Lymph fluorescence tracing system based on isothioblue injection

Abstract

The invention provides a lymphatic fluorescence tracing system based on an isothioblue injection, the lymphatic fluorescence tracing system comprises a light-closed cover body, a laser generator, a light source, a camera, a processor unit and a projection unit, the light source and the camera are integrated in the light-closed cover body, and the light source is connected with the laser generator; the processor unit is connected with the camera and the projection unit. The lymphatic vessel is subjected to fluorescence development through the isothioblue injection, a developed image is shot through the camera, the image is processed through the processor unit and then sent to the projection unit, and a lymphatic vessel tracing image is directly and accurately projected to a tracing area through the projection unit. According to the system, in the fluorescent tracing process, a light source in an operating room does not need to be turned off, a lymphatic vessel fluorescent path can be directly and accurately projected and imaged to a skin lymphatic tracing position, and medical staff can conveniently and visually observe and capture the lymphatic vessel tracing position.

Description

Technical Field

[0001] The present invention mainly relates to the technical field related to fluorescence tracing, and specifically to a lymphatic fluorescence tracing system based on isosulfan blue injection. Background Art

[0002] The metastasis of cancer cells to lymph nodes through lymphatic vessels is the main way for breast cancer to spread locally. The axillary lymph node staging determined by surgical removal of axillary lymph nodes is an important basis for breast cancer prognosis and adjuvant treatment decisions. Sentinel lymph node biopsy has become the first choice for axillary staging of early breast cancer patients with negative axillary lymph nodes, allowing many early breast cancer patients to avoid complications such as upper limb edema.

[0003] Sentinel lymph nodes are defined as the first lymph nodes where tumors metastasize. In actual operation, the location of sentinel lymph nodes can be tracked by fluorescent tracing. The general operation is to first inject a tracer near the breast. The tracer has a fluorescent imaging function and can produce a fluorescent effect under the irradiation of incident light of a specific wavelength. Then, the first lymph node to be removed can be found according to the guidance of the fluorescent path, and the metastasis can be determined by pathological diagnosis. The key to ensuring the success rate and accuracy of sentinel lymph node biopsy lies in the tracing technology.

[0004] As mentioned above, in current clinical technology, the application of fluorescent tracers mainly involves medical personnel injecting fluorescent tracers and then irradiating them with incident light to draw lines on the patient's skin surface according to the fluorescent tracing path. They then follow the guidance of the drawn lines to dissect the relevant tissues along the path to find the sentinel lymph nodes. The main technical problems in the current existing technology are as follows: since fluorescence requires incident light irradiation, and the light such as the shadowless lamp used in the operating room will seriously affect the incident light and the fluorescence effect, therefore, in order for clinical medical staff to clearly observe the fluorescent tracing route of the lymphatic vessels in the skin, they need to turn off the light source in the operating room. According to investigation, each tracing process requires the light source in the operating room to be turned on and off multiple times, which is very inconvenient to operate; the existing fluorescent tracing system mainly includes a laser generator, a light source, a camera and a display screen. The light source is connected to the laser generator, and the light source and the camera are integrated on a handle. Medical staff operate the handle to make the light source irradiate the tracing area, thereby developing fluorescence. However, fluorescence is difficult to observe intuitively with the naked eye, so it is necessary to collect the light beam through the photosensitive element of the camera and convert it into an electrical signal to send it to the computer, which is displayed on the display screen on one side. Medical staff need to observe the fluorescent route on the display screen and then draw a line on the skin. Because the drawing position is different from the display position, it is time-consuming and laborious to draw the line and it is difficult to accurately capture the accurate fluorescence path. Summary of the invention

[0005] In order to solve the shortcomings of current technology, the present invention combines the existing technology and provides a lymphatic fluorescence tracing system based on isosulfan blue injection from the perspective of practical application. This system makes it possible to directly and accurately project the fluorescent path of the lymphatic vessels onto the skin lymphatic tracing position without turning off the light source in the operating room during the fluorescence tracing process, thereby facilitating medical personnel to intuitively observe and capture the position of the traced lymphatic vessels.

[0006] The technical solution of the present invention is as follows: The lymphatic fluorescence tracing system based on isosulfan blue injection includes: a light-blocking cover, a laser generator, a light source, a camera, a processor unit and a projection unit. The light source and the camera are integrated in the light-blocking cover. The light source is connected to the laser generator; the processor unit is connected to the camera and the projection unit. The tracing method of the tracing system is as follows: (1) Mark multiple points in the area where lymphatic tracing is required to form basic positioning points, inject isosulfan blue injection in the tracing area, cover the area where tracing is required with a light-tight cover body that integrates a light source and a camera, turn on the laser generator to allow the light source to emit incident light, and obtain the initial image I1 of the tracing area under the light-tight cover body through the camera and send it to the processor unit for storage. The initial image I1 includes the fluorescent lymphatic vessel path and basic positioning point information; (2) The light-blocking cover is removed, and the initial image stored in the processor unit is projected onto the tracing area through the projection unit. The image to be processed I2 is obtained through the camera integrated in the projection unit. The image to be processed I2 includes the projection image and basic positioning point information. The image to be processed I2 is sent to the processor unit for processing; (3) The processor unit constructs a translation vector matrix and a rotation angle vector matrix based on the position information of the projection positioning point and the basic positioning point in the image to be processed I2, wherein the projection positioning point is the basic positioning point in the projection image, and transforms any pixel point in the projection image based on the translation vector matrix and the rotation angle vector matrix to obtain an adjusted projection correction image; (4) The processor unit transmits the processed projection correction image to the projection unit, and the projection unit re-projects the projection correction image to obtain a lymphatic vessel tracing route image in the tracing area.

[0007] Furthermore, step (3) also includes constructing a thin plate spline interpolation TPS model based on the translation vector matrix and the rotation angle vector matrix, and transforming any pixel point in the projection image based on the thin plate spline interpolation TPS model.

[0008] Further, step (3) specifically includes: (3.1) Based on feature point detection and matching, the coordinate information of the projection positioning points and the basic positioning points are extracted from the image to be processed I2; (3.2) Calculate the translation vector for each projection positioning point and the corresponding base positioning point, and construct the translation vector matrix; (3.3) For any two adjacent projection positioning points and basic positioning points, calculate the vectors to obtain the projection positioning point vector and the basic positioning point vector. Use the vector dot product to calculate the cosine value of the angle between the projection positioning point and the basic positioning point to obtain the rotation angle and construct the rotation angle vector matrix. (3.4) constructing a thin plate spline interpolation TPS model and obtaining a model parameter vector, wherein the model parameter vector includes a weight vector associated with a radial basis function and an affine transformation parameter vector; (3.5) For any pixel point in the projection image, the coordinates after thin plate spline interpolation transformation are calculated to obtain the transformed projection correction image.

[0009] Further, step (3.4) specifically includes: (3.4.1) Calculate the distance matrix between each pair of basic positioning points to obtain the radial basis function matrix; (3.4.2) For two adjacent basic positioning points , , and the corresponding projection positioning points are , , calculate the rotation angle difference between adjacent projection positioning points , and the translation difference , ; (3.4.3) Construct a new distance metric: , in, , , is the weight coefficient; (3.4.4) Use Recalculate the radial basis function matrix ， ; (3.4.5) Construct constraint matrix , based on the constraint matrix , radial basis function matrix , translation vector matrix , The linear equations are solved to obtain the weight vector associated with the radial basis function , and the affine transformation parameter vector , .

[0010] Further, step (3.5) specifically includes: (3.5.1) For any pixel in the projected image , calculate its relative to each projection location point Direction vector And the direction vector and Angle of positive axis ; (3.5.2) Define the adjustment function , , calculate the rotation correction term , ; (3.5.3) Calculate the coordinates after TPS transformation , ,in , The calculation is as follows: + , , + , ; , , , The calculation is as follows: , + + + , , + + + ; in, , They are , The direction is the weight vector associated with the radial basis function, , is the affine transformation parameter vector associated with the translation transformation, , yes Affine transformation parameter vector related to orientation scaling and rotation, , yes Affine transformation parameter vector related to orientation scaling and rotation.

[0011] Furthermore, the rotation correction term in step (3.5.2) , The calculation method is as follows: , .

[0012] Furthermore, in step (2), before the initial image stored in the processor unit is projected to the tracing area by the projection unit, the initial image is first segmented, and the lymphatic vessel fluorescence path and the basic positioning point are extracted as the foreground category to obtain the lymphatic vessel fluorescence path. , basic positioning point Binarized mask image of lymphatic vessel fluorescence path , basic positioning point The color value of the projected image is extracted after setting the color value.

[0013] Furthermore, the tracing method of the tracing system further comprises the following steps: (6) The information of the projection correction image is obtained again through the camera, and the average position difference between the projection positioning point and the basic positioning point in the projection correction image is calculated. A threshold is set. When the average difference is less than the set threshold, the projection correction image is retained. When the average difference is greater than or equal to the set threshold, the process returns to step (3) for iterative processing.

[0014] Furthermore, the wavelength range of the incident light emitted by the laser generator is 620nm-690nm.

[0015] Furthermore, at least three basic positioning points are set and are scattered and marked in the lymphatic tracing area by a marker.

[0016] Beneficial effects of the present invention: 1. Isosulfan blue is a safe and highly sensitive dye. After subcutaneous injection, it can be selectively absorbed by lymphatic vessels and quickly enter the lymphatic system. It has low diffusion in soft tissues and is simple, fast and intuitive. The lymphatic fluorescent tracing system provided by the present invention is based on isosulfan blue injection. The new use of isosulfan blue as a fluorescent tracer has been clinically verified. This use retains the dye tracing of isosulfan blue and uses its fluorescent tracing function at the same time.

[0017] 2. When the lymphatic fluorescence tracing system of the present invention is in use, the camera and the light source are integrated into the light-proof cover through the setting of the light-proof cover. Therefore, when fluorescence development and collection are required, they are not affected by external light sources. Therefore, there is no need to turn off the light source in the operating room, which facilitates the operation of medical staff.

[0018] 3. When using the lymphatic fluorescence tracing system of the present invention, taking into account the problem that lines cannot be drawn directly on the skin due to the influence of the light-blocking cover, the collected images are processed by projection means, so that the fluorescent path of the lymphatic vessels can be directly and accurately projected on the skin, which is convenient for medical personnel to draw lines or perform other operations.

[0019] 4. The lymphatic fluorescence tracing system of the present invention fully considers the skin deformation and projection deformation caused by the light-blocking cover. By setting multiple reference positioning points, the projection image is compared and adjusted to ensure the accuracy of the projection image and the accurate projection of the lymphatic fluorescence path. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of the system of the present invention.

[0021] Figure 2 The figure is a flow chart of the tracing method of the present invention.

[0022] Figure 3 Schematic diagram of the image taken by the camera after fluorescent development of lymphatic vessels.

[0023] Figure 4 Schematic diagram of lymphatic vessels based on the tracer path. DETAILED DESCRIPTION

[0024] The present invention will be further described with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope limited by the application.

[0025] This embodiment provides a lymphatic fluorescence tracing system based on isosulfan blue injection, the main structural principle is as follows Figure 1 As shown, it includes: a light-blocking cover 1, a laser generator 2, a light source 3, a camera 4, a processor unit 5 and a projection unit 6.

[0026] Among them, the light source 3 and the camera 4 are integrated in the light-blocking housing 1, and the light source 3 is connected to the laser generator 2 for emitting incident light; the processor unit 5 is connected to the camera 4 and the projection unit 6, and is used to obtain and save image data, perform algorithm processing on the image data, etc., and the camera is also integrated on the projection unit 6. In this tracing system, the bottom surface of the light-blocking housing 1 can be set as a flexible and deformable structure, which can have a certain curvature or special-shaped structure so as to adapt to the size characteristics of the breast to the armpit position. The light-blocking housing 1 should be able to fit tightly to the skin surface, completely cover the area to be traced, and completely isolate the external light source. The light source 3 and the camera 4 are both located in the light-blocking housing 1. The light source 3 generates incident light to irradiate the tracing area to develop fluorescence. The camera 4 uses an autofocus camera, and its photosensitive element can collect fluorescent light beams. After the image collected by the camera 4 is transmitted to the processor unit 5, the processor unit 5 can display the fluorescent path. The projection unit 6 can be set on a multi-degree-of-freedom adjustment mechanism, so that the projection image can be located in the accurate tracing area as much as possible through mechanical adjustment of the position of the projection unit 6.

[0027] The fluorescent tracing system of this embodiment refers to Figure 2 As shown, the main tracing methods are as follows.

[0028] (1) Mark multiple points in the area where lymphatic tracing is required to form basic positioning points, inject isosulfan blue injection in the tracing area, cover the area where tracing is required with a light-tight cover body that integrates a light source and a camera, turn on the laser generator to allow the light source to emit incident light, and obtain the initial image I1 of the tracing area under the light-tight cover body through the camera and send it to the processor unit for storage. The initial image I1 includes the fluorescent lymphatic vessel path and basic positioning point information.

[0029] (1.1) Medical staff can roughly determine the location of lymphatic vessels based on their experience. They first use a brightly colored marker to draw scattered dots near the lymphatic vessels. There should be at least three dots as basic positioning points. The basic positioning points are located on the skin and need to be clearly visible and can be captured by the camera.

[0030] (1.2) Inject 1 ml of 10-fold diluted isosulfan blue injection at the corresponding points around the areola. Isosulfan blue is a safe and highly sensitive dye. After subcutaneous injection, it can be selectively absorbed by lymphatic vessels and quickly enter the lymphatic system. It has low diffusion in soft tissues and is simple, fast, and intuitive. It is currently the only sentinel lymph node tracer approved by the FDA, with an accuracy rate of more than 97%. The main side effect in clinical application is allergy (0.25-2%), which generally returns to normal within 60 minutes. The blue dyeing of the skin tattoo can be eliminated in a short time. The mechanism of action is based on the physiological lymphatic system circulation. After subcutaneous injection, isosulfan blue injection binds to serum proteins, the lymphatic circulation system recycles proteins, and the lymphatic vessels are displayed by blue dye. Isosulfan blue also has a fluorescence effect. When the wavelength of the incident light is in the range of 620nm-690nm, isosulfan blue injection produces fluorescence with a wavelength of 690nm-790nm. Therefore, it is feasible to use isosulfan blue injection as a fluorescent tracer. The use of isosulfan blue injection makes it have both dye and fluorescent properties, which can improve the tracing efficiency in lymphatic vessel tracing.

[0031] (1.3) The light-blocking hood 1 is covered over the area to be traced. At this time, the traced area is completely dark and not affected by the light in the operating room. The laser generator is turned on to generate incident light of 620nm-690nm. The incident light is irradiated to the traced area through the light source 3. At the same time, the camera 4 collects the image inside the light-blocking hood 1. The image is defined as the initial image I1 and sent to the processor unit 5 for storage and processing. The information of the initial image I1, the fluorescent development path information, also includes the information of multiple basic positioning points.

[0032] (1.4) The processor unit 5 processes the initial image I1. This processing step mainly extracts the fluorescence path and basic positioning points separately for projection, so as to facilitate the subsequent calibration of the image. At the same time, it also makes the projected image only contain the required image information, which is convenient for the subsequent operation of medical personnel. During the processing, the initial image I1 is first segmented, and the lymphatic fluorescence path and basic positioning points are used as the foreground category, and other images are used as the background for training. The lymphatic fluorescence path is extracted separately based on the image sample. , basic positioning point It should be understood that the current image segmentation technology based on deep learning is already very mature. According to image training, the lymphatic vessel fluorescence path is extracted from the image separately. , basic positioning point In order to make it easier for the camera to capture the projected image and for medical staff to observe after it is projected onto the skin, the lymphatic fluorescence path can also be set. , basic positioning point The color value is achieved by assigning the color channel of the image.

[0033] (2) The light-blocking cover 1 is removed, and the initial image stored in the processor unit is projected onto the tracing area through the projection unit. The image to be processed I2 is obtained through the camera integrated in the projection unit. The image to be processed I2 includes the projected image and basic positioning point information. The image to be processed I2 is sent to the processor unit for processing.

[0034] In step (1), the required raw image data is obtained through the light-blocking cover 1 without turning off the external light source. At this time, the light-blocking cover 1 can be removed to expose the tracing area. The processed raw image data is projected onto the skin through the projection unit 6. The projection unit 6 can first adjust the position and focal length of the projection unit 6 through a multi-degree-of-freedom adjustment platform or manual adjustment, so that the basic positioning point (hereinafter referred to as the projection positioning point) in the image projected by the projection unit 6 coincides with the basic positioning point marked on the skin as much as possible. However, because the initial image is collected in the light-blocking cover 1, the light-blocking cover 1 is close to the skin and may cause skin deformation to a certain extent. On the other hand, due to the position of the projection unit 6 and the influence of the projection unit 6 itself, the projected image will not be a completely real image. Therefore, it is necessary to process the projection image based on the previously marked basic positioning point.

[0035] To process the projected image, it is necessary to capture the image again through the camera. The image should include the information of the basic positioning points marked on the skin and the information of the projected image. The captured image is defined as the image to be processed I2 and is sent to the processor unit for processing.

[0036] (3) The processor unit constructs a translation vector matrix and a rotation angle vector matrix based on the position information of the projection positioning point and the basic positioning point in the image to be processed I2, wherein the projection positioning point is the basic positioning point in the projection image, and transforms any pixel point in the projection image based on the translation vector matrix and the rotation angle vector matrix to obtain an adjusted projection correction image.

[0037] (3.1) Based on feature point detection and matching, the coordinate information of the projection positioning points and the basic positioning points are extracted in the image to be processed I2. Through an image analysis algorithm (such as an algorithm based on feature point detection and matching, such as a feature extraction algorithm such as SURF, ORB, combined with a method such as RANSAC for accurate point matching), the coordinates of the basic positioning points marked on the skin and the coordinates of the projection positioning points are extracted in the image to be processed I2.

[0038] (3.2) Calculate the translation vector for each projection positioning point and the corresponding base positioning point, and construct the translation vector matrix.

[0039] For each projection point and the corresponding marked base anchor point Calculate the translation vector: - , - ; Constructing the translation vector matrix and .

[0040] (3.3) For any two adjacent projection positioning points and basic positioning points, calculate the vectors to obtain the projection positioning point vector and the basic positioning point vector. Use the vector dot product to calculate the cosine value of the angle between the projection positioning point and the basic positioning point to obtain the rotation angle and construct the rotation angle vector matrix. The angle pre-selection calculation method is: ,in, is the projection positioning point vector, Construct the rotation angle vector matrix based on the basic positioning point vector .

[0041] (3.4) Construct a thin plate spline interpolation TPS model and obtain a model parameter vector, where the model parameter vector includes a weight vector related to the radial basis function and an affine transformation parameter vector.

[0042] (3.4.1) Calculate the distance matrix between the two basic positioning points to obtain the radial basis function matrix, the distance matrix ; (3.4.2) For two adjacent basic positioning points , , and the corresponding projection positioning points are , , calculate the rotation angle difference between adjacent projection positioning points , and the translation difference , ; The elements in the rotation angle vector matrix calculate the rotation angle difference , and calculate the translation difference in the same way , .

[0043] (3.4.3) Construct a new distance metric: , in, , , is a weight coefficient determined by experiments or prior knowledge, which is used to balance the contribution of spatial distance, rotation angle difference, and translation difference in the new distance metric.

[0044] (3.4.4) Use Recalculate the radial basis function matrix ， ; (3.4.5) Construct constraint matrix , based on the constraint matrix , radial basis function matrix , translation vector matrix , The linear equations are solved to obtain the weight vector associated with the radial basis function , and the affine transformation parameter vector , .

[0045] Among them, the constraint matrix , the constructed linear equations are as follows: = , = .

[0046] By solving the equation, we get the parameter vector , as well as , . , , , Related to the translation transformation, in the form of an affine transformation it can be viewed as and The translation constant part in the direction, , Mainly When considering rotation and scaling, these coefficients are related to The coordinates of the points are combined to change For example, in a rotation operation, they work together with the elements of the rotation matrix to determine the position and shape of a point in the direction; during a scaling operation, they determine the The scaling factor for the direction. , Mainly In the calculation process of the TPS model, these affine transformation parameters work together with the weight vector to accurately calculate the transformed coordinates of any pixel in the projected image, thereby achieving precise adjustment of the projected image to adapt to the effects of skin deformation (including translation, rotation, scaling, etc.).

[0047] (3.5) For any pixel point in the projection image, the coordinates after thin plate spline interpolation transformation are calculated to obtain the transformed projection correction image.

[0048] (3.5.1) For any pixel in the projected image , calculate its relative to each projection location point Direction vector And the direction vector and Angle of positive axis ; ( - , - ), ( - , - ).

[0049] (3.5.2) Define the adjustment function , (The function can be a linear function), calculate the rotation correction term , ; , .

[0050] Here is the calculation When the direction rotation correction term is used, the input parameter of the adjustment function is added , the purpose is to take into account Direction and There is a 90° phase difference when the direction rotates. Through such adjustment, The rotation correction term of the direction can more accurately reflect the rotation of The influence of coordinates is different from x Correction of direction.

[0051] (3.5.3) Calculate the coordinates after TPS transformation , ,in , The calculation is as follows: + , , + , .

[0052] , , , The calculation is as follows: , + + + , , + + + .

[0053] in, , They are , The direction is the weight vector associated with the radial basis function, , is the affine transformation parameter vector associated with the translation transformation, , yes Affine transformation parameter vector related to orientation scaling and rotation, , yes Affine transformation parameter vectors related to directional scaling and rotation. When calculating the translation vector, this technology mainly focuses on the actual movement of the positioning point position, without involving changes in shape and size. However, in the affine transformation parameters of the TPS model, the concept of scaling can be reflected. When these parameters are combined with the pixel coordinates for calculation, the image can be scaled to adapt to possible deformations such as skin stretching or compression, thereby providing image processing accuracy.

[0054] After the coordinate transformation is performed as described above, all pixel points in the projected image are transformed in coordinates, thereby achieving automatic adjustment of the image.

[0055] (4) The processor unit transmits the processed projection correction image to the projection unit, and the projection unit re-projects the projection correction image to obtain a lymphatic vessel tracing route image in the tracing area.

[0056] (5) In order to further ensure the authenticity of the projected image, the projected image can be processed multiple times. For example, the information of the projection correction image can be obtained again through the camera, and the average position difference between the projection positioning point and the basic positioning point in the projection correction image is calculated. A threshold is set. When the average difference is less than the set threshold, the projection correction image is retained. When the average difference is greater than or equal to the set threshold, return to step (3) and iterate until the average difference is less than the set threshold.

[0057] The images processed by the above steps have one-to-one correspondence between the projection positioning points in the projection correction image and the basic positioning points on the actual skin, thereby ensuring the authenticity of the projection image. The lymphatic tracing route projected on the skin is displayed on the skin in clear colors and is exactly located at the actual lymphatic route. Medical personnel can quickly determine the actual position of the lymphatic vessel based on the projection image, thereby performing other operations such as accurate line drawing or direct surgery. Figure 3 An example of drawing a line along a lymphatic vessel is shown in the figure.

Claims

1. A lymphatic fluorescence tracing system based on isosulfan blue injection, characterized in that: include: A light-blocking housing, a laser generator, a light source, a camera, a processor unit and a projection unit. The light source and the camera are integrated in the light-blocking housing. The light source is connected to the laser generator. The processor unit is connected to the camera and the projection unit. The tracing method of the tracing system is as follows: (1) Mark multiple points in the area where lymphatic tracing is required to form basic positioning points, inject isosulfan blue injection in the tracing area, cover the area where tracing is required with a light-tight cover body that integrates a light source and a camera, turn on the laser generator to allow the light source to emit incident light, and obtain the initial image I1 of the tracing area under the light-tight cover body through the camera and send it to the processor unit for storage. The initial image I1 includes the fluorescent lymphatic vessel path and basic positioning point information; (2) The light-blocking cover is removed, and the initial image stored in the processor unit is projected onto the tracing area through the projection unit. The image to be processed I2 is obtained through the camera integrated in the projection unit. The image to be processed I2 includes the projection image and basic positioning point information. The image to be processed I2 is sent to the processor unit for processing; (3) The processor unit constructs a translation vector matrix and a rotation angle vector matrix based on the position information of the projection positioning point and the basic positioning point in the image to be processed I2, wherein the projection positioning point is the basic positioning point in the projection image, and transforms any pixel point in the projection image based on the translation vector matrix and the rotation angle vector matrix to obtain an adjusted projection correction image; (4) The processor unit transmits the processed projection correction image to the projection unit, and the projection unit re-projects the projection correction image to obtain a lymphatic vessel tracing route image in the tracing area.

2. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 1, characterized in that: Step (3) also includes constructing a thin plate spline interpolation TPS model based on the translation vector matrix and the rotation angle vector matrix, and transforming any pixel point in the projection image based on the thin plate spline interpolation TPS model.

3. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 2, characterized in that: Step (3) specifically includes: (3.1) Based on feature point detection and matching, the coordinate information of the projection positioning points and the basic positioning points are extracted from the image to be processed I2; (3.2) Calculate the translation vector for each projection positioning point and the corresponding base positioning point, and construct the translation vector matrix; (3.3) For any two adjacent projection positioning points and basic positioning points, calculate the vectors to obtain the projection positioning point vector and the basic positioning point vector. Use the vector dot product to calculate the cosine value of the angle between the projection positioning point and the basic positioning point to obtain the rotation angle and construct the rotation angle vector matrix. (3.4) constructing a thin plate spline interpolation TPS model and obtaining a model parameter vector, wherein the model parameter vector includes a weight vector associated with a radial basis function and an affine transformation parameter vector; (3.5) For any pixel point in the projection image, the coordinates after thin plate spline interpolation transformation are calculated to obtain the transformed projection correction image.

4. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 3, characterized in that: Step (3.4) specifically includes: (3.4.1) Calculate the distance matrix between each pair of basic positioning points to obtain the radial basis function matrix; (3.4.2) For two adjacent basic positioning points , , and the corresponding projection positioning points are , , calculate the rotation angle difference between adjacent projection positioning points , and the translation difference , ; (3.4.3) Construct a new distance metric: ， in, , , is the weight coefficient; (3.4.4) Use Recalculate the radial basis function matrix ， ; (3.4.5) Construct constraint matrix , based on the constraint matrix , radial basis function matrix , translation vector matrix , The linear equations are solved to obtain the weight vector associated with the radial basis function , and the affine transformation parameter vector , .

5. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 4, characterized in that: Step (3.5) specifically includes: (3.5.1) For any pixel in the projected image , calculate its relative to each projection location point Direction vector And the direction vector and Angle of positive axis ; (3.5.2) Define the adjustment function , , calculate the rotation correction term , ; (3.5.3) Calculate the coordinates after TPS transformation , ,in , The calculation is as follows: + ， ， + ， ； , , , The calculation is as follows: ， + + + ， ， + + + ； in, , They are , The direction is the weight vector associated with the radial basis function, , is the affine transformation parameter vector associated with the translation transformation, , yes Affine transformation parameter vector related to orientation scaling and rotation, , yes Affine transformation parameter vector related to orientation scaling and rotation.

6. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 5, characterized in that: The rotation correction term in step (3.5.2) , The calculation method is as follows: ， 。 7. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 1, characterized in that: In step (2), before the initial image stored in the processor unit is projected to the tracing area by the projection unit, the initial image is segmented, and the lymphatic vessel fluorescence path and the basic positioning point are extracted as the foreground category to obtain the lymphatic vessel fluorescence path. , basic positioning point Binarized mask image of lymphatic vessel fluorescence path , basic positioning point The color value of the projected image is extracted after setting the color value.

8. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 1, characterized in that: The tracing method of the tracing system also includes the following steps: (5) The information of the projection correction image is obtained again through the camera, and the average position difference between the projection positioning point and the basic positioning point in the projection correction image is calculated. A threshold is set. When the average difference is less than the set threshold, the projection correction image is retained. When the average difference is greater than or equal to the set threshold, the process returns to step (3) for iterative processing.

9. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 1, characterized in that: The wavelength range of the incident light emitted by the laser generator is 620nm-690nm.

10. The lymphatic fluorescence tracing system based on isosulfan blue injection according to claim 1, characterized in that: At least three basic positioning points are set and are scattered and marked in the lymphatic tracing area by a marker.

Images