Treatment thermal field visualization method based on OCT (Optical Coherence Tomography)

Abstract

The invention discloses an OCT (Optical Coherence Tomography)-based treatment thermal field visualization method, which relates to the technical field of biomedical imaging and comprises the following steps: S100, acquiring an original signal; s200, preprocessing the original signal; s300, OCT signal processing is carried out; s400, calculating a speckle decorrelation value; s500, calculating a speckle decorrelation rate; s600, speckle decorrelation rate signal processing is carried out; and S700, structure and thermal field image visualization is carried out. Compared with the prior art, the method has the advantages that the robustness to speckles and decorrelation noise is higher, the sensitivity to time interval selection is reduced, the synchronous real-time display of the OCT structure and the thermal field image is realized, the real-time synchronous monitoring of the structure and thermal field imaging in the tissue thermal treatment process is met, and the popularization of OCT in minimally invasive surgery and thermal treatment is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of biomedical imaging technology, and in particular to an OCT-based therapeutic thermal field visualization method. Background Art

[0002] Optical coherence tomography (OCT) has gained widespread recognition as an important diagnostic tool in ophthalmology due to its high-resolution imaging and non-invasive nature. Its applications have also expanded to other medical fields, such as cardiology, dermatology, and gastroenterology, where it provides detailed tomographic images of internal structures.

[0003] OCT is currently transitioning from a diagnostic tool to a visualization tool for adjunctive therapy. It is already being used to monitor various types of treatments. Among them, thermal ablation therapies, including laser and radiofrequency ablation, are known for their minimally invasive nature, reduced risk of complications, and adaptability to a variety of therapeutic areas. These therapies have been used to treat various types of tumors and cardiovascular diseases such as atherosclerosis and atrial fibrillation. By providing real-time imaging, OCT can observe the thermal ablation process, potentially improving the precision and safety of the procedure by ensuring accurate targeting of the heating energy, thereby minimizing the impact on adjacent healthy tissue.

[0004] However, visualizing therapeutic thermal fields using OCT structural imaging faces challenges, primarily due to the fact that the optical properties of tissues change little with increasing temperature until near the degeneration point of 50°C, as described in the paper "Thermal therapy modalities for cancer treatment: A review and future perspectives" (Mellal, Idir, et al. International Journal of Applied Science - Research and Review. 4.02 (2017): 1-11). This limitation, inspired by the success of OCT angiography (OCTA), has prompted the exploration of alternative OCT-based approaches that can visualize blood flow without labels, as described in the paper "Optical coherencetomography-based angiography" (Chen, Chieh-Li, and Ruikang K. Wang. Biomedicaloptics express 8.2 (2017): 1056-1082). Recent studies have employed algorithms based on speckle decorrelation, a popular technique in OCTA, to monitor thermal therapy. For example, Lo et al. applied the complex differential variance (CDV) algorithm to achieve two-dimensional visualization of thermal coagulation dynamics during laser ablation, as described in the paper “Laser thermal therapy monitoring using complex differential variance in optical coherencetomography” (Lo, William CY, et al. Journal of biophotonics 10.1 (2017): 84-91). Guo et al. used the speckle variance (SV) algorithm to track real-time temperature changes in human skin tissue samples removed during laser ablation, as described in the paper “Intraoperative speckle variance optical coherencetomography for tissue temperature monitoring during cutaneous laser therapy” (Guo, Shoujing, et al. IEEE journal of translational engineering in health and medicine 7 (2019): 1-8).In addition, Maltais-Tariant et al. calculated the speckle intensity decorrelation that occurs during tissue coagulation, providing a method for monitoring treatment depth, as described in the paper "Real-time co-localized OCT surveillance of laser therapy using motion corrected speckledecorrelation" (Maltais-Tariant, Raphaël, et al. Biomedical Optics Express 11.6 (2020): 2925-2950.).

[0005] Compared to structural OCT, speckle decorrelation-based methods provide enhanced visualization of areas of elevated temperature, but they still have some limitations. First, they often suffer from low signal-to-noise ratios (SNRs), which can mask subtle changes in the speckle pattern that indicate temperature changes within the tissue, leading to inaccurate assessments of the effectiveness of thermal treatments. Second, the method is sensitive to the choice of the time interval used in the decorrelation calculation. The optimal time interval should be long enough to capture changes in the speckle pattern due to thermal effects, but not so long that it includes excessive noise or uncorrelated tissue motion. Choosing an inappropriate time interval can lead to underestimation or overestimation of the extent of thermal damage.

[0006] Therefore, those skilled in the art are committed to developing an OCT-based method for visualizing therapeutic thermal fields. Summary of the Invention

[0007] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is how to improve the signal-to-noise ratio and reduce the sensitivity to time interval selection, while meeting the requirements of real-time synchronous monitoring of the structure and thermal field imaging during tissue thermal treatment.

[0008] Research has found that the decorrelation rate of OCT speckle can be used as a new contrast mechanism to visualize the therapeutic thermal field. It has many advantages. First, because it is a decorrelation trend rather than a decorrelation value at a specific time, it is more robust to speckle and decorrelation noise, and can characterize therapeutic temperature changes with a high signal-to-noise ratio; second, it shows better insensitivity to the choice of time interval in the decorrelation calculation, making it more widely used in thermal ablation treatments with different heating conditions (power, duty cycle, etc.). In addition, considering that monitoring thermal ablation treatments requires real-time processing speed for visualization methods, the simultaneous real-time display of OCT structure and thermal field images is achieved to meet the real-time synchronous monitoring of the structure and thermal field imaging during tissue thermal treatment.

[0009] In one embodiment of the present invention, a method for visualizing a therapeutic thermal field based on OCT is provided, comprising the following steps: S100, original signal acquisition, continuously acquiring interference spectrum signals of the heating treatment thermal field area at the same position during tissue thermal treatment, and saving the signals; S200, preprocessing the original signal, performing signal preprocessing on the saved interference spectrum signal to obtain a preprocessed OCT complex domain signal, and saving the signal; S300, OCT signal processing, performing modulo and logarithmic transformation on the pre-processed OCT complex domain signal, mapping it into tissue light scattering B-scan OCT signal data, and standardizing it to obtain standardized B-scan OCT signal data, and saving it; S400, speckle decorrelation value calculation, using the pre-processed OCT complex domain signal to synchronously calculate speckle decorrelation values ​​corresponding to different time intervals at the current moment; S500, speckle decorrelation rate calculation, speckle decorrelation values ​​corresponding to different time intervals are used to calculate the speckle decorrelation rate through linear regression fitting, and the slope of the fitting line is used to obtain speckle decorrelation rate signal data; S600, speckle decorrelation rate signal processing, removing outliers from the speckle decorrelation rate signal data to obtain outlier-removed speckle decorrelation rate signal data, and saving the data; S700, structural and thermal field image visualization, performs grayscale mapping operation on the standardized B-scan OCT signal data, outputs it in image form, and displays OCT structural image information in real time; performs pseudo-color mapping operation on the speckle decorrelation rate signal data after removing outliers, outputs it in image form, and displays OCT thermal field image information synchronized with the OCT structural image information in real time.

[0010] Furthermore, in the OCT-based therapeutic thermal field visualization method in the above embodiment, signal preprocessing includes: Spectral k-domain linear correction, uniform sampling in the wavenumber k-domain, using multiple spline interpolation calculations; DC term removal, baseline artifact removal, averaging the elements at the same depth in multiple continuously acquired interferometric spectral signals to obtain an average value, and then subtracting the average value from each element of the interferometric spectral signal; Dispersion compensation, which compensates for the optical material and optical path difference between the sample arm and reference arm of the OCT system; Spectrum window shaping: select the window function and multiply the interference spectrum signal to perform spectrum shaping to suppress the generation of sidelobe noise; Fast Fourier transform maps the interference spectrum signal from the wavenumber k domain to the depth coordinate information of the sample.

[0011] Furthermore, in the OCT-based therapeutic thermal field visualization method in the above embodiment, the dispersion compensation method includes hardware compensation or polynomial compensation.

[0012] Furthermore, in the OCT-based therapeutic thermal field visualization method in the above embodiment, hardware compensation is to add a dispersion compensation element in the reference arm to match the dispersion characteristics of the sample arm.

[0013] Furthermore, in the OCT-based therapeutic thermal field visualization method in the above embodiment, the dispersion compensation element includes but is not limited to compensation materials, gratings or fiber stretchers.

[0014] Furthermore, in the OCT-based therapeutic thermal field visualization method in the above embodiment, the window function includes but is not limited to a Hanning window, a Hamming window, or a Gaussian window.

[0015] Optionally, in the OCT-based therapeutic thermal field visualization method in any of the above embodiments, the B-scan OCT signal data is normalized using the following formula: in, and Respectively expressed as i Row, No. j Columns of normalized B-scan OCT signal data before and after and They are B-scan OCT signal data and their average values, respectively.

[0016] Optionally, in the OCT-based therapeutic thermal field visualization method in the above embodiment, the B-scan OCT signal data is standardized to an 8-bit unsigned integer data type in the range of 0-255.

[0017] Optionally, in the OCT-based therapeutic thermal field visualization method in any of the above embodiments, the speckle decorrelation value uses the following formula: in, x and z are the lateral and axial positions in the B-scan section of OCT, A is the amplitude of the OCT speckle signal, It is the inverse of the OCT system frame rate, that is, the time interval between two adjacent B scans. and are the current moment and the time interval used for decorrelation calculation, yes A positive integer multiple of N is the number of B scans used for averaging, which helps improve the signal-to-noise ratio. n From 0 toN-1 An integer.

[0018] Optionally, in the OCT-based therapeutic thermal field visualization method in any of the above embodiments, the speckle decorrelation rate is calculated using the following formula: in, M is the number of time points used to calculate the slope, express, m From 1 to M An integer.

[0019] Optionally, in the OCT-based therapeutic thermal field visualization method in any of the above embodiments, removing outliers from the speckle decorrelation rate signal data includes reassigning speckle decorrelation rate signal data less than 0 to a region centered on itself and taking the average of 3×3 neighborhood element values.

[0020] This method utilizes the decorrelation rate of OCT speckle to visualize the therapeutic thermal field, demonstrating enhanced robustness to speckle and decorrelation noise, enabling characterization of therapeutic temperature changes with a high signal-to-noise ratio. The decorrelation results, based on successive time intervals, reduce sensitivity to time interval selection, enabling broader application in thermal ablation treatments with varying heating conditions (power, duty cycle, etc.). This method enables simultaneous, real-time display of OCT structural and thermal field images, enabling real-time, simultaneous monitoring of tissue structure and thermal field imaging during thermal treatment, contributing to the widespread adoption of OCT in minimally invasive surgery and thermal therapy.

[0021] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart illustrating a method for visualizing a therapeutic thermal field based on OCT according to an exemplary embodiment; Figure 2 is a schematic diagram illustrating the difference between decorrelation and decorrelation rate visualization of a therapeutic thermal field based on OCT speckle according to an exemplary embodiment; Figure 3 . A schematic diagram illustrating a complex differential variance algorithm using speckle decorrelation calculation in an OCT-based therapeutic thermal field visualization method according to an exemplary embodiment is shown as a control diagram. DETAILED DESCRIPTION

[0023] The following describes several preferred embodiments of the present invention with reference to the accompanying drawings to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0024] In the drawings, components with identical structures are denoted by the same reference numerals, and components with similar structures or functions are denoted by similar reference numerals. The size and thickness of each component shown in the drawings are arbitrary and are not limited by the present invention. To enhance clarity, the thickness of components in some places in the drawings is schematically exaggerated.

[0025] The inventors designed a method for visualizing the therapeutic thermal field based on OCT, such as Figure 1 As shown, the following steps are included: S100, original signal acquisition: during the tissue thermal treatment process, an AD acquisition card is used to continuously acquire the interference spectrum signal of the heating treatment thermal field area at the same position and save it.

[0026] S200, preprocessing the original signal, performing signal preprocessing on the saved interference spectrum signal to obtain a preprocessed OCT complex domain signal, and saving the signal. The signal preprocessing includes: Spectral k-domain linear correction, uniform sampling in the wavenumber k-domain, using multiple spline interpolation calculations; DC term removal, baseline artifact removal, averaging the elements at the same depth in multiple continuously acquired interferometric spectral signals to obtain an average value, and then subtracting the average value from each element of the interferometric spectral signal; Dispersion compensation is performed to compensate for the optical material and optical path difference between the sample arm and reference arm of the OCT system. Dispersion compensation methods include hardware compensation or polynomial compensation. Hardware compensation involves adding dispersion compensation elements, including but not limited to compensation materials, gratings, or fiber stretchers, to the reference arm to match the dispersion characteristics of the sample arm. In this embodiment, polynomial compensation is used. Spectral window shaping: select a window function to multiply the interference spectrum signal to perform spectrum shaping to suppress sidelobe noise. The window function includes but is not limited to a Hanning window, a Hamming window, or a Gaussian window. In this embodiment, a Gaussian window is used as the window function. Fast Fourier transform maps the interference spectrum signal from the wavenumber k domain to the depth coordinate information of the sample.

[0027] S300, OCT signal processing, modulo and logarithmically transform the preprocessed OCT complex domain signal, map it to tissue light scattering B-scan OCT signal data, and normalize it to obtain and save the standardized B-scan OCT signal data. The B-scan OCT signal data is normalized to an 8-bit unsigned integer data type in the range of 0-255 using the following formula: in, and Respectively expressed asi Row, No. j Columns of normalized B-scan OCT signal data before and after and They are B-scan OCT signal data and their average values, respectively.

[0028] S400, speckle decorrelation value calculation, the pre-processed OCT complex domain signal is used to parallel and synchronously calculate the speckle decorrelation values ​​corresponding to different time intervals at the current moment, using the following formula: in, x and z are the lateral and axial positions in the B-scan section of OCT, A is the amplitude of the OCT speckle signal, It is the inverse of the OCT system frame rate, that is, the time interval between two adjacent B scans. and are the current moment and the time interval used for decorrelation calculation, yes A positive integer multiple of N is the number of B scans used for averaging, which helps improve the signal-to-noise ratio. n From 0 to N-1 An integer.

[0029] S500, speckle decorrelation rate calculation, speckle decorrelation values ​​corresponding to different time intervals are fitted by linear regression to calculate the speckle decorrelation rate, and the slope of the fitting line is used to obtain speckle decorrelation rate signal data; the speckle decorrelation rate is calculated using the following formula: in, M is the number of time points used to calculate the slope, express, m From 1 to M An integer.

[0030] S600: Processing the speckle decorrelation rate signal to remove outliers from the speckle decorrelation rate signal data. This includes re-assigning the speckle decorrelation rate signal data that is less than 0 to the center of the speckle decorrelation rate signal data, taking the average of the 3×3 neighborhood element values, and obtaining the speckle decorrelation rate signal data with the outliers removed, and saving the data.

[0031] S700, structural and thermal field image visualization, performs grayscale mapping operation on the standardized B-scan OCT signal data, outputs it in image form, and displays OCT structural image information in real time; performs pseudo-color mapping operation on the speckle decorrelation rate signal data after removing outliers, outputs it in image form, and displays OCT thermal field image information synchronized with the OCT structural image information in real time.

[0032] like Figure 2 As shown, the dots represent the decorrelation values ​​of the OCT speckle signal, the dashed line represents the linear regression fit of the decorrelation values, and the solid curve represents the decorrelation trend. When laser ablation is performed on tissue in real time, the decorrelation trend of the OCT speckle signal is such that the decorrelation value increases with increasing temperature. However, the decorrelation values ​​of the OCT speckle signal fluctuate and jump due to noise, resulting in inaccurate results when using the decorrelation values ​​to represent the thermal field. However, the decorrelation rate value derived by fitting the decorrelation values ​​corresponding to a time interval is less susceptible to noise and other errors than the decorrelation values ​​corresponding to a single time interval, making it more robust for characterizing the therapeutic thermal field.

[0033] To verify the effectiveness of the present invention, an experiment involving laser ablation of ex vivo biological tissue was conducted. An infrared thermal imager was used as a temperature reference, detecting infrared radiation emitted by an object and converting it into a visual image representing temperature changes. Infrared thermal imagers do not have tomographic imaging capabilities, so a special setup was designed to capture OCT B-scans and infrared thermal imager imaging on the same plane. First, the ex vivo tissue was processed so that it had a flat, vertical outer surface parallel to the imaging and treatment laser beams. The coaxial OCT imaging and treatment laser beams were then brought as close as possible to the aforementioned outer surface. Coupling was achieved using a dichroic mirror. Consequently, it was concluded that the information from the OCT B-scan results was consistent with that obtained from the infrared thermal imager.

[0034] The experiment employed a 532 nm continuous-wave diode laser for optical heating. The OCT spectral domain system used had a central wavelength of 840 nm and an A-line rate of 80 kHz. Imaging was performed using a 50 mm focal length objective lens and a two-axis galvanometer scanning mirror for flying-spot scanning. Because continuous B-scans were required at the same position to calculate decorrelation parameters, only the fast axis of the galvanometer scanning mirror was set to line scanning, while the slow axis was fixed. The scan length and corresponding data points were set to 5 mm and 1024, respectively. Set to 16.6 ms and N to 10.

[0035] First, a qualitative study of pig liver was conducted, such as Figure 3As shown, the complex differential variance (CDV) algorithm for speckle decorrelation calculation was used as a control. Time intervals M were selected at 3, 5, and 8 seconds, with the first and second columns showing OCT B-scans and infrared thermography images under continuous laser heating at different times. During the initial phase of the experiment, particularly within the first 5 seconds, the OCT B-scan images showed no substantial changes. Meanwhile, the infrared thermography images revealed a gradual increase in tissue temperature. As the experiment progressed to the 8th and 15th seconds, the tissue underwent thermal denaturation and coagulation due to the rising temperature, resulting in the formation of tiny structural protrusions in the central region of the laser focus.

[0036] In thermal field visualization based on speckle decorrelation, regions experiencing elevated temperature are indicated by increased signal intensity. This approach provides a visual advantage compared to structural OCT images and could potentially offer enhanced treatment monitoring capabilities. However, the signal-to-noise ratio (SNR) within these elevated temperature regions is limited compared to surrounding areas. Furthermore, the SNR within these elevated temperature regions was observed to decrease with decreasing time intervals over which decorrelation was calculated, demonstrating sensitivity to this time interval. In contrast, a method using the speckle decorrelation rate also identified elevated temperature regions with high signal intensity. This method significantly improved the SNR compared to surrounding regions. Furthermore, this method consistently achieved higher SNR representations of post-treatment thermal field regions across different time intervals. Furthermore, after tissue denaturation and coagulation, at 15 seconds, both speckle decorrelation and the speckle decorrelation rate exhibited a shift toward lower signal intensity. At this stage, these parameters no longer maintained a positive correlation with temperature changes. This observation highlights the importance of considering temporal dynamics in the analysis of tissue thermal effects.

[0037] Further quantitative studies were conducted, using pig skin and chicken breast to verify its effectiveness on different tissue types, and proved that it outperformed other speckle decorrelation-based algorithms (amplitude decorrelation AD and speckle variance SV). Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were used as quantitative indicators, defined as: in I max is the maximum pixel value of the entire image. µ d 、 µ s is the average value of the dynamic heating zone and the static zone, σ d σ s is the variance of the dynamic heating zone and the static zone. The results are summarized in Table 1.

[0038] Table 1. Quantitative comparison of different methods for observing the therapeutic thermal field in different tissue types. Unit: dB.

[0039] Consistent with the qualitative results above, the method based on speckle decorrelation rate can achieve higher SNR and CNR. This conclusion also applies to different types of tissues.

[0040] Furthermore, monitoring thermal ablation therapy requires fast visualization processing speed, so the processing speed of the method was further evaluated. Table 2 lists the average processing time for processing each B-scan moment.

[0041] Table 2. Average processing time per B-scan using different methods. Unit: milliseconds (ms) For comparison, a workstation equipped with an Intel Core i9-10920X CPU and an NVIDIA GeForce RTX4090 GPU was used. It can be seen that when using all CPUs, the speckle decorrelation rate method significantly increases the time cost due to the need for multiple decorrelation calculations and linear fitting. By utilizing the OCT architecture and real-time thermal imaging processing flow of the present invention, the average processing time per B-scan using a GPU can be reduced to approximately 10 ms, meeting real-time requirements.

[0042] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A method for visualizing therapeutic thermal fields based on OCT, characterized in that: The steps include: S100, original signal acquisition, during the tissue thermal treatment process, the interference spectrum signal of the heating treatment thermal field area is continuously acquired at the same position and saved; S200, preprocessing the original signal, performing signal preprocessing on the saved interference spectrum signal to obtain a preprocessed OCT complex domain signal, and saving the preprocessed signal; S300, OCT signal processing, performing modulus and logarithmic transformation on the preprocessed OCT complex domain signal, mapping it into B-scan OCT signal data of tissue light scattering and standardizing it, obtaining standardized B-scan OCT signal data, and saving it; S400, calculating speckle decorrelation values, calculating speckle decorrelation values ​​corresponding to different time intervals in parallel and synchronously for the preprocessed OCT complex domain signal at the current moment; S500, speckle decorrelation rate calculation, calculating the speckle decorrelation rate by performing linear regression fitting on the speckle decorrelation values ​​corresponding to the different time intervals, and obtaining speckle decorrelation rate signal data by the slope of the fitting line; S600, speckle decorrelation rate signal processing, removing outliers from the speckle decorrelation rate signal data to obtain speckle decorrelation rate signal data with outliers removed, and saving the data; S700, structural and thermal field image visualization, performing grayscale mapping operation on the standardized B-scan OCT signal data, outputting in the form of an image, and displaying OCT structural image information in real time; A pseudo-color mapping operation is performed on the speckle decorrelation rate signal data with outliers removed, and the data is output in the form of an image, so as to display the OCT thermal field image information synchronized with the OCT structural image information in real time.

2. The OCT-based therapeutic thermal field visualization method according to claim 1, characterized in that: The signal preprocessing comprises: Spectral k-domain linear correction, uniform sampling in the wavenumber k-domain, using multiple spline interpolation calculations; DC term removal, baseline artifact removal, averaging the elements at the same depth in a plurality of continuously collected interference spectrum signals to obtain an average value, and then subtracting the average value from each element of the interference spectrum signal; Dispersion compensation, which compensates for the optical material and optical path difference between the sample arm and the reference arm of the OCT system; Spectrum windowing and shaping, selecting a window function to multiply the interference spectrum signal to perform spectrum shaping to suppress the generation of sidelobe noise; Fast Fourier transform maps the interference spectrum signal from the wave number k domain to the depth coordinate information of the sample.

3. The OCT-based therapeutic thermal field visualization method according to claim 2, characterized in that: The dispersion compensation method includes hardware compensation or polynomial compensation.

4. The OCT-based therapeutic thermal field visualization method according to claim 3, characterized in that: The hardware compensation is to add a dispersion compensation element in the reference arm to match the dispersion characteristics of the sample arm of the OCT system.

5. The OCT-based therapeutic thermal field visualization method according to claim 2, characterized in that: The window function includes but is not limited to a Hanning window, a Hamming window or a Gaussian window.

6. The OCT-based therapeutic thermal field visualization method according to claim 1, characterized in that: The B-scan OCT signal data were normalized using the following formula: 。 7. The OCT-based therapeutic thermal field visualization method according to claim 6, characterized in that: The B-scan OCT signal data is normalized to an 8-bit unsigned integer data type with a range of 0-255.

8. The OCT-based therapeutic thermal field visualization method according to claim 1, characterized in that: The speckle decorrelation value uses the following formula: 。 9. The OCT-based therapeutic thermal field visualization method according to claim 1, characterized in that: The speckle decorrelation rate is calculated using the following formula: 。 10. The OCT-based therapeutic thermal field visualization method according to claim 1, characterized in that: The removing of outliers from the speckle decorrelation rate signal data comprises reassigning the speckle decorrelation rate signal data less than 0 to the average of the values ​​of 3×3 neighborhood elements centered on the speckle decorrelation rate signal data.

Images