Multi-line endoscopic OCT anti-interference microvascular imaging algorithm and imaging system

Abstract

The present invention discloses a multi-line endoscopic OCT anti-interference microvascular imaging algorithm and imaging system. The algorithm includes utilizing a constructed multi-line endoscopic OCT imaging system for data acquisition; processing the data acquired in the current frame, calculating the spectrum information of the OCT structural imaging, and generating an OCT structural image; extracting and calculating blood flow information from the OCT structural image to generate a blood flow image; normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with a structural diagram; projecting the superimposed blood flow information structural diagram with principal component information to generate a vascular projection image; and performing three-dimensional spatial coordinate conversion on the projected information to convert it into three-dimensional spatial information with a tubular structure. The present application can utilize data acquired in a single frame for spectrum and structural analysis to obtain a blood flow image, overcome probe and tissue jitter, and has a simple structure that can effectively extract endoscopic microvascular imaging information.

Description

Technical Field

[0001] The present invention relates to the technical field of OCT imaging algorithms, and in particular to a multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm and imaging system. Background Art

[0002] Optical coherence tomography (OCT) is a noninvasive, high-resolution imaging technique developed in the 1990s. It uses the principle of optical low coherence to image light signals diffusely reflected from tissue. Optical coherence tomography (OCT) has been continuously evolving, from initial time-domain OCT to current frequency-domain OCT, offering faster imaging speeds and higher resolution. OCT's high resolution and penetrating properties have enabled its application not only in structural imaging but also in functional imaging. Based on OCT imaging and algorithm design, newly developed microvascular imaging techniques can image microvasculature within tissues. The basic principle is to acquire multiple frames at the same location and perform subtraction, retaining only motion information while removing static tissue structures. Because blood is flowing within blood vessels, this preserves vascular information, resulting in microvascular imaging. Existing microvascular imaging techniques fall into two main categories: amplitude-based methods, represented by the SSADA (Optical Spectral Separation Adaptation) algorithm; and phase-addition-based imaging algorithms, such as OMAG (Optical Microangiography). These two techniques are currently being applied in two fields: ophthalmic OCT and skin OCT. Both ophthalmic OCT and skin OCT utilize galvanometer-based scanning probes. During scanning, the probe can be kept stationary in space. Scanning is completed by simply keeping the sample stationary relative to the probe. Microvascular imaging is achieved through inter-frame information calculation. Both ophthalmic OCT and skin OCT microvascular imaging are performed in vitro. Currently, no technology exists for endoscopic microvascular imaging. This is because the endoscopic imaging probe is in free space and cannot be fixed, and the endoscopic tissue is constantly in motion with heartbeat and respiration. Using the two inter-frame calculation methods mentioned above, it is difficult to obtain stable and clear endoscopic OCT microvascular images. Patent CN108042125A proposes using a dual-balloon stabilization system to maintain inter-frame stability for endoscopic microvascular imaging, thereby utilizing frame correlation to obtain microvascular images. This method is feasible, but requires blood flow occlusion and complex catheter design. Patent CN111493930A also proposes endoscopic microvascular imaging, but the imaging method also utilizes frame-to-frame difference imaging and lacks any anti-interference or probe-tethering methods, making it ineffective. Patent CN111493832A proposes an en-face projection-based endoscopic microvascular imaging method. This method is also based on an inter-frame approach and cannot overcome interference imaging. This paper proposes a new endoscopic OCT anti-interference microvascular imaging technology that can overcome probe and tissue jitter and overcome the need for complex balloon catheters to achieve endoscopic OCT microvascular imaging. Summary of the Invention

[0003] Therefore, the purpose of the present invention is to provide an anti-interference microvascular imaging algorithm and system based on multi-line endoscopic OCT, which can use single-frame acquired data to perform spectral and structured analysis to obtain blood flow images, overcome the interference of probe and tissue jitter on the imaging results, and has a simple structure that can effectively extract endoscopic microvascular imaging information.

[0004] To achieve the above objectives, the present invention provides a multi-line endoscopic OCT anti-interference microvascular imaging algorithm, comprising the following steps:

[0005] S1. Data acquisition using the constructed multi-line endoscopic OCT imaging system;

[0006] S2. Processing the data collected in the current frame according to the preset parameters of the anti-interference algorithm, calculating the spectrum information of the OCT structural imaging, and generating an OCT structural image;

[0007] S3, extracting and calculating blood flow information from the OCT structural image to generate a blood flow image;

[0008] S4. normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with the structural diagram;

[0009] S5. Projecting the principal component information of the superimposed blood flow information structure map to generate a blood vessel projection image;

[0010] S6. Performing coordinate conversion of three-dimensional spatial information on the projected information, converting the two-dimensional image information into three-dimensional spatial information having a tubular structure.

[0011] Furthermore, in S2, the preset parameters of the anti-interference algorithm include setting the imaging dynamic range, setting the microvascular blood flow signal display range, and setting the spectral window transition region width.

[0012] Furthermore, in S2, the spectrum information of OCT structural imaging is calculated and the following formula is used to generate the OCT structural image:

[0013] ;

[0014] in, is the window function matrix for each line, For a B scan, including 5000 or 10000 lines, represents the fast Fourier transform, For OCT structural imaging.

[0015] Further preferably, in S3, when performing blood flow information extraction and calculation on the structural imaging to generate a blood flow image, the following steps are included:

[0016] The distal structure image information in the OCT structure image that deviates from the center by more than a preset distance is truncated;

[0017] Perform Fourier transform on the truncated image and then filter it;

[0018] Perform inverse Fourier transform on the filtered image.

[0019] Furthermore, blood flow information is extracted and calculated for structural imaging. When generating a blood flow image, the following formula is used:

[0020] ;

[0021] Among them, imcropRg is the defined truncation depth, which is used to truncate the structural image information farther away from the center of the image. is the defined filter function.

[0022] Furthermore, the filter function is calculated using the following formula:

[0023] ;

[0024] in, is the filter function The component factors of To generate a matrix function, The factors are expanded in the vertical direction. right The height of the image after truncation.

[0025] Furthermore, the The factor is expressed as follows:

[0026] ;

[0027] in, Indicates the spectral window width; represents the width of the spectral window transition region; n X represents the number of scan lines; is the Hamming window function; is a function of the spectral window width, It is a function of the spectral window width and the window width transition area.

[0028] Furthermore, in S4, before normalizing the blood flow image, the following formula is used to perform logarithm calculation on the structural image and the blood flow image, and then perform dynamic range calculation:

[0029] ;

[0030] ;

[0031] in, and Adjust the lower and upper limits of the dynamic range for the structure graph; and Adjust the lower and upper limits of the dynamic range for blood flow graphs; To take the logarithm For OCT structural imaging, is the depth of the defined cutoff; To take the structure diagram after logarithmic transformation and dynamic range calculation, Blood flow map after logarithmic transformation and dynamic range calculation.

[0032] The present invention also provides a multi-line-based endoscopic OCT anti-interference microvascular imaging system for implementing the multi-line-based endoscopic OCT anti-interference microvascular imaging method, comprising a swept frequency OCT imaging engine, a DOC motion control module, an image processing module, and an endoscopic probe;

[0033] The swept frequency OCT imaging engine is used to provide a high-speed swept frequency light source when the endoscopic probe performs endoscopic imaging;

[0034] The DOC motion control module is used to set the motion speed and retraction speed to control the motion of the endoscope probe;

[0035] The image processing module has a built-in anti-interference microvascular imaging algorithm for processing the data collected in the current frame, calculating the spectrum information of the OCT structural imaging, and generating the OCT structural image;

[0036] Perform blood flow information extraction and calculation on OCT structural images to generate blood flow images;

[0037] Normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with the structural image;

[0038] Perform principal component information projection on the superimposed blood flow information structure map to generate blood vessel projection imaging;

[0039] The projected information is subjected to coordinate transformation of three-dimensional spatial information, converting the two-dimensional image information into three-dimensional spatial information with a tubular structure.

[0040] Furthermore, before data acquisition, each module of the imaging system is initialized and set up as follows:

[0041] Set the sweep speed range of the swept frequency OCT imaging engine to: 200KHz~800KHz;

[0042] The rotation speed range of the DOC motion control module when controlling the movement speed and retraction speed is set to 40fps~160fps; the retraction speed is set to 10mm / s~40mm / s.

[0043] The multi-line endoscopic OCT anti-interference microvascular imaging algorithm and imaging system disclosed in this application have at least the following advantages over the existing technology:

[0044] The multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm and imaging system provided in this application can use single-frame acquired data to perform spectral and structural analysis to obtain blood flow images, overcoming the interference of probe and tissue jitter on the imaging results.

[0045] In the multi-line endoscopic OCT anti-interference microvascular imaging algorithm provided in this application, a matrix function is formed after window function transformation, and the matrix function is transformed into a filter function, thereby realizing the final conversion of spectral data into a blood flow image, simplifying the system structure, having a simple structure, and being able to effectively extract endoscopic microvascular imaging information. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of the multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm provided by the present invention.

[0047] Figure 2 This is a comparison diagram of image changes in lip endoscopic imaging scanned by an OCT catheter in an embodiment of the present invention.

[0048] Figure 3 This is a projection diagram of blood flow information in the lips scanned by the OCT catheter.

[0049] Figure 4 Endoscopic OCT blood flow image showing the three-dimensional spatial information of the lips scanned by the OCT catheter.

[0050] Figure 5 This is a projection of blood flow information in the palm of the hand scanned by the OCT catheter.

[0051] Figure 6 Endoscopic OCT blood flow image of the palm of the hand scanned by the OCT catheter for three-dimensional spatial information. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] like Figure 1 As shown, an embodiment of one aspect of the present invention provides a multi-line endoscopic OCT anti-interference microvascular imaging algorithm, comprising the following steps:

[0054] S1. Data acquisition using the constructed multi-line endoscopic OCT imaging system;

[0055] S2. Processing the data collected in the current frame according to the preset parameters of the anti-interference algorithm, calculating the spectrum information of the OCT structural imaging, and generating an OCT structural image;

[0056] S3, extracting and calculating blood flow information from the OCT structural image to generate a blood flow image;

[0057] S4. normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with the structural diagram;

[0058] S5. Projecting the principal component information of the superimposed blood flow information structure map to generate a blood vessel projection image;

[0059] S6. Performing coordinate conversion of three-dimensional spatial information on the projected information, converting the two-dimensional image information into three-dimensional spatial information having a tubular structure.

[0060] The multi-line endoscopic OCT imaging system mentioned above includes a swept-frequency OCT imaging engine, a DOC motion control module, an image processing module, and an endoscopic probe;

[0061] The swept frequency OCT imaging engine is used to provide a high-speed swept frequency light source when the endoscopic probe performs endoscopic imaging;

[0062] The DOC motion control module is used to set the motion speed and retraction speed to control the motion of the endoscope probe;

[0063] The image processing module has a built-in anti-interference microvascular imaging algorithm for processing the data collected in the current frame according to the anti-interference microvascular imaging algorithm.

[0064] The swept-frequency OCT imaging engine integrates a high-speed swept-frequency light source with a sweep speed of 200KHz to 800KHz. It also integrates modules such as a balanced detector and an interferometer delay line to form an electrically controllable Michelson interferometer system. The DOC motion control module sets the motion speed and retraction speed. The rotation speed is set to 40fps to 160fps, corresponding to the speed of the swept-frequency light source of 200KHz to 800KHz. The retraction speed is set to 10mm / s to 40mm / s, corresponding to the above rotation speed setting. An optical imaging catheter is connected to the front end of the DOC motion module, and an endoscopic probe is set at the distal end of the catheter. The DOC motion module and the optical imaging catheter form a sample arm.

[0065] Example 1

[0066] Here, two lips and the palm of a hand clenched into a closed cylindrical shape are used as specific collection and processing cases for illustration. Of course, the application scope of this method is not limited to lips and palms, and can be applied to other living parts.

[0067] Step 1: Use the constructed multi-line endoscopic OCT imaging system to collect data from the lips and palms;

[0068] The anti-interference microvascular imaging algorithm sets specific data parameters. The number of lines per frame is set between 5,000 and 10,000. The specific number of lines should be set based on the size of the scanning lumen, the speed of the swept light source, and the retraction distance. When the vessel lumen diameter is less than 2.5 mm, the number of lines per frame can be set to 5,000. When the vessel lumen diameter is greater than 2.5 mm, the number of lines per frame should be set to 10,000. The number of points per line is set based on the characteristics of the swept light source. After the settings are completed, the data of the lips and palms is collected and saved.

[0069] Step 2: Process the data collected in the current frame according to the preset parameters of the anti-interference algorithm, calculate the spectrum information of the OCT structural imaging, and generate the OCT structural image;

[0070] The anti-interference algorithm preset parameters include setting the imaging dynamic range, setting the microvascular blood flow signal display range, and setting the spectral window transition region width. The dynamic range Rgstr is defined to constrain the display effect of OCT structural imaging. The dynamic range Rgflw is used to constrain the display effect of microvascular blood flow signals, and RgtransitWin is set to determine the spectral window transition region width. Setting RgstruWin is used to define the spectral window width.

[0071] After completing the above parameter definitions, the next step is to calculate the OCT scan structure imaging of the lips and palms to generate the OCT structural image. The calculation formula is as follows:

[0072] ;

[0073] Where Bwin is the window function, which is defined as the window function matrix of each line; Bfram is a B scan, including 5000 or 10000 lines. The number of lines will be represented by nX in the following text. is fast Fourier transform; Bimgs is OCT structural imaging.

[0074] Step 3: Extract and calculate blood flow information from the OCT structural image to generate a blood flow image, including the following steps:

[0075] The distal structure image information in the OCT structure image that deviates from the center by more than a preset distance is truncated;

[0076] Perform Fourier transform on the truncated image and then filter it;

[0077] Perform inverse Fourier transform on the filtered image.

[0078] Blood flow information extraction and calculation are performed on the frequency domain information of lip and palm structure imaging:

[0079] ;

[0080] imcroprg defines a truncated depth, which cuts off the information farther from the center of the image.

[0081] in, is the defined filter function. The above formula means that the truncated image is subjected to Fourier transform and then filtered, and then the inverse Fourier transform is performed. The filter function is defined as follows:

[0082] ;

[0083] in, is the filter function The component factors. To generate the matrix function, The factors are expanded into matrix functions in the vertical direction. The definition is as follows:

[0084] ;

[0085] in, is the Hamming window function; is a function of the spectral window width, is a function of the spectral window width and the window width transition region. Indicates the spectral window width; Indicates the width of the spectral window transition region.

[0086] Logarithm calculation is performed on the structural image and blood flow image of the lips and palm data, and then dynamic range calculation is performed as follows:

[0087] ;

[0088] ;

[0089] in, and Adjust the lower and upper limits of the dynamic range for the structure graph; and Adjust the lower and upper limits of the dynamic range for blood flow graphs; To take the logarithm For OCT structural imaging, is the depth of the defined cutoff; To take the structure diagram after logarithmic transformation and dynamic range calculation, Blood flow map after logarithmic transformation and dynamic range calculation.

[0090] Step 4: normalize the blood flow image, color-label the normalized blood flow information, and overlay the labeled blood flow information with the structural diagram;

[0091] Normalize the blood flow images of the lips and palm data.

[0092] ;

[0093] in The normalization parameters for displaying blood flow images are specified.

[0094] The normalized blood flow information of the lips and palms is color-labeled, and the labeled blood flow information is superimposed on the structural diagram.

[0095] The process of steps 2-4 above is as follows Figure 2 As shown, the upper left is the OCT structure image after logarithmization, the lower left is the blood flow information image after logarithmization, the upper right is the normalized blood flow information image, and the lower right is the superposition of the normalized blood flow information image and the OCT structure image.

[0096] Step 5. Project the principal component information of the blood flow information obtained from the lips and palms. The principal component information here refers to the separation of the intensity components of the obtained blood flow information and then projection. The special principal component is defined as the difference between the signal information and the background information, which is greater than 0.2 compared to the background information. The background information is calculated as follows: set a threshold so that the intensity value greater than the threshold is the signal, and the average value of all intensity values ​​less than the threshold is taken as the background information, otherwise it is noise information. The effect of the projection of the blood flow information of the lips scanned by the OCT catheter is as follows Figure 3 The effect of the blood flow information of the palm scanned by the OCT catheter is shown in the following figure: Figure 5 shown.

[0097] Step 6: Perform coordinate conversion of the projected information into three-dimensional spatial information, converting the two-dimensional image information into three-dimensional spatial information with a tubular structure.

[0098] After conversion with Figure 3 The corresponding endoscopic OCT blood flow image of the lips using the OCT catheter to scan the three-dimensional spatial information is as follows: Figure 4 shown; with Figure 5 The corresponding OCT catheter scans the endoscopic OCT blood flow image of the three-dimensional spatial information of the palm of the hand as shown in the figure. Figure 6 shown.

[0099] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A multi-line endoscopic OCT anti-interference microvascular imaging algorithm, characterized by: The following steps are involved: S1. Data acquisition using the constructed multi-line endoscopic OCT imaging system; S2. Processing the data collected in the current frame according to the preset parameters of the anti-interference algorithm, calculating the spectrum information of the OCT structural imaging, and generating an OCT structural image; S3, extracting and calculating blood flow information from the OCT structural image to generate a blood flow image; comprising the following steps: truncating distal structural image information that deviates from the center by more than a preset distance in the OCT structural image; The truncated image is subjected to Fourier transform and then filtered. When filtering, the filter function is calculated using the following formula: BwenT=f rep (BwinThalf(x)| x＝1:end ,BwinThalf(x)| x＝end:1 ,nZcrop) Among them, BwinThalf is the component factor of the filter function BwenT, f rep To generate a matrix function for expanding the BwinThalf factor in the vertical direction, nZcrop represents the image height after the Bimgs structure image is truncated; Perform inverse Fourier transform on the filtered image; S4. normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with the structural diagram; S5. Projecting the principal component information of the superimposed blood flow information structure map to generate a blood vessel projection image; S6. Performing coordinate conversion of three-dimensional spatial information on the projected information, converting the two-dimensional image information into three-dimensional spatial information having a tubular structure.

2. The multi-line endoscopic OCT anti-interference microvascular imaging algorithm according to claim 1, characterized in that: In S2, the preset parameters of the anti-interference algorithm include setting the imaging dynamic range, setting the microvascular blood flow signal display range, and setting the spectral window transition region width.

3. The multi-line endoscopic OCT anti-interference microvascular imaging algorithm according to claim 1, characterized in that: In S2, the spectrum information of OCT structural imaging is calculated and the following formula is used to generate the OCT structural image: Among them, Bwin is the window function matrix of each line, Bfram is a B scan, including 5000 or 10000 lines, represents fast Fourier transform, and Bimgs represents OCT structural imaging.

4. The multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm according to claim 1, characterized in that: When extracting and calculating blood flow information from structural imaging and generating blood flow images, the following formula is used: Among them, imcropRg is the defined truncation depth, which is used to truncate the structural image information farther away from the center of the image, and BwenT is the defined filter function.

5. The multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm according to claim 1, characterized in that: The BwinThalf factor is expressed by the following formula: BwinThalf=f(f0(RgstruWin),winhann(RgtransitWin),f1(nX / 2-RgstruWin -RgtransitWin)) Where RgstruWin represents the spectral window width; RgtransitWin represents the width of the spectral window transition region; nX represents the number of scan lines; winhann is the Hamming window function; f0 is the function of the spectral window width, and f1 is the function of the spectral window width and the window width transition region.

6. The multi-line-based endoscopic OCT anti-interference microvascular imaging algorithm according to claim 1, characterized in that: In S4, before normalizing the blood flow image, the following formula is used to perform logarithm calculation on the structural image and the blood flow image, and then perform dynamic range calculation: imSint=f log ((|Bimgs(imcropRg)|)-RgStr(1)) / (RgStr(2)-RgStr(1)) imFint=f log ((|Bimgf|)-RgFlw(1)) / (RgFlw(2)-RgFlw(1)) Among them, RgStr(1) and RgStr(2) are the lower limit and upper limit of the dynamic range interval of the structure diagram adjustment; RgFlw(1) and RgFlw(2) are the lower limit and upper limit of the dynamic range interval of the blood flow diagram adjustment; f log is the logarithmic operation, Bimgs is the OCT structural imaging, imcropRg is the defined cutoff depth; imSint is the structural image after logarithmic transformation and dynamic range calculation, and imFint is the blood flow image after logarithmic transformation and dynamic range calculation.

7. A multi-line endoscopic OCT anti-interference microvascular imaging system, used to implement the multi-line endoscopic OCT anti-interference microvascular imaging method according to any one of claims 1 to 6, characterized in that: Includes swept frequency OCT imaging engine, DOC motion control module, image processing module and endoscopic probe; The swept frequency OCT imaging engine is used to provide a high-speed swept frequency light source when the endoscopic probe performs endoscopic imaging; The DOC motion control module is used to set the motion speed and retraction speed to control the motion of the endoscope probe; The image processing module has a built-in anti-interference microvascular imaging algorithm for processing the data collected in the current frame, calculating the spectrum information of the OCT structural imaging, and generating the OCT structural image; Perform blood flow information extraction and calculation on OCT structural images to generate blood flow images; Normalizing the blood flow image, color-labeling the normalized blood flow information, and superimposing the labeled blood flow information with the structural image; Perform principal component information projection on the superimposed blood flow information structure map to generate blood vessel projection imaging; The projected information is subjected to coordinate transformation of three-dimensional spatial information, converting the two-dimensional image information into three-dimensional spatial information with a tubular structure.

8. The multi-line endoscopic OCT anti-interference microvascular imaging system according to claim 7, characterized in that: Before data acquisition, each module of the imaging system is initialized as follows: Set the sweep speed range of the sweep OCT imaging engine to: 200KHz ~ 800KHz; The rotation speed range of the DOC motion control module when controlling the motion speed and retraction speed is set to 40fps~160fps; the retraction speed is set to 10mm / s~40mm / s.

Images