is determined by using an image recognition algorithm on the image, and the NURD area is sent to the rotary retraction device 10 .

[0089] Alternatively, the PC 32 may also be used to determine the NURD area using an image recognition algorithm on the image, and send the NURD area to the rotation and retraction device 10 .

[0090] It should be noted that no matter whether the ultrasonic acquisition card 31 in the image acquisition device 30 recognizes the NURD region in the image or the PC 32 recognizes the NURD region in the image, the image recognition algorithm used is the same.

[0091] Among them, the image recognition algorithm used by the image acquisition device 30 of the present application to identify the NURD area in the image includes but is not limited to the correlation between lines in the image sequence. When the correlation is greater than a certain threshold, it is determined to be a NURD area, or the image sequence is identified using machine learning or deep learning methods. NURD area.

[0092] It should be noted that, in practical applications, the image acquisition device 30 may also only include the ultrasound acquisition card 31 .

[0093] When the image acquisition device 30 only includes the ultrasound acquisition card 31 , the image acquisition device 30 can be used as an independent device, or integrated into the rotating and retracting device 10 , or integrated into a PC.

[0094] In one embodiment, see Figure 4 , a flowchart of a method for an image acquisition device to determine a NURD area of ​​an image using an image recognition algorithm disclosed in an embodiment of the present invention, the method may include:

[0095] Step S101: Obtain a grayscale change matrix based on the correlation of each row in the image pixel matrix corresponding to the image.

[0096] Step S102: Obtain the NURD region based on the correlation of each column in the grayscale change matrix.

[0097] For ease of understanding, the implementation process of step S101 and step S102 is as follows:

[0098] Step S101 may specifically include:

[0099] (1) Calculate the row variance of each row of the image pixel matrix of the image, and set the pixel values ​​whose row variance is greater than a first variance threshold to 1, and set the pixel values ​​whose row variance is not greater than the first variance threshold to 0, thereby obtaining a first sequence as a valid image area.

[0100] The dimension of the image pixel matrix of the image in this application is [M×N], where M is the number of pixels and N is the number of lines of the image, that is, the number of image lines contained in the image.

[0101] Assume that the sequence of variances of each row of the image pixel matrix is , Represents the mean value in the calculation sequence, the first variance threshold The calculation formula is as follows:

[0102] .

[0103] (2) Smoothing the first sequence using a window function, and determining the starting position and the ending position of the longest subsequence in the smoothed first sequence.

[0104] The window function in the present application is a window function with a length of a preset value. For example, the window function is a window function with a length of 5.

[0105] The starting and ending positions of the longest subsequence can be expressed using [start, end].

[0106] (3) When the length of the longest subsequence is greater than 0, a corresponding row is found in the image pixel matrix based on the starting position and the ending position of the longest subsequence.

[0107] Among them, when the length of the longest subsequence is not greater than 0, it indicates that there is no NURD area in the image.

[0108] (4) Calculate the grayscale change of each row of image pixels and the grayscale change matrix of the previous and next rows of image pixels.

[0109] The grayscale change matrix B obtained by calculating the grayscale change of the pixels in the previous and next rows of each row of image pixels has a dimension of [M×N]. The grayscale change matrix B is calculated as follows:

[0110] ;

[0111] Where Q represents the image pixel matrix, A represents the first sequence as the effective image area, and the diff function represents the difference between adjacent elements.

[0112] The above formula means: according to the elements in the first sequence A, the corresponding row is found in the image pixel matrix Q, and the difference between adjacent elements in this row is calculated.

[0113] The implementation process of step S102 is as follows:

[0114] (1) Calculate the column variance of each column of the grayscale change matrix, set the column whose column variance is greater than the second variance threshold to 1, and set the column whose column variance is not greater than the second variance threshold to 0, to obtain a second sequence.

[0115] Assume that the sequence composed of the variance of each column of the grayscale change matrix B is , the second variance threshold The calculation formula is as follows:

[0116] .

[0117] (2) Smoothing the second sequence using the window function, and determining a target sequence in the smoothed second sequence whose length is greater than a preset length as the NURD region.

[0118] The window function in the present application is a window function with a length of a preset value, for example, the window function is a window function with a length of 5.

[0119] In practical applications, the motor moves by receiving pulses. Assuming that the motor rotates one circle when it receives M pulses, the number of pulses generated by the positioning mechanism 111 when it rotates one circle is N. There is a fixed conversion relationship between M and N. For the target speed , it can be understood that the motor needs to complete a circle within a fixed time. Assuming that the pulse time interval of the motor input is The time for one rotation is , and The relationship is as follows:

[0120] ;

[0121] and The relationship is as follows:

[0122] .

[0123] It should be noted that in practical applications, in addition to using Figure 4 The method for determining the NURD region described in the illustrated embodiment can also be used in other ways. For example, the image acquisition system can be combined with images from other modalities (such as magnetic resonance images and CT images) to assist in determining the NURD region. By fusing relevant information from images of different modalities and utilizing information on tissue structure and lesion characteristics from other images, the information that may be missing from the images in some cases can be supplemented, thereby more accurately identifying and defining the NURD region and improving the reliability and accuracy of the diagnosis.

[0124] In addition, based on the image recognition algorithm of this application, it is also possible to increase the recognition of NURD areas based on deep learning or machine learning, including but not limited to using deep learning or machine learning to detect NURD areas or using machine learning or deep learning as an alternative to any of the steps described in this application.

[0125] In one embodiment, see Figure 5 , an embodiment of the present invention discloses a flow chart of a method for adjusting a motor control signal at a next moment based on a current motor control signal collected by a positioning mechanism, when the motor control signal is a motor control pulse, by rotating a retraction device based on reducing the NURD area. The method may include:

[0126] Step S201: Get the NURD area.

[0127] Among them, the NURD area can be expressed as follows:

[0128] ;

[0129] Where, Indicates the NURD region sequence, Indicates the NURD region sequence corresponding to ID number j, Indicates the NURD region sequence corresponding to ID number k.

[0130] Sequences j and k are between 1 and M. j and k represent the starting ID number and the ending ID number of the NURD region, respectively.

[0131] Step S202: determining a Normal region corresponding to a motor control pulse sequence based on the NURD region.

[0132] Divide the M pulse intervals into motor control pulse sequences, then we have , represents the motor control pulse train, Indicates the motor control pulse with position ID 1. Representing the motor control pulse with position ID M, the following formula can be obtained:

[0133] ;

[0134] Where, Indicates the pulse time interval of the pulse control sequence in which the NURD area is currently located. represents the pulse at the i-th position, represents the pulse time interval at the i-th position.

[0135] i represents the number of generations used in the calculation of the accumulated symbols, and the sequences j and k are contained between 1 and M. j and k represent the starting ID number and the ending ID number of the NURD area, respectively.

[0136] For sequences without NURD regions Denoted as:

[0137] ;

[0138] Similarly, the expression of the time interval is as follows:

[0139] ;

[0140] At this point, the area corresponding to the motor control pulse sequence is divided into the Normal area and the NURD area.

[0141] Step S203: determining the current adjustment factor of the motor control pulse according to a preset condition.

[0142] The preset conditions at least include: a change trend of the NURD area.

[0143] For the regulating factor , its value is adjusted according to the changing trend of the NURD area.

[0144] First, determine whether the NURD area is larger than the set detection threshold If it is greater than the set detection threshold Then calculate the length of its NURD region sequence. Here and at the initial moment, the length of the NURD region sequence is .

[0145] Regulatory factors Assign an initial value, the initial value is calculated as follows:

[0146] ;

[0147] Where, It is an experience value ranging from 0 to 1.

[0148] In this application, the current moment adjustment factor is determined based on the NURD region sequence length at the previous moment, the NURD region sequence length at the current moment, and the previous moment adjustment factor.

[0149] Specifically, for the current moment adjustment factor The calculation formula is as follows:

[0150] ;

[0151] Where, For the previous moment sequence length, For the current moment sequence length, is the adjustment factor at the previous moment.

[0152] Step S204 : adjusting the motor control pulse at the next moment based on the current moment adjustment factor.

[0153] In order to achieve speed regulation, the regulation factor is set to .

[0154] The formula After transformation, we get the following formula:

[0155] .

[0156] To facilitate calculation, each subsequence and Time interval within If the spacing is set to be equal, the formula obtained by the above deformation can be used to reduce time to achieve the purpose of accelerating the NURD area. Combined with the formula The pulse interval within the sequence interval can be calculated, and similarly, the The pulse interval within the interval is combined into the pulse sequence at the current moment according to the positioning mechanism sequence. Then it is output to the motor to complete a round of motor control process.

[0157] The entire intelligent control system uses the NURD area as a feedback condition to adjust the width of the motor pulse sequence at the next moment.

[0158] In one embodiment, see Figure 6 , a schematic structural diagram of an imaging device disclosed in an embodiment of the present invention, the device may include: a catheter head end 21, a transducer 22 and a metal spring 23.

[0159] The catheter tip 21 is connected to the rotation and retraction device 10 , the transducer 22 is connected to the image acquisition device 30 , one end of the metal spring 23 is connected to the catheter tip 21 , and the other end of the metal spring 23 is connected to the transducer 22 .

[0160] In this application, the images captured by the imaging device can be found in Figure 7 As shown, in this application, Figure 7 The NURD regions identified from the images are shown in Figure 8 shown.

[0161] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0162] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0163] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An image acquisition system, characterized in that: include: A rotating retraction device, an imaging device and an image acquisition device, wherein a positioning mechanism is installed on the rotating motor of the rotating retraction device; The rotation and retraction device is used to drive the imaging device to perform rotation and retraction movement in the blood vessel upon receiving a start instruction, and simultaneously trigger the positioning mechanism to send an image acquisition instruction to the image acquisition device; The image acquisition device is used to acquire an image from the imaging device after receiving the image acquisition instruction, and use an image recognition algorithm to determine a NURD area on the image, and send the NURD area to the rotation and retraction device; The rotary retraction device is further used to adjust the motor control signal at the next moment according to the current motor control signal collected by the positioning mechanism based on reducing the NURD area, so as to dynamically adjust the motor speed of the rotating motor.

2. The image acquisition system according to claim 1, wherein: The rotary retraction device comprises: a rotary retraction device body and a retraction guide rail; The rotating and retracting device body includes: the positioning mechanism, the driver, the rotating motor and the retracting motor; The driver is used to drive the rotating motor and the retracting motor to work; The positioning mechanism is mounted on the rotating motor and sends the image acquisition instruction to the image acquisition device when the rotating motor rotates; The rotating motor is used to drive the slip ring and the conduit connecting plug to rotate; The retraction guide rail is used to install the rotating retraction device body, so that the retraction motor drives the rotating retraction device body to move along the retraction guide rail in the opposite direction of the imaging device.

3. The image acquisition system according to claim 2, characterized in that: The rotary retraction device body further comprises: a slip ring and a catheter connecting plug; The slip ring is used to transmit signals between the imaging device and the image acquisition device; The catheter connecting plug is used to drive the imaging device to perform a rotational retraction movement.

4. The image acquisition system according to any one of claims 1 to 3, characterized in that: The image acquisition device is also used for: After receiving each image acquisition subcommand in the image acquisition instruction, acquiring each image line from the imaging device, wherein each image line corresponds to a position of the positioning mechanism; After all the image acquisition subcommands in the image acquisition instruction have acquired corresponding lines of the image, all acquired lines of the image are used to generate the image corresponding to the image acquisition instruction.

5. The image acquisition system according to claim 1, wherein: The process of the image acquisition device using an image recognition algorithm to determine the NURD area includes: Obtaining a grayscale change matrix based on the correlation of each row in the image pixel matrix corresponding to the image; The NURD region is obtained based on the correlation of each column in the grayscale change matrix.

6. The image acquisition system according to claim 5, characterized in that: The step of obtaining a grayscale change matrix based on the correlation of each row in the image pixel matrix corresponding to the image includes: Calculating the row variance of each row of the image pixel matrix of the image, and setting pixel values ​​whose row variance is greater than a first variance threshold to 1, and setting pixel values ​​whose row variance is not greater than the first variance threshold to 0, to obtain a first sequence as a valid image region; Performing smoothing on the first sequence using a window function, and determining the starting position and the ending position of the longest subsequence in the smoothed first sequence; When the length of the longest subsequence is greater than 0, searching for a corresponding row in the image pixel matrix based on the starting position and the ending position of the longest subsequence; The grayscale change matrix is ​​obtained by calculating the grayscale change of the pixels in the previous and next rows of each row of image pixels.

7. The image acquisition system according to claim 5 or 6, characterized in that: The obtaining of the NURD region based on the correlation of each column in the grayscale change matrix includes: Calculating the column variance of each column of the grayscale change matrix, and setting the columns whose column variance is greater than the second variance threshold to 1, and setting the columns whose column variance is not greater than the second variance threshold to 0, to obtain a second sequence; The second sequence is smoothed using the window function, and a target sequence having a length greater than a preset length in the smoothed second sequence is determined as the NURD region.

8. The image acquisition system according to claim 1, wherein: When the motor control signal is a motor control pulse, the rotation and retraction device is further configured to adjust the motor control signal at the next moment based on the motor control signal at the current moment collected by the positioning mechanism based on reducing the NURD area. The process includes: Get the NURD area; Determine a Normal area corresponding to a motor control pulse area based on the NURD area; Determining a current moment adjustment factor of the motor control pulse according to preset conditions, wherein the preset conditions at least include: a change trend of the NURD region; The motor control pulse at the next moment is adjusted based on the current moment adjustment factor.

9. The image acquisition system according to claim 8, characterized in that: The process of the rotary retraction device determining the current moment adjustment factor of the motor control pulse according to the preset conditions includes: The current moment adjustment factor is determined according to the NURD region sequence length at the previous moment, the NURD region sequence length at the current moment, and the previous moment adjustment factor.

10. The image acquisition system according to claim 1, wherein: The image acquisition system also includes: a PC; The image acquisition device is used to acquire images from the imaging device and send the images to the PC for display.

11. The image acquisition system according to claim 1, wherein: The image acquisition device is further configured to use an image recognition algorithm to determine the NURD area on the image, and send the NURD area to the rotation and retraction device.

12. The image acquisition system according to claim 10, wherein: The PC is further configured to determine the NURD area using an image recognition algorithm on the image, and send the NURD area to the rotary retraction device.

Images