WO2016180287A1 - Method and system for processing oct signal using general-purpose graphics processing unit - Google Patents

Abstract

A method for processing an OCT signal using a general-purpose graphics processing unit (GPGPU) in an OCT system. The method comprises four steps: (1) data collection; (2) data transmission; (3) data processing; and (4) transfer to an image display library. At the same time of data transmission, the GPGPU is used to perform parallel processing on OCT original data that is transmitted to a device memory a previous time, and the processed data is placed in a memory of the image display library. The method may improve efficiency of processing an FD-OCT digital signal, thereby implementing real-time processing and display of an image, lowering a requirement for and a cost of a digital signal processing device, and improving portability of software and reducing a development cost of the software; meanwhile, seamless combination with existing popular image display libraries are implemented, thereby improving flexibility of software display.

Description

Method and system for processing OCT signal by using general image processor Technical field

The invention relates to the technical field of medical instruments, in particular to an image processing method for optical coherence tomography scanning imaging, and provides a data which is simple, efficient, highly portable and compatible with popular image display libraries (such as OpenGL, DirectX). Processing method and system.

Background technique

Optical Coherence Tomography (OCT) has been widely used in the field of ophthalmology diagnosis. It is based on the science of optics, electronics and computer technology. It is a collection of optoelectronic and high-speed data acquisition and image processing. With a number of cutting-edge disciplines as a new imaging technology, OCT has attracted much attention due to its high-resolution, high-speed imaging and other aspects, and has begun to receive attention and application in the field of biomedical and clinical diagnostics.

Compared with other CT, ultrasound, MRI and other imaging methods, OCT has a very high resolution, compared with the traditional laser confocal microscope, OCT imaging depth has obvious advantages. Most of the core technologies of traditional optical probes use fiber bundles for light transmission and imaging, or CCD technology for imaging. Such endoscopic probes can only detect lesions on the surface of tissues. However, the symptoms of early cancer occur 1-3 mm below the epidermis. The depth of the traditional optical endoscopic probe is not enough. At present, there are endoscopic probes for medical imaging through the principle of ultrasound. Although the tissue information below the surface of the biological tissue can be obtained, the resolution is only on the order of millimeters, which may cause missed diagnosis for early cancer.

Endoscopic OCT technology is an OCT branch technology that was born and flourished with the development of OCT technology in the past decade. Its core goal is to miniaturize OCT optical imaging equipment without providing resolution, and provide internal organs. High resolution OCT image of the lumen. This technology greatly expands the application of OCT technology, making OCT inspection objects from body surface organs or biopsy samples to human internal organs such as blood vessels, digestive tract and respiratory tract. At present, in clinical aspects, OCT endoscopic techniques have been initially applied in the examination of atherosclerosis and the examination of vascular stent placement.

The traditional OCT system is a time domain OCT system (TD-OCT), and the scanning speed of the time domain OCT system generally cannot exceed 2000 lines per second, thereby limiting the imaging speed. In recent years, the second-generation frequency domain OCT (FD-OCT) has become more and more popular due to its high scanning speed and high detection sensitivity. Vision.

In FD-OCT, in order to obtain depth image information, multi-step signal processing of the original signal is required. Among them, digital resampling and Fourier transform are computationally intensive digital signal processing. These digital signal processing for the original signal of the FD-OCT is a typical single instruction multiple data stream (SIMD) processing method, and the processing efficiency is low with the general-purpose CPU, and real-time signal processing cannot be realized for the high-speed OCT system.

In order to perform real-time signal processing for high-speed OCT systems, it is usually necessary to use a dedicated digital signal processing chip to achieve this function. The disadvantages of using this method are: 1) high hardware cost and development cost; 2) low portability and 3 The processed signal still needs to be transmitted to the image display device via the bus, thereby reducing efficiency, so there is still a need for improvement.

Summary of the invention

In order to improve the efficiency of FD-OCT digital signal processing, real-time image processing and display, reduce the requirements and cost of digital signal processing equipment, and improve software portability and software development cost, the present invention provides a utilization The general image processor is a method for processing OCT signals in an OCT endoscopic scanning imaging system that is simple, efficient, highly portable, and compatible with popular image display libraries such as OpenGL and DirectX.

The present invention provides a method for processing an OCT signal in an OCT endoscopic scanning imaging system using a general purpose image processor (GPGPU), the method comprising: (1) data acquisition; (2) data transmission; (3) data processing; (4) Four steps to the image display library.

Wherein, (1) data collection, the invention obtains FD-OCT raw data through an external collection device;

(2) Data transmission, the FD-OCT raw data obtained in the data acquisition step is placed in the computer system or embedded system memory, and the data is stored in the system memory in frame units, when certain conditions are met (such as data accumulation) For one or more frames), the data can be transferred to the device memory of the general-purpose image processor via a data bus (such as PCI Express); because the bus transfer speed is relatively slow, the general-purpose image processor will be on the same while the data is being transferred. The OCT raw data transferred to the device memory at one time is processed in parallel. The method has high-efficiency parallel signal processing capability, can realize real-time digital signal processing, greatly improves transmission efficiency, and saves bus resources;

(3) Data processing, digital signal processing in a general-purpose image processor is divided into three steps: one-dimensional digital resampling, one-dimensional fast Fourier transform (FFT), and calculation of amplitude and normalization. One-dimensional digital re-sampling step achieves fast one-dimensional cubic interpolation by two linear texture search to improve re-collection Accuracy

(4) Passed to the image display library, the processed data is placed in the memory of the image display library, and the image display library can be directly called, no need to transmit through the bus, which greatly improves the transmission efficiency, saves the bus resources, and is efficient. Parallel signal processing capability for real-time digital signal processing, high portability, seamless integration with popular image display libraries, and improved flexibility of software display (eg, image processing via a general-purpose image processor) Post-processing) enables lower hardware and software development costs.

Another object of the present invention is to provide an OCT endoscopic scanning imaging system, including a frequency sweeping laser module, an interference module, a detector module, a data acquisition module, a data processing module, an image display module, an actuator, a balloon catheter, and an OCT. Microprobes and charge and discharge devices, and a method of processing OCT signals using a general purpose image processor. among them:

The frequency sweeping laser module comprises a high speed swept laser, a fiber isolator and a fiber coupler, and the optical signal output from the swept laser is isolated from the subsequent optical path to prevent the optical signal returned by the subsequent optical path from interfering with the normal operation of the laser; the interference module A fiber-optic Mach-Zehnder interferometer (MZI) or a fiber-optic Michelson interferometer structure can be used. The Mach-Zehnder interferometer structure is mainly composed of two fiber couplers, two fiber circulators and two fiber polarization controllers. The first fiber coupler generally uses an asymmetric fiber coupler to drive most of the laser. a microprobe that is output to the sample arm; a fiber optic circulator is placed in both the reference arm and the sample arm to collect optical signals reflected or scattered back from the two arms; the second fiber coupler can be coupled using a symmetric 2×2 fiber The device (ie, the split ratio is 50/50) to generate an optical interference signal and reduce the DC common mode signal. The fiber polarization controller is symmetrically placed in the reference arm and the sample arm to adjust the polarization state of the two arms to obtain the most Good optical interference signal. The Michelson interferometer structure consists of a symmetrical 2×2 fiber coupler, a fiber circulator, and two optical polarization controllers. The swept laser first passes through the fiber circulator and enters the fiber coupler, from the reference arm and sample. The optical signal reflected or scattered back by the arm generates an interference signal through the same fiber coupler. The fiber polarization controller is symmetrically placed in the reference arm and the sample arm to adjust the polarization state of the two arms for optimal optics. Interference signal. The Mach-Zehnder Interferometer (MZI) has the advantages of structural symmetry, simple dispersion management, and high detection sensitivity. The advantage of the Michelson interferometer is that it is simple in structure and does not introduce a positive mode dispersion (PMD). The commonality of both is that the optical path difference between the two arms determines the free spectral region where the optical clock occurs ( FSR), which also determines the maximum imaging depth of the OCT image; the detector module can use a high-speed balanced photodetector, mainly for The interference optical signal outputted by the interference module is converted into an electrical signal; the data acquisition module is a high-speed analog-digital acquisition card, which is mainly used for converting an analog electrical signal into a digital electrical signal, and providing the digital signal to the data processing module for digital signal processing. The data processing module is a chip with digital signal processing capability (such as CPU, GPGPU, DSP, FPGA, etc.), and is mainly used for processing original signals and converting into final image signals; the image display module is mainly used for The image signal is displayed and is responsible for the post-processing of the image and the measurement work; the actuator is composed of a fiber optic rotary connector, a motor and an electric translation stage, and the rotary motor in the actuator drives the OCT micro-probe to perform a rotational scan while the electric translation stage is driven to execute The mechanism moves in a certain direction, at which time the software reconstructs the acquired rotational scan data and the translation stage movement data, that is, generates a 3D image; the OCT micro probe is mainly used to enter the internal organs of the human body to transmit the swept laser and collect Optical signal from backscattering in biological tissue; the balloon catheter is used The internal organs of the human body are expanded to eliminate wrinkles and stabilize the OCT microprobe at the center of the balloon; the charge and discharge device is mainly used to expand the balloon catheter.

Preferably, the OCT microprobe comprises a spring tube capable of providing sufficient torque to keep a probe of a certain length synchronized at both ends when rotating, and the spring tube is interspersed with a single mode fiber that transmits an optical signal; a lens assembly that concentrates light propagating through an optical fiber at a predetermined working distance, the lens assembly including a glass rod and a self-focusing lens, and the working distance of the OCT micro-probe can be changed by changing the bonding distance between the glass rod and the single-mode optical fiber, It can improve the lateral resolution of the OCT probe; the bonding of the autofocus lens to the glass rod increases the clear aperture of the autofocus lens, thereby increasing the numerical aperture and lateral resolution of the OCT probe and optimizing the physical size of the probe. The OCT microprobe may also include a mirror, a support stainless steel tube, and a slotted stainless steel tube, the end faces of which are glued with optical glue.

Wherein, one end of the single-mode fiber has a fiber optic standard connector, and the connector can be connected to the optical fiber rotating end of the OCT system, the single-mode fiber is sleeved in the spring tube (covered with a PTFE film), and the spring tube can effectively protect the single The mode fiber reduces the resistance of the probe when rotating, and makes the overall scanning of the OCT micro-probe more smooth and smooth. The optical fiber standard connector has a supporting stainless steel tube, which supports the OCT micro-probe when scanning, so that the whole The probe rotates more smoothly when rotated. The other end of the single-mode optical fiber is a beveled surface, and is glued to one end surface of the glass rod which is also beveled. The inclination of the glued surface effectively reduces the interference of the reflected light on the signal light, and can change the bonding distance between the glass rod and the single-mode optical fiber. To change the working distance of the OCT microprobe to achieve the required working distance required. The other end of the glass rod is glued to the self-focusing lens at an angle of 0° and then encapsulated in a slotted stainless steel tube. By changing the length of the autofocus lens, the working distance of the OCT probe is changed to achieve the specified working distance. Get the best horizontal To the resolution, the numerical aperture of the OCT probe can be increased by increasing the length of the glass rod, thereby improving the lateral resolution, that is, the use of the glass rod not only increases the working distance of the micro probe, but also increases the numerical aperture of the micro probe. The increase in numerical aperture also leads to an increase in lateral resolution. At the same time, this design also greatly shortens the length of the autofocus lens, ensuring the cornering of the microprobe, so that the entire microprobe can be clamped through the endoscope. Enter the human esophagus directly with the catheter. The mirror can change the forward light into a lateral direction, and the mirror is a cylindrical mirror, which can change the astigmatism effect of the protective sleeve outside the OCT probe on the concentrated beam. The self-focusing lens is glued to the glass rod, wherein the surface of the self-focusing lens in contact with the air is coated with an anti-reflection film, which can reduce the reflection of light between the optical surfaces and increase the light transmission performance, thereby reducing the optical surface. The influence of reflected light on the signal light improves the sensitivity of the OCT micro-probe. The surface of the autofocus lens in contact with air can be processed into a 4°-8° slope. This design can further reduce the interference signal formed by light passing through this surface. . The angle of the glued slope of the single mode fiber to the glass rod is 4°-12°. The reflecting surface of the mirror is packaged in a stainless steel tube toward the slot of the stainless steel tube. In order to reduce the influence of the astigmatism of the light source through the cylindrical inner tube on the imaging, the mirror here can be based on the inner and outer diameters of the inner cylindrical tube and the inner tube. The refractive index of the material is designed to be a cylindrical mirror. When the mirror is added, when the light is incident on the mirror, not only the light path of the light is reflected but also changed, and the mirror is specially designed to converge the light. Thereby offsetting the influence of the astigmatism of the inner tube, correcting the shape of the spot, and achieving the purpose of improving the image quality.

Preferably, the balloon catheter comprises: a handle, one interface of the handle is a host interface, the other interface is a ventilation interface; a double lumen tube, the dual lumen tube can allow an OCT optical probe to pass; a balloon, the The front end of the balloon is sealed and has a scale on the balloon; the inner tube is fixed in length according to the length of the balloon, and its length is shorter than the balloon, the balloon and the soft head When welding, the balloon is pushed down a certain distance to be flush with the inner tube and fixed, and then welded. Since the inner tube is too thick, the sharpness of the scanned image is affected, and too thin may affect the rotation and concentricity of the probe. The inner tube is specifically designed for an OCT probe, the inner tube has an inner diameter of 1.4 mm, and the inner tube has an outer diameter of 1.65 mm. The concentricity of the inner tube and the balloon deviates from the rated atmospheric pressure by no more than 500 micrometers, preferably The rated atmospheric pressure is 3-5 atmospheres, and the concentricity of the inner tube and the balloon deviates from 500 micrometers at 3 atmospheres; the membrane sleeve is located in the balloon and the double lumen tube The joint controls the floating of the inner tube inside the lumen, thereby ensuring The center of the capsule does not deviate; the soft head is a solid structure, wherein the double lumen tube is connected to the handle at one end, and the other end is connected to the inner tube and the balloon, the balloon The other end of the inner tube is connected to the soft head.

Conventional balloon catheters require guidewire support and guidance. Guidewire diameters are typically 0.018 in, 0.035 in, 0.014 in, 0.038 in. The balloon catheter of the present invention can pass through a 0.055 in OCT optical microprobe. The balloon has an ink printing scale, the line width is ≤0.1mm, and the direction of the probe scanning can be discriminated, which does not affect the scanning judgment of the normal image and can also distinguish the scanning position on the display screen. The balloon front end seal Blocking can prevent the body fluid from affecting the optical scanning, prevent the corrosion of the body fluid to the precision equipment, and at the same time, the sealing material is a soft structure, which will not scratch the tissue and the cavity of the measured object during the operation, and enhance the equipment. safety. The soft head has a solid structure and can prevent body fluid from entering. Preferably, the nominal atmospheric pressure is 3-5 atmospheres. Preferably, the nominal atmospheric pressure is 3 atmospheres, and the concentricity of the inner tube and the balloon deviates by no more than 500 microns at 3 atmospheres. The balloon is rated at 3 atmospheres and does not cause damage to the normal esophagus at lower pressures. At the same time, the heat setting process and welding process of the balloon ensure the concentricity of the inner tube and the balloon at a rated pressure of 3 atmospheres. Deviation does not exceed 500 microns for optical imaging. The soft head has a solid structure and can prevent body fluid from entering. The double lumen tube is connected to the handle by a UV adhesive, and the other components are connected by a welding process. The length of the inner tube is fixed according to the length of the balloon, and the length thereof is shorter than the balloon. When the balloon is welded to the soft head, the balloon is pushed down by a certain distance to make it The inner tube is flush and fixed and welded, so that the balloon has an elongation margin when filling, thereby matching the inner tube to stretch and maintain concentricity. The method of welding and fixing the balloon can prevent the inner tube from being directly inserted into the balloon without being fixed. Due to the movement of the patient and the organ cavity and the excessive insertion of the catheter into the balloon, the relative movement of the catheter and the balloon is caused. The problem of optical probe eccentricity. Preferably, the balloon has a folding curl temperature of 40° to 45° and a setting time of 4 to 5 hours. Compared with the conventional balloon folding process, the process can maintain the concentricity while maintaining the memory of the balloon. characteristic. In addition, in the present invention, the handle material may be polycarbonate, the double lumen tube and the soft head material may be block polyether amide, and the balloon and inner tube material may be nylon and its modified polymer.

The method of using the balloon catheter for OCT endoscopic scanning imaging includes: inserting a balloon catheter with an OCT microprobe into the caliper and pushing the balloon to the desired scanning position; then placing the handle with an interface and OCT The device is connected, and the other interface is connected to the inflating device for inflation. After the inflation is completed, the image is scanned, the inhalation is performed after the scanning is finished, and finally the balloon catheter is withdrawn after the end of inhalation.

Preferably, the automatic charging and discharging device is applied to the OCT endoscopic scanning imaging system, that is, includes a frequency sweeping laser module, an interference module, a detector module, a data acquisition module, a data processing module, an image display module, an actuator, and a balloon. a catheter, an OCT microprobe, and a charge and discharge device, the charge and discharge device being an automatic charge and discharge device, the automatic charge and discharge device comprising: a control and display module, a gas pump, and an inflation device Solenoid valve, bleed solenoid valve, pressure sensor, explosion-proof pressure sensor, mechanical pressure switch. The automatic charging and deflation device is applied to an OCT endoscopic scanning imaging system to realize automatic charging and discharging and precise air pressure control, wherein the air pump is connected to the throttle valve through a gas filling solenoid valve and a bleed solenoid valve, the throttle valve Connected to the foregoing balloon, the pressure sensor is plural, wherein at least one pressure sensor is connected to the balloon, and at least one pressure sensor is disposed between the throttle valve and the inflation solenoid valve and the deflation solenoid valve, the explosion-proof pressure sensor and the ball The capsule is connected, the mechanical pressure switch is connected to the balloon, and the control and display module can set the balloon air pressure and inflation time, collect pressure, control the air pump to start and stop, and control the working state of the solenoid valve;

The use of the automatic charge and discharge device includes:

During the inflation process, the balloon pressure and inflation time are set in the control and display module, and the air pump is inflated until the set air pressure value;

During the deflation process, the balloon pressure and deflation time are set in the control and display module, and the air pump inhales until the set pressure value;

During the charging and discharging process, the system is over-pressure protected by an explosion-proof pressure sensor, the air pump is turned off when the balloon exceeds the set air pressure, and the pressure is released by the mechanical pressure switch when the balloon exceeds the set air pressure. The automatic charging and discharging device realizes automatic inflation and inhalation, and has the function of setting different air pressure parameters, and can charge and deflate the balloon of different specifications, and the device reaches the set air pressure during inflation of the balloon. After the value, the inflation is stopped and the overvoltage protection function is provided. The achievable effect is: firstly, the doctor's manual filling and deflation operation of the balloon is eliminated, the time for the doctor to charge and deflate is shortened, the safety is improved, and the risk of overfilling and exploding the balloon is avoided; secondly, Accurate air pressure control ensures that the shape consistency of the balloon after inflation is ensured. Since optical imaging is sensitive to the shape of the scanned object supported by the balloon, the repeatability of multiple scans of the same scanned object is better. The doctor can compare the scanned image data; again, in the emergency case, the doctor can perform other operations while automatically deflated.

Preferably, the OCT endoscopic scanning imaging system comprises an optical clock module composed of the interference module, the detector module and an optical clock conversion circuit module, wherein the interferometer module can adopt all-fiber Mach-Zehnder interference The instrument (MZI) structure is mainly composed of two fiber couplers, wherein the second coupler is a symmetric 2×2 fiber coupler, which is first divided into two paths of light at the first fiber coupler, and the two paths respectively pass through Two sections of the first optical fiber and the second optical fiber having a fixed optical path difference interfere with each other at the second optical fiber coupler. The detector module can employ a high-speed balanced photodetector, which is mainly used to convert an interference optical signal output from the interference module into an electrical signal. The optical interference signal generated from the MZI is converted into an electrical signal by a balanced photodetector, and then passed through the optical clock conversion circuit module, that is, sequentially passed through the broadband The 90 degree phase shifter, the zero crossing comparator, the exclusive OR gate, the OR gate, and the optical clock signal output module are converted into optical clock signals that are uniform in the frequency domain and variable in frequency domain. Among them, the wide-band 90-degree phase shifter is mainly used to shift the phase of the MZI electrical signal by 90 degrees, which can increase the available spectral interval width of the original signal, enrich the spectrum distribution resources of the sampling clock signal, and optimize the acquired sampling clock signal. The zero-crossing comparator is mainly used for zero-crossing comparison between the original MZI electrical signal and the phase-shifted MZI electrical signal to be converted into a digital signal, and the zero point of the MZI signal is evenly distributed in the frequency domain, so the number generated after the zero-cross comparison is generated. The rising or falling edge of the signal is also evenly distributed in the frequency domain. The XOR gate is mainly used to combine two digital clock signals to obtain two clock signals in one free spectral region (FSR). Increasing the FSR increases the maximum imaging depth of the OCT and reduces the jitter generated by the optical signal. Moreover, since the swept laser always has some idle time between two adjacent scans, the optical clock signal needs to fill in some false clock signals in the blank through an OR gate to ensure that the high speed analog capture card can work normally. The OR gate implements the function of combining the real optical clock signal with the fake clock signal. The optical clock signal output module is mainly used to deliver the combined real optical clock signal and the fake clock signal to the data acquisition module. By using the optical clock module in the OCT endoscopic scanning imaging system, the requirements for the data acquisition and processing system can be reduced, the redundant information collection can be reduced, the burden on the storage system can be reduced, and the integration degree of the entire OCT system can be improved. In turn, the system cost is reduced, and the signal-to-noise ratio of the image signal can be improved, and the attenuation of the detection sensitivity can be reduced, thereby improving the sharpness of the image.

DRAWINGS

1 is a schematic diagram of a FD-OCT signal processing step of the present invention;

2 is a schematic diagram of parallel generation of data transmission and signal processing of the GPGPU of the present invention;

3 is a schematic diagram of an OCT endoscopic scanning imaging system for processing an OCT signal using a general purpose image processor of the present invention;

Figure 4 is a diagram showing the physical components of the OCT micro-probe of the present invention;

5a and 5b are enlarged cross-sectional views of key parts of the OCT microprobe of the present invention;

Figure 6 is a schematic view showing the structure of a balloon catheter of the present invention;

Figure 7 is a balloon with a film sleeve

Figure 8 is the imaging effect of the inner tube and balloon eccentricity of about 500 microns

Figure 9 is the imaging effect of the inner tube and the balloon eccentricity is less than 500 microns.

Figure 10 is the imaging effect of the inner tube and the balloon eccentricity of more than 500 microns

Figure 11 shows the concentricity scan results of a product on the market.

Figure 12 is a schematic structural view of the automatic charging and discharging device of the present invention;

Figure 13 is a flow chart showing the operation of the automatic charging and discharging device of the present invention;

Figure 14 is a schematic view of an optical clock module of the present invention;

Figure 15 is a schematic diagram showing the process of generating an optical clock signal of the present invention;

16 is a schematic diagram of an OCT endoscopic scanning imaging system with an optical clock module of the present invention;

Figure 17 is a view showing the overall construction of the present invention;

Figure 18 is a graph showing the relationship between the working distance, the bonding distance and the lateral resolution curve of the OCT microprobe of the present invention;

Figure 19 is a view showing the esophagus of a healthy animal of the present invention;

Figure 20 is a partial enlarged view of the eczema scan of a healthy animal of the present invention;

Figure 21 is a 3D image of a healthy animal's esophagus of the present invention.

Description of the reference numerals

1, single mode fiber, 2, spring tube, 3, glass rod, 4, self-focusing lens, 5, mirror, 6, slotted stainless steel tube, 7, support stainless steel tube, 8, soft head, 9, inner tube, 10, balloon, 11, double lumen, 12, handle, 13, ventilation interface, 14, host interface, 15, membrane sleeve,

81, squamous epithelial layer (SE), 82, lamina propria (LP), 83, muscle mucosa (MM), 84, submucosa (SM), 85, intrinsic base layer (MP),

91. MZI electrical signal with 90 degree phase shift after wide frequency 90 degree phase shifter, 92, MZI electrical signal without phase shift, 93, digital signal of signal 91 undergoing zero crossing, 94, signal 92 Zero-compared digital signal, 95, false clock signal, 96, digital signal 93 and 94 are combined by XOR gate,

101, air pump, 102, pneumatic solenoid valve, 103, pressure sensor, 104, throttle valve, 105, pressure sensor, 106, explosion-proof pressure sensor, 107, deflation solenoid valve, 108, mechanical pressure switch.

detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The present invention will be further described below in conjunction with the drawings and embodiments, but the present invention is by no means limited to the embodiments described below.

Example 1

As shown in FIG. 1, the scheme for processing an OCT signal by using a general-purpose image processor according to the present invention includes (1) data acquisition, (2) data transmission, (3) data processing, and (4) delivery to an image display. The library is a few steps.

In the data transmission, because the bus transmission speed is relatively slow, while the data is being transmitted, the general image processor performs the parallel processing of the OCT raw data transmitted to the device memory last time, and the parallel transmission and processing process is as shown in FIG. 2 . The data processing process is divided into three steps: one-dimensional digital resampling, one-dimensional fast Fourier transform (FFT), and calculation of amplitude and normalization, in which one-dimensional digital resampling step is achieved by two linear texture lookups. One-dimensional cubic interpolation at a time to improve the accuracy of resampling.

As shown in FIG. 1, a scheme for processing an OCT signal by using a general-purpose image processor includes (1) data acquisition, (2) data transmission, (3) data processing, and (4) delivery to an image display library. step.

Wherein, (1) data collection, the invention obtains FD-OCT raw data through an external collection device;

(2) Data transmission, the FD-OCT raw data obtained in the data acquisition step is placed in the computer system or embedded system memory, and the data is stored in the system memory in frame units, when certain conditions are met (such as data accumulation) One or more frames), this data can be transferred to the device memory of the General Purpose Image Processor (GPGPU) via a data bus (such as PCI Express); due to the relatively slow bus transfer speed, general image processing while transmitting data The OCT raw data that was last transferred to the device memory is processed in parallel.

For example, as shown in FIG. 2, when the original data of the nth frame is transmitted to the memory of the general image processor device, the original data of the n-1th frame is simultaneously digitally processed in the general image processor, and After the data transmission and processing ends, the frame synchronization is performed, that is, whether the data transmission or the data processing completion party will wait for the completion of the completion of the next frame operation, the parallel signal transmission/processing model can be effectively used. Improve the data processing speed of the general image processor.

(3) Data processing, as shown in Figure 1, digital signal processing in a general-purpose image processor is divided into three steps: one-dimensional digital resampling, one-dimensional fast Fourier transform (FFT), and calculation of amplitude and normalization. . Digital resampling can be realized by the texture search function built in the image processor (GPU). The texture search function built in the image processor can automatically realize two-dimensional linear interpolation for interpolation, and the image processor's texture search module has special interpolation. Hardware optimization, faster interpolation than general general image processor, especially for non-equidistant interpolation in OCT signal processing; by precisely setting the search point in one dimension, it can be realized by the texture search function built in the image processor. One-dimensional line interpolation. On this basis, fast one-dimensional cubic interpolation is realized by two linear texture search, so as to improve the accuracy of resampling. Because the texture search module specially optimizes non-equidistant interpolation, this method is more than direct cubic interpolation. The general-purpose image processor is implemented with less computation and higher computational efficiency; the FFT can be implemented by a common commercial image processor-based numerical calculation library (such as nVidia's cuFFT library or OpenCL FFT library); the amplitude is calculated and normalized. You can write your own image processor program. For example, you can use the CUDA library provided by nVidia to write the corresponding kernel function to achieve fast traversal of 2D data to achieve amplitude and normalization calculation.

(4) Passed to the image display library, the processed data is placed in the memory of the image display library, and the image display library can be directly called, no need to transmit through the bus, which greatly improves the transmission efficiency, saves the bus resources, and is efficient. Parallel signal processing capability for real-time digital signal processing, high portability, seamless integration with popular image display libraries, and improved flexibility of software display (eg, image processing via a general-purpose image processor) Post-processing) enables lower hardware and software development costs.

Example 2

As shown in FIG. 3, an OCT endoscopic scanning imaging system includes a frequency sweeping laser module, an interference module, a detector module, a data acquisition module, a data processing module, an image display module, an actuator, a balloon catheter, and an OCT microprobe. And a charge and discharge device, wherein: the method for processing the OCT signal in the OCT endoscopic scanning imaging system is the method described in Embodiment 1, wherein:

The frequency sweeping laser module comprises a high speed swept laser, a fiber isolator and a fiber coupler, and the optical signal output from the swept laser is isolated from the subsequent optical path to prevent the optical signal returned by the subsequent optical path from interfering with the normal operation of the laser; the interference module A fiber-optic Mach-Zehnder interferometer (MZI) or a fiber-optic Michelson interferometer structure can be used. The Mach-Zehnder interferometer structure is mainly composed of two fiber couplers, two fiber circulators and two fiber polarization controllers. The first fiber coupler generally uses an asymmetric fiber coupler to drive most of the laser. a microprobe that is output to the sample arm; a fiber optic circulator is placed in both the reference arm and the sample arm to collect optical signals reflected or scattered back from the two arms; the second fiber coupler can be coupled using a symmetric 2×2 fiber The device (ie, the split ratio is 50/50) to generate an optical interference signal and reduce the DC common mode signal. The fiber polarization controller is symmetrically placed in the reference arm and the sample arm to adjust the polarization state of the two arms to obtain the most Good optical interference signal. The Michelson interferometer structure consists of a symmetric 2×2 fiber coupler, a fiber circulator and two optical polarization controllers. The swept laser first passes through the fiber circulator and then enters the fiber coupling. The optical signal reflected or scattered back from the reference arm and the sample arm generates an interference signal through the same fiber coupler, and the fiber polarization controller is symmetrically placed in the reference arm and the sample arm for adjusting the polarization of the two arms. State to obtain the best optical interference signal. The Mach-Zehnder Interferometer (MZI) has the advantages of structural symmetry, simple dispersion management, and high detection sensitivity. The advantage of the Michelson interferometer is that it is simple in structure and does not introduce a positive mode dispersion (PMD). The commonality of both is that the optical path difference between the two arms determines the free spectral region where the optical clock occurs ( FSR), which also determines the maximum imaging depth of the OCT image; the detector module can use a high-speed balanced photodetector, which is mainly used to convert the interference optical signal output from the interference module into an electrical signal; the data acquisition module is a high-speed mode The data acquisition card is mainly used for converting an analog electrical signal into a digital electrical signal, and providing the digital signal to a data processing module for digital signal processing; the data processing module is a chip with digital signal processing capability (such as CPU, GPGPU, DSP, FPGA, etc.) are mainly used to process the original signal and convert it into a final image signal; the image display module is mainly used for displaying an image signal and is responsible for post-processing and measurement work of the image; the actuator is rotated by the optical fiber The connector, the motor and the electric translation stage are mainly used to drive the mechanical spiral scan of the OCT micro probe to obtain the OCT. The OCT microprobe is mainly used to enter internal organs of the human body to transmit the swept laser and collect optical signals backscattered from the biological tissue; the balloon catheter is used to expand the internal organs of the human body to eliminate wrinkles. The OCT microprobe is stabilized at the center of the balloon; the charge and deflation device is primarily used to expand the balloon catheter.

Example 3

An OCT endoscopic scanning imaging system similar to that of Embodiment 2, except that the OCT microprobe is as shown in FIG. 4 and FIGS. 5a and 5b: the spring tube 2 is covered with a single mode fiber 1, which is a spring tube 2 can provide enough torque, so that a certain length of the probe keeps the near and far ends synchronous when rotating, the spring tube 2 can effectively protect the fragile optical fiber while reducing the resistance when the probe rotates; the glass rod 3 end and the self-focusing lens 4 The zero-degree angle is glued, and the other end is obliquely combined with the single-mode fiber 1. The working distance of the OCT probe can be changed by changing the bonding distance between the two end faces of the glass rod 3 and the single-mode fiber 1 to achieve the expected working distance, thereby improving The numerical aperture and lateral resolution of the OCT probe; the mirror 5 is a cylindrical mirror and is enclosed in a slotted stainless steel tube 6, and the reflecting surface of the mirror 5 faces the slot of the slotted stainless steel tube 6, which can reduce the light source through the cylinder The effect of astigmatism of the inner tube on imaging.

4 and 5a and 5b, an OCT microprobe of an OCT endoscopic scanning imaging system, comprising a single mode fiber 1, a spring tube 2, a glass rod 3, a self-focusing lens 4, a mirror 5, a slotted stainless steel tube 6 And supporting the stainless steel tube 7, the end faces of these optical elements are glued with optical glue. Specifically, the mirror 5 is mounted Put into the slotted stainless steel tube 6, then place the A/B glue on the tooling, then point the UV glue under the microscope, assemble the glass rod 3 and the self-focusing lens 4; insert the fiber 1 into the spring tube 2, and then The assembled glass rod 3 and the self-focusing lens 4, the optical fiber 1 and the spring tube 2 are assembled using a point UV glue, and finally the assembled spring tube assembly is loaded into the slotted stainless steel tube 6, and the A/B glue is used at the edge gap. Fill up.

The single-mode fiber 1 is externally sheathed with a stainless steel spring tube 2 (coated with a PTFE membrane), which effectively protects the fragile fiber while reducing the resistance of the probe when rotating, making the micro-probe scan smoother and smoother. The main function of the supporting stainless steel tube 7 is to support the OCT probe during scanning, so that the entire probe is rotated more smoothly, and the notch of the slotted stainless steel tube 6 allows the beam to be irradiated onto the sample to be tested through the slot.

The antireflection film can reduce the reflection of light between the optical surfaces and increase the light transmission performance, thereby reducing the influence of the reflected light of the optical surface on the signal light. Therefore, in this embodiment, the length of the self-focusing lens can be changed. The working distance of the OCT probe is to achieve the best lateral resolution under the specified working distance. The surface of the autofocus lens 4 in contact with the air is coated with an anti-reflection film. In addition, the surface of the autofocus lens in contact with the air can be processed into 4 °-8° bevel, this design can further reduce the interference signal formed by light passing through this surface. The difference between Fig. 5a and Fig. 5b is: whether the self-focusing lens 4 has a tilt angle design, and Figure 5b has an exit surface. Tilted 4°-8° design, this design minimizes the unwanted light signal reflected back from this surface and improves the imaging quality of the OCT probe. At the same time, the glass rod 3 is glued to the zero-degree end surface of the self-focusing lens 4, which improves the sensitivity and resolution of the micro-probe. The other end of the glass rod 3 has a certain inclination angle with the bonding surface of the single-mode optical fiber 1. In this embodiment, the inclination angle of the bonding surface can be 8°, and the inclination of the bonding surface effectively reduces the reflected light to the signal light. Interference, the design of the glass rod 3, increases the numerical aperture of the OCT probe, thereby increasing the lateral resolution, and can change the working distance of the OCT probe by changing the bonding distance of the two end faces to achieve the desired working distance. Since the mirror 5 is mounted at an angle of 45°, the incident light and the reflected light are perpendicular to each other to cause light interference. In the embodiment, the front end of the micro probe is mounted with a 40° angle mirror 5, and the mirror 5 is encapsulated in the slotted stainless steel tube 6 and reflected. Faced toward the notch of the slotted stainless steel tube 6, and in order to reduce the influence of the astigmatism of the light source through the cylindrical inner tube on the imaging, the mirror here in this embodiment is based on the inner and outer diameters of the inner cylindrical tube and the inner tube material. The refractive index of the cylindrical mirror is designed, the mirror can change the forward light into a lateral direction, and the mirror is a cylindrical mirror, which can change the protective sleeve of the OCT probe to the concentrated beam. Astigmatic effects.

Example 4

An OCT endoscopic scanning imaging system similar to Embodiment 2 or 3, except that the balloon catheter, as shown in FIG. 6, includes a handle 12, one interface of the handle 12 is a host interface 14, and the other interface is The ventilating interface 13; the double lumen tube 11 can allow the OCT optical probe to pass; the balloon 10, the front end of the balloon 10 is blocked and the balloon has a scale, and the front end sealing is a solid welded structure, and the blocking is sleeved Inside the tube, when the device moves inside the cavity, the cavity rubs against the inner tube and the balloon, and does not contact the plugging soft head. Even if it contacts the soft head, the frictional force tends to press the soft head into the inner tube. The inner tube 9, the concentricity of the inner tube 9 and the balloon 10 deviates by no more than 500 micrometers at 3 atmospheres; the membrane sleeve 15, as shown in Fig. 7, the membrane sleeve is located at the junction of the balloon and the double lumen tube, Controlling the floating of the inner tube inside the lumen to ensure no deviation from the center of the balloon; the soft head 8 is a solid structure, wherein one end of the double lumen 11 is connected to the handle 12, and the other end is connected to the inner tube 9 and the balloon 10 connected, the balloon 10 and the other end of the inner tube 9 are connected with the soft head 8, and the double lumen tube 11 passes the UV glue with the hand 12 is connected to each of the other parts are connected by welding.

The material of the handle 12 is polycarbonate, the material of the double lumen tube 11 and the soft head 8 is block polyether amide, and the material of the balloon 10 and the inner tube 9 is nylon and its modified polymer.

The balloon 10 has an ink printing scale with a line width of ≤0.1 mm, which can distinguish the direction of the probe scanning, and does not affect the scanning judgment of the normal image, but also can distinguish the scanning position on the display screen. The double lumen tube 11 is printed with a scale so that the doctor can judge the position of the scan.

Conventional balloon catheters require guidewire support and guidance. The diameter of the guidewire is generally 0.018 in, 0.035 in, 0.014 in, 0.038 in. In this embodiment, the double lumen tube 11 for the OCT balloon catheter can pass the 0.055 inOCT optical probe. . For optical imaging, the probe beam from the probe needs to pass through the inner tube wall to reach the object to be measured. In addition, the OCT probe needs to be rotated inside the inner tube to detect the object under test. In this process, if the inner tube wall is too thick, the detection light will cause attenuation when passing through the inner tube wall, resulting in loss of detection light energy, resulting in unclear imaging; if the inner tube wall is too thin, due to the OCT probe High-speed rotation inside the inner tube, too thin tube wall will cause the probe to be stuck, the inner tube is deformed, and the imaging is uneven. Therefore, the inner diameter and the outer diameter of the inner tube are limited in this application, thereby determining the thickness of the inner tube. In the present embodiment, the inner tube 9 has an inner diameter of 1.4 mm and an outer diameter of 1.65 mm, and is specifically designed for the OCT probe. This inner tube thickness is most suitable for OCT probe imaging, which can ensure that the light energy attenuation is small under this thickness, and the inner tube strength is ensured, and the imaging unevenness and unclearness caused by the deformation of the probe and the inner tube are not caused. The problem.

The length of the inner tube is fixed according to the length of the balloon, and the length thereof is shorter than the balloon. When the balloon is welded to the soft head, the balloon is pushed down by a certain distance to make it Inner tube is flush and fixed after welding The connection allows the balloon to have an elongation margin when filling, thereby matching the inner tube's stretch and maintaining concentricity.

The addition of the membrane sleeve controls the floating of the inner tube inside the lumen, thereby ensuring that there is no deviation from the center of the balloon, as shown in FIG.

In optical imaging, the balloon needs to be opened to open the cavity of the object to be measured, so that the image is completely formed. If the air pressure is low, the surface of the balloon will form wrinkles, which will inevitably affect the imaging effect, but if the air pressure is too high, it may be Damage and damage to the object to be tested, therefore, in order to ensure that the device does not cause damage and damage to the object under test, while ensuring that the balloon can be fully filled without wrinkles, the applicant preferably uses a rated atmospheric pressure of 3 atmospheres. The concentricity of the inner tube with the balloon deviates by no more than 500 microns at 3 atmospheres. In addition, this indicator can also limit the length of the balloon and the length of the inner tube. The process of the balloon of the present application ensures that the concentricity of the inner tube from the balloon does not exceed 500 microns at 3 atmospheres for optical imaging.

If the concentricity of the inner tube and the balloon deviates from 500 micrometers, the imaging effect is good. If the inner tube and the balloon deviate more than 500 micrometers, the image displayed is incomplete. Figure 8 shows a standard circle with an eccentricity of about 500 microns. Figures 9 and 10 show images scanned when the inner tube and the balloon are less than 500 microns eccentric and the eccentricity is more than 500 microns. From the above two figures, the eccentricity of less than 500 microns can meet the full imaging requirements. In contrast, the general product on the market does not control this indicator, and can not meet the requirement that the concentricity of the inner tube and the balloon deviate from 500 micrometers at 3 atmospheres. We select a product on the market as a comparison, and its imaging The effect is shown in Fig. 11, in which the upper right corner area is incompletely imaged and cannot meet the actual scanning requirements.

Example 5

An OCT endoscopic scanning imaging system is similar to Embodiment 2, 3 or 4, except that the charging and discharging device is an automatic charging and discharging device, as shown in FIG. 12, the automatic charging and discharging device includes The power supply part, the control and display part, and the air pump and its control system part, the air pump and its control system part include: air pump, inflation solenoid valve, bleed solenoid valve, throttle valve, pressure sensor, explosion-proof pressure sensor, mechanical pressure switch.

The pressure sensor of the present application is plural, wherein at least one pressure sensor is connected to the balloon, and at least one pressure sensor is disposed between the throttle valve and the inflation solenoid valve and the deflation solenoid valve, and the pressure sensor functions to charge and discharge The pipeline air pressure and balloon pressure are monitored in real time during the gas process. The pressure sensor is composed of two pressure sensors 103 and 105, which cooperate with the plurality of solenoid valve assemblies in the present application to form a pressure detecting system for charging and discharging the gas process, and the specific structure is: between the air pump 101 and the pressure sensor 103 A pneumatic solenoid valve 102 and a deflation solenoid valve 107 are provided. The pressure sensors 103, 105 are separated by a throttle valve 104, and the pressure sensor 105 is coupled to the balloon. Through this structure setting, the following three functions can be achieved:

(1) Two pressure sensors 103 and 105 are arranged, and after being separated by the throttle valve 104, the pressure sensor 105 is directly connected to the balloon, and the acquired data is more accurate, and for the pressure sensor 103, the pipeline can be monitored. The air pressure is proofed with 105 air pressure, and in the process of charging and discharging gas, after the inflation solenoid valve 102 and the bleed air valve 107 are opened, before the throttle valve 104 is opened, the air pressure of the air pump can be monitored for early warning. The role of the pressure, such as pressure problems can be automatically controlled to close the solenoid valve in time, to prevent excessive pressure on the balloon and even the patient's body cavity caused unnecessary damage, greatly enhancing the safety of the system.

(2) The pressure sensors 103 and 104 are connected to the air pump through the electromagnetic valve, and the electromagnetic pump isolates the air pump and the external air pressure to interfere with the pressure monitoring.

(3) Double monitoring of the balloon and the official air pressure, even if the single sensor failure system can still work normally, enhance system stability.

The pressure monitoring system of the present application further includes an explosion-proof pressure sensor connected to the balloon. The function of the explosion-proof pressure sensor is to monitor the air pressure in real time during the charging and discharging process, and the monitoring data is independent of the pressure sensor when the monitoring data is Exceeding the preset upper limit of the air pressure will activate the pressure relief mechanism. If the monitoring data remains unchanged for a long time, it will prevent the air pump from continuing to pressurize. The pressure monitoring system of the present application is divided into a pressure sensor and an explosion-proof pressure sensor, which realizes multiple monitoring of the air pressure by setting a plurality of sensors, and enhances the stability of the system; in addition, different sensors are disposed at different positions. The proofreading can also be performed between the sensors. The electromagnetic valve is also isolated between the sensors to avoid the interference of the charging and discharging device and the pressure relief device on the pressure measurement, and the accuracy of the pressure measurement is improved. Finally, the plurality of pressure sensors cooperate with the solenoid valve stroke pressure. With the support of the automatic control function, the monitoring system can prevent excessive charging and discharging air pressure from causing unnecessary damage to the balloon or even the patient's body cavity, and enhance the safety of the system.

As shown in FIG. 13, the use process of the automatic charging and discharging device includes:

1) Inflation process: Firstly, the user sets the parameters of the balloon 10 air pressure, inflation time and the like, and issues an inflation command. The control system reads the data of the pressure sensors 103 and 105. If it is less than the air pressure set by the user, the air pump 101 is activated to open the inflation. The solenoid valve 102 and the throttle valve 104 read the feedback values of the pressure sensors 103 and 105 in real time during the inflation process until the set air pressure value is reached, the air pump 101 is turned off, and the throttle valve 104 and the gas filling solenoid valve 102 are closed;

2) Deflating process: Firstly, the user sets parameters such as the balloon 10 air pressure and deflation time, and the control system reads the data of the pressure sensor 105. If it is greater than the air pressure set by the user, the air pump 101 is activated, and the deflation solenoid valve 107 is opened. The throttle valve 104 reads the feedback value of the pressure sensor 105 in real time during the deflation until the set air pressure value is reached, the gas 101 pump is turned off, and the throttle valve 104 and the deflation solenoid valve 107 are closed.

During the charging and discharging process, the system monitors the feedback value of the explosion-proof pressure sensor 106 in real time. If the user's air pressure upper limit setting value is exceeded, the software program protection is immediately performed, the air pump 101 is turned off, and the alarm is given; the mechanical pressure switch 108 is hardware protected. If the set value is exceeded, the switch is opened and the pressure is released for protection.

The pressure relief protection mechanism is realized by the automatic control function to guide the entire charging and discharging device system. It monitors the trigger through the explosion-proof pressure sensor and is implemented in three ways: First, software protection, that is, when the explosion-proof pressure sensor detects the pressure overrun The air pump charging function will be turned off, and the pneumatic solenoid valve will be closed at the same time. If necessary, the throttle valve and the deflation solenoid valve will be opened by the air pump to actively pump the air pressure. During this process, the pressure sensor will cooperate with the pressure monitoring, when the pressure is normal. After that, the solenoid valve and the air pump will be closed to protect the pressure relief, so as to prevent danger caused by excessive pressure relief. Second, the hardware protection, after the explosion pressure switch detects the pressure limit, the mechanical switch will automatically open for pressure relief protection; Third, the pressure relief valve protection, when the equipment can not operate normally, you can also manually depressurize by opening the switch on the pressure relief valve. The application combined with the explosion-proof pressure sensor provides software protection, hardware protection and pressure relief valve protection triple protection path for the system over-voltage risk, and achieves the effect of strengthening the safety and stability of the system and improving the accuracy of the pressure relief system.

In addition, the automatic charging and discharging device of the present application has the following advantages:

1. The user can set the required pressure value and inflation time on the control and display module at any time, and upload relevant parameters to the data system in real time. The system will monitor and operate the system according to the new parameters in real time. All kinds of emergencies in the situation are timely and strained to enhance system security. At the same time, the precise air pressure control makes the shape consistency of the balloon inflated, which makes the image data obtained by scanning the object multiple times more relevant and easier to compare.

2, because the air pump does not change or change the air pressure for a long time, it is very likely that the balloon is ruptured or leaking. If the pump continues to inflate or pumping gas, it may cause damage to the patient's body cavity. Therefore, if the pressure system monitors the length of the pump When the air is inflated or deflated, the air pressure has not changed or changed little, and the protection mechanism is automatically activated to close the air pump and strengthen the system safety.

3. This application can realize the rapid opening and closing of the electromagnetic valve by means of automatic control, reduce the interference of the charging and discharging gas to the monitoring system, and increase the accuracy of the monitoring system.

4. This method of inflation and pumping is two different channels, which can quickly fill and deflate two processes. Switching, especially during pressure relief protection, rapid pumping can prevent the damage caused by excessive inflation on the balloon and the patient's body cavity, and enhance system safety.

5. This application adopts an automatic charge and discharge air pump, which has a large inflation pressure range. According to the body cavity environment adopted in the present application, the pressure range is generally controlled at 1 to 5 atmospheres. The pressure pressure range that can be provided by manual pressure supply is much smaller than the pressure range of the present application.

Example 6

As shown in FIG. 14, an optical clock module used in an OCT endoscopic scanning imaging system includes an interference module, a detector module, and an optical clock conversion circuit module, the optical clock conversion circuit module including a wide frequency 90 degree phase shifting , zero-crossing comparators, and circuits consisting of XOR gates, OR gates, and optical clock signal output modules. The interference module adopts an all-fiber Mach-Zehnder interferometer (MZI) structure, which is mainly composed of two fiber couplers, wherein the second coupler is a symmetric 2×2 fiber coupler, first coupled in the first fiber. The device is divided into two paths of light, and the two paths of light respectively pass through two segments of the first optical fiber and the second optical fiber with a fixed optical path difference, and interference occurs at the second fiber coupler. The detector module is composed of a high-speed balanced photodetector and is mainly used for converting an interference optical signal output from the interference module into an electrical signal. The MZI electrical signal converted by the detector module is partially transmitted to the broadband 90-degree phase shifter, and the other portion is transmitted to the zero-crossing comparator, and the phase shift of the electrical signal transmitted to the broadband 90-degree phase shifter is 90 degrees. The comparator is mainly used for zero-crossing comparison of signals in which phase shift occurs and phase shift has not occurred to be converted into a digital signal. The XOR gate is mainly used to combine two digital clock signals to obtain two clock signals in one free spectral region (FSR), which increases the maximum imaging depth of the OCT without increasing the FSR, and reduces the optical The jitter produced by the signal. Since the swept laser always has some idle time between two adjacent scans, the optical clock signal needs to fill in some false clock signals in the blank through an OR gate to ensure that the high speed analog capture card can work normally, or The gate implements the function of combining a real optical clock signal with a false clock signal. The optical clock signal output module is mainly used for transmitting the combined real optical clock signal and the fake clock signal to the data acquisition module.

As shown in FIG. 15, 91 is an MZI electrical signal that undergoes a 90-degree phase shift after a wide-band 90-degree phase shifter, 92 is an MZI electrical signal in which no phase shift occurs, and 93 is a zero-cross comparison of the MZI signal 91 in which phase shift occurs. The subsequent digital signal, 94 is a digital signal after the zero-comparison of the MZI signal 92 without phase shift, because the zero point of the MZI signal is evenly distributed in the frequency domain, so the rising edge or falling of the digital signal generated after the zero-crossing comparison The edges are also evenly distributed in the frequency domain, 96 is a uniformly distributed number in the frequency domain. The signals after the word signals 93 and 94 are combined by the exclusive OR gate, 95 is a false clock signal, and 95 and 96 together constitute an optical clock signal after the OR gate merge.

As shown in FIG. 16, an OCT endoscopic scanning imaging system is similar to Embodiment 2-5 except that the frequency sweeping laser module, the optical clock module, the data acquisition module, the data processing module, the image display module, and the execution are performed. The mechanism, the OCT micro probe and the balloon catheter, and the charging and discharging device, wherein: the optical clock module is as shown in FIG. 14 , and includes an interference module, a detector module and an optical clock conversion circuit module, and the optical clock conversion circuit module Including a wide frequency 90 degree phase shifter, a zero-crossing comparator, and an exclusive OR gate, or a circuit composed of an OR gate and an optical clock signal output module, the optical interference signal generated from the MZI is converted into an electrical signal by a balanced photodetector, and then passed through a A circuit consisting of a wide-band 90-degree phase shifter, a zero-crossing comparator, and an exclusive OR gate is converted into an optical clock signal that is uniform in the frequency domain and frequency-variant in the time domain, or a gate-to-real optical clock signal and a false clock signal. Combine the normal operation of the high speed analog acquisition card.

The frequency sweeping laser module comprises a high speed swept laser, an optical isolator and a fiber coupler, and the optical signal output from the swept laser is isolated from the subsequent optical path, preventing the optical signal returned by the subsequent optical path from interfering with the normal operation of the laser, and A small portion of the swept laser output is output to the optical clock module, and most of the laser continues to output; the optical clock module includes an interference module, a detector module, and an optical clock conversion circuit module, which are mainly used to obtain uniformity and time in the frequency domain. The optical clock signal of the variable frequency in the domain; the data acquisition module can adopt a high-speed analog-digital acquisition card, and the original image signal is mainly collected based on the optical clock signal output by the optical clock module, and is provided to the data processing module for processing; The data processing module is a chip with digital signal processing capability (such as CPU, GPGPU, DSP, FPGA, etc.), and is mainly used for processing and converting the original signal into a final image signal; the image display module is mainly used for displaying an image signal. And responsible for the post-processing of the image and the measurement work; the actuator is light The rotary connector, the motor and the electric translation stage are mainly used for driving the mechanical spiral scan of the OCT micro probe to obtain an OCT image; the OCT micro probe is mainly used for entering the internal organs of the human body to transmit the swept laser and collecting from the biological tissue. Backscattered optical signal; the balloon catheter is used to expand the internal organ conduit of the human body, eliminate wrinkles and stabilize the OCT microprobe at the center of the balloon; the charge and discharge device is primarily used to expand the balloon catheter.

Example 7

As shown in FIG. 17, a microprobe for OCT imaging scanning of a human esophagus inserts the microprobe from the proximal guidewire lumen of the balloon into the balloon inner tube 9, and connects the balloon handle ventilation interface 13 to An automatic inflation pump (not shown) inflates the balloon 10 to a rated air pressure to expand the esophagus. The balloon 10 and the inner tube 9 are made of an optically transparent material and have excellent transparency. Light performance. The balloon 10 functions to expand the esophagus to reduce esophageal wrinkles and to secure the OCT microprobe within its working distance. The balloon 10 expands to a radius of between about 8 and 10 mm, which is the radius after the esophagus is fully deployed. Therefore, a relatively large working distance (approximately 8-10 mm) is a must for the OCT microprobe. When the working distance is large, the OCT micro-probe needs to select a long self-focusing lens, which is not easy to bend, and causes inconvenience when used in a body cavity, and the lens may be broken when excessively bent, which has a safety hazard. Therefore, by setting the glass rod to change the working distance of the self-focusing lens, making it a multi-section structure, the micro-probe can ensure better cornering, so that the entire micro-probe can enter a narrow channel when the working distance is large.

In this embodiment, the glass rod 3 is specially designed. The foregoing has mentioned that changing the bonding distance of the glass rod 3 and the single-mode optical fiber 1 can change the working distance of the micro-probe. In this example, the micro-probe is used for the human esophagus, and the working distance is about 8 -10mm, by the calculation and testing of the glass rod 3 and the single-mode fiber 1 on both sides of the glue distance should be less than 0.3mm. The design of the glass rod 3 not only enables the OCT micro-probe to work over a long working distance range, but also changes the numerical aperture and lateral resolution of the micro-probe, thereby increasing the bonding distance between the glass rod and the single-mode fiber. The distance increases the numerical aperture and increases the lateral resolution of the microprobe.

The relationship between the numerical aperture and the aperture of the optical element is as follows:

Where D is the clear aperture of the optical component, WD is the working distance, and NA is the numerical aperture. When the working distance WD is constant, the numerical aperture is proportional to the aperture (D) of the optical component, due to the processing process of the self-focusing lens itself. The defect has a clear aperture of only about 80% of its diameter. However, since the single-mode fiber is thin, in practical applications, the pass-through aperture that can be actually used by directly connecting the self-focusing lens through a single-mode fiber has only a self-focusing lens. The diameter of the self-focusing lens is less than 10%. In order to increase the clear aperture of the self-focusing lens, the present invention incorporates the glass rod. Since the light is diffused and transmitted in the glass rod, the glass rod expands the light. The aperture of the aperture that is actually used when the optical fiber passes through the self-focusing lens is increased. .

The relationship between resolution and numerical aperture is as follows:

Where λ is the incident light wavelength is a fixed value, and the lateral resolution ΔX is proportional to the numerical aperture (N.A), that is, the larger the numerical aperture, the higher the lateral resolution (the smaller the value).

In summary, the use of the glass rod 3 not only increases the working distance of the microprobe, but also increases the numerical aperture of the microprobe, and the increase of the numerical aperture also leads to an increase in lateral resolution, and this design is also extremely large. The length of the self-focusing lens is shortened to ensure the bending of the micro-probe, so that the entire micro-probe can still directly enter the human esophagus through the endoscopic forceps together with the balloon catheter. This effective design allows the lateral resolution of the probe to be approximately 10-30 microns and the working distance to be 8-10 mm. The relationship between the working distance and the lateral resolution of the microprobe is shown in Figure 18. The diameter of the entire microprobe is less than 1.5 mm. For example, a self-focusing lens with a diameter of 1.0 mm is used. The diameter of the whole microprobe is less than 1.3 mm. For a self-focusing lens with a diameter of 0.7 mm, the diameter of the entire microprobe is less than 1.0 mm; if the diameter is 0.5 mm Self-focusing lens, the entire microprobe has a diameter of less than 0.7 mm. The application enables the OCT microprobe to be applied to a narrow space with a large working distance, a large numerical aperture and a high resolution.

Figure 19 is a section of a healthy animal esophagus obtained by an OCT balloon catheter endoscope with an image size of 1200 lateral scans x 4096 longitudinal scans, a scan rate of 0.2 cm/3 s, and a scale of 1 mm. Figure 20 is a partial enlarged view of the esophageal image of the healthy animal of Figure 19, the distinguishable layers including 81: squamous epithelial layer (SE), 82: lamina propria (LP), 83: muscle mucosa (MM), 84: mucosa Lower layer (SM) and 85: intrinsic base layer (MP). Figure 21 is a 3D image of a healthy animal's esophagus produced by OCT endoscopic scanning imaging system for lumen surface and depth scan, and then software reconstruction of the scan data.

The above description is only a preferred embodiment of the present application, so that those skilled in the art can understand or implement the invention of the present application. Various modifications and combinations of these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the application. . Therefore, the present application is not limited to the embodiments shown herein, but the broadest scope consistent with the principles and novel features disclosed herein.

Claims (6)

1. A method for processing an OCT signal in an OCT system using a general-purpose image processor, the method comprising: a, data acquisition; b, data transmission; c, data processing; and d, passing to an image display library, wherein the method comprises At the same time of transmitting data, the OCT raw data transmitted to the device memory last time is processed in parallel by the general image processor, and the processed data is placed in the memory of the image display library.
2. The method of processing an OCT signal in an OCT system using a general-purpose image processor according to claim 1, wherein the data processing comprises the following steps:

   C1, one-dimensional digital resampling;

   C2, one-dimensional fast Fourier transform;

   C3, calculate the amplitude and normalize.
3. The method for processing an OCT signal in an OCT system using a general-purpose image processor according to claim 1, wherein a one-dimensional digital re-sampling step performs fast one-dimensional cubic interpolation by two linear texture search to improve resampling Precision.
4. An OCT endoscopic scanning imaging system, the OCT system comprising a frequency sweeping laser module, an interference module, a detector module, a data acquisition module, a data processing module, an image display module, an actuator, a balloon catheter, a charging and discharging device, It is characterized in that the OCT endoscopic scanning imaging system performs OCT signal processing by the method according to any one of claims 1 to 3.
5. The OCT endoscopic scanning imaging system according to claim 4, wherein the actuator drives the OCT microprobe to perform a rotational scan to generate a 3D image.
6. The OCT endoscopic scanning imaging system according to claim 4, wherein the charging and discharging device is an automatic charging and discharging device, and the automatic charging and discharging device comprises: a control and display module, an air pump, a pneumatic solenoid valve, Venting solenoid valve, pressure sensor, explosion-proof pressure sensor, mechanical pressure switch.

Images