A portable fatty liver detection device based on ultrasound and data processing method thereof

Abstract

The present invention belongs to the technical field of medical equipment, and specifically relates to a portable fatty liver detection device based on ultrasound and a data processing method thereof. In view of the difficulties in miniaturization and portability of the ultrasonic fatty liver detection device in the prior art, the present invention provides a data processing method for a portable fatty liver detection device based on ultrasound, which includes: making the ultrasonic probe emit ultrasonic waves of at least two frequencies and receiving ultrasonic echo data, and obtaining ultrasonic RF data after beamforming the ultrasonic echo data; finally, according to the user's instructions, the ultrasonic RF data is output, or the ultrasonic RF data is post-processed in the portable fatty liver detection device. The present invention also provides a portable fatty liver detection device with a simple structure, low power and small size. The present invention enables the portable device to detect fatty liver by obtaining results such as ultrasonic attenuation coefficient or elastic imaging while having extremely low requirements on the chip structure.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical equipment, and in particular relates to a portable fatty liver detection device based on ultrasound and a data processing method thereof. Background Art

[0002] Fatty liver refers to a disease in which normal liver tissue is replaced by more than 5% to 6% fat. In this disease, the accumulation of fat leads to inflammation, cell death and scar formation, which we also call non-alcoholic steatohepatitis. If left untreated, steatohepatitis can lead to liver fibrosis, which causes reduced blood flow throughout the liver and the formation of scar tissue. Further, liver fibrosis may lead to cirrhosis, liver failure and liver cancer. From a global perspective, about 1 / 4 of the population suffers from non-alcoholic fatty liver disease, which varies slightly from country to country and from different populations to different populations. The prevalence of fatty liver in my country is slightly higher than the global average.

[0003] Common examination methods include liver function blood tests, abdominal B-ultrasound examinations, abdominal CT examinations, MRI examinations, ultrasound elastography, magnetic resonance elastography or liver biopsy, among which ultrasound B-ultrasound examinations are the most popular. The treatment of fatty liver or liver fibrosis usually lies in reasonable prevention, slowing down or reversing the progression of liver fibrosis. The solutions include lifestyle changes, drug treatment, and cholesterol and diabetes management.

[0004] At present, the most common way to detect fatty liver is to have doctors perform routine B-ultrasound scans of the liver for diagnosis. With the in-depth study of the characteristics and detection methods of fatty liver, many new methods have also begun to gain attention and be applied to detection equipment. For example, elastic imaging mainly measures the hardness of the liver by measuring the strain generated when the liver is vibrated, and then determines the severity of fatty liver. This method has been widely used, but the related equipment is expensive and bulky, which is not conducive to the popularization of fatty liver detection. Some scholars also perform fatty liver detection based on the change of ultrasonic attenuation coefficient, but their data is based on large-scale ultrasonic equipment, and an effective system equipment solution has not yet been formed, let alone portable and low-cost detection equipment.

[0005] In ultrasonic detection devices for fatty liver, in order to increase the amount of channel data, a more complex processing system is usually required. In digital beamforming systems, multiple high-speed and high-resolution ADCs are usually used. On the other hand, given that the sampling rate of the ADC directly affects the resolution and the accuracy of the phase delay adjustment between channels, the higher the sampling rate, the finer the phase delay; but correspondingly, increasing the sampling rate will lead to increased power consumption. The above two problems have led to difficulties in the design of portable devices. For example, the usual 128-channel all-digital beamforming requires 16 8-channel high-speed ADCs, and its power consumption and heat dissipation requirements far exceed the design requirements of portable devices.

[0006] It can be seen that a major factor restricting the miniaturization and portability of ultrasonic detection devices for fatty liver in the existing technology is that the processing process of ultrasonic signals is relatively complicated, resulting in relatively complex components such as the chipset of the device, which has an adverse effect on the power supply, heat dissipation and occupied space of the device. Summary of the invention

[0007] In view of the difficulties in miniaturization and portability of ultrasonic-based fatty liver detection devices in the prior art, the present invention provides a portable ultrasonic-based fatty liver detection device and a data processing method thereof, the purpose of which is to reduce the requirements for chips in the device by improving the processing method of ultrasonic data, thereby reducing the power supply, heat dissipation and space requirements of the device, making the device more suitable for portable devices.

[0008] A data processing method for a portable fatty liver detection device based on ultrasound, comprising: enabling an ultrasound probe to emit ultrasound waves of at least two frequencies and receive ultrasound echo data, performing beam synthesis on the ultrasound echo data to obtain ultrasound RF data; and finally, outputting the ultrasound RF data according to a user's instruction, or post-processing the ultrasound RF data in the portable fatty liver detection device.

[0009] Preferably, the ultrasonic echo data is collected by emitting ultrasonic waves of two frequencies, and the magnitude relationship between the two frequencies is an integer multiple; further preferably, the two frequencies are 3-5 MHz and 6-10 MHz, respectively.

[0010] Preferably, it specifically comprises the following steps:

[0011] (1) The ultrasonic probe transmits cylindrical waves in a synthetic transmission mode and receives ultrasonic echo data;

[0012] (2) Aperture synthesis technology is used to reconstruct the transmitted cylindrical wave and the received echo data to obtain beam data;

[0013] (3) using beam parallel synthesis technology, changing the array combination of the ultrasound probe for transmission and / or reception, and repeating steps (1) and (2), and superimposing the beam data obtained in each repetition with the previous beam data to obtain new beam data; until all the array combinations for transmission and / or reception are used once, the beam data finally obtained is the ultrasound RF data;

[0014] (4) According to the user's instruction, the ultrasonic RF data is output, or the ultrasonic RF data is post-processed in the portable fatty liver detection device.

[0015] Preferably, in step (4), the post-processing comprises the following steps:

[0016] (a) The ultrasonic RF data obtained from the cylindrical wave of each frequency are band-pass filtered separately;

[0017] (b) performing data demodulation on the data obtained in step (a);

[0018] (c) performing low-pass filtering and signal downsampling on the data obtained in step (b);

[0019] (d) performing envelope processing on the data obtained in step (c) to obtain an attenuation curve corresponding to the ultrasonic RF data obtained by the cylindrical wave of each frequency;

[0020] (e) After logarithmic compression of the attenuation curves, the difference obtained by subtraction is linearly fitted to obtain the ultrasonic attenuation coefficient α.

[0021] Preferably, in step (b), the data demodulation method is orthogonal demodulation to obtain data of the I channel and the Q channel; in step (c), the data of the I channel and the Q channel are respectively subjected to low-pass filtering and signal downsampling; in step (d), the data of the I channel and the Q channel obtained after processing in step (c) are subjected to envelope processing.

[0022] Preferably, in step (e), the ultrasonic attenuation coefficient α is calculated as follows:

[0023]

[0024] Among them, f1 and f2 are the probe transmission frequencies, I(d) is the ultrasonic signal intensity, and d is the ultrasonic propagation depth.

[0025] The present invention also provides a portable ultrasonic fatty liver detection device, comprising a front-end ultrasonic transmitting and receiving device, a processing chip and a power supply; the front-end ultrasonic transmitting and receiving device comprises an ultrasonic probe, a high-voltage pulse generator and an ultrasonic transmitting and receiving chip;

[0026] The ultrasonic transmitting and receiving chip and the processing chip are used to perform data processing according to the above method.

[0027] Preferably, the ultrasound transmitting and receiving chip is an AFE5805 chip, which is used to control the ultrasound probe to transmit cylindrical waves and receive ultrasonic echo data; and / or, the processing chip is an FPGA chip, which is used to implement the data processing process of data reconstruction, superposition and output in the above method.

[0028] Preferably, the FPGA chip is provided with a selector for controlling the direction of RF data, and the direction of the RF data is to be directly output or processed into an image signal.

[0029] Preferably, the ultrasound probe is a multi-element ultrasound probe.

[0030] Preferably, the ultrasound transmitting and receiving chip includes an ADC, and the number of the ADC is one.

[0031] The method of the present invention utilizes beamforming technology to realize the reuse of array elements and processing chips on the ultrasonic probe, thereby reducing the complexity of the ultrasonic probe and the processing chip without losing the amount of channel data, and reducing the transmission power of the ultrasonic probe and the working power of the processing chip. Therefore, the technical solution of the present invention can greatly reduce the size and power of the device, making it possible to miniaturize and port the device.

[0032] Conventional 128-channel full digital beamforming requires 16 8-channel high-speed ADCs, whose power consumption and heat dissipation requirements far exceed the design requirements of portable devices. In the preferred embodiment of the present invention, 128-channel full digital beamforming can be completed by 1 ADC.

[0033] Obviously, according to the above contents of the present invention, in accordance with common technical knowledge and customary means in the art, without departing from the above basic technical ideas of the present invention, other various forms of modification, replacement or change may be made.

[0034] The above contents of the present invention are further described in detail below through specific implementation methods in the form of embodiments. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following examples. All technologies realized based on the above contents of the present invention belong to the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the system structure of the portable ultrasonic fatty liver detection device in Example 1;

[0036] Figure 2 This is a schematic diagram of the principle of the selector in Example 1;

[0037] Figure 3 is an example of the grayscale image of RF data in Example 2;

[0038] Figure 4 Schematic diagram of the processing flow of RF data in Example 2;

[0039] Figure 5 These are two attenuation curves obtained before and after manually compressing the detection area in Example 2. DETAILED DESCRIPTION

[0040] The technical solution of the present invention is further described below through specific embodiments. It should be particularly noted that the algorithms of the steps of data collection, transmission, storage and processing not specifically described in the embodiments, as well as the hardware structure, circuit connection, etc. not specifically described can be implemented by the disclosed contents of the prior art.

[0041] Example 1: Ultrasound-based portable fatty liver detection device

[0042] The system structure of the device in this embodiment is as follows Figure 1 As shown, it includes a front-end ultrasonic transmitting and receiving device, a processing chip and a power supply. The front-end ultrasonic transmitting and receiving device includes a probe, a high-voltage pulse generator and an ultrasonic transmitting and receiving chip.

[0043] In this embodiment, the ultrasonic transmitting and receiving chip is an AFE5805 chip. The AFE5805 chip is used to control the transmission and reception of ultrasonic signals. The ultrasonic transmitting and receiving chip includes an ADC, and the number of the ADC is one.

[0044] The ultrasonic probe of this embodiment is a multi-element probe, which adopts a frequency doubling setting, that is, the two frequencies emitted are integer multiples of each other. The two frequencies can be emitted simultaneously or successively with a time difference.

[0045] The processing chip is an FPGA chip. The FPGA chip is used to process the collected ultrasonic signals and convert them into RF data, and at the same time control the front-end ultrasonic transmitting and receiving device. The FPGA issues instructions to control the high-voltage pulse generator to transmit electrical signals, and receives the digital signals transmitted by the AFE5805, and then beam synthesizes the RF data to transmit it to the PC through the USB interface. In this embodiment, the FPGA chip contains a multi-channel bandpass filter for signal reception, uses a digital filter to filter the received echo signal, decomposes and analyzes the received multi-channel bandwidth signal, and performs gain compensation on the ultrasonic signal. As a preferred solution, a selector for controlling the direction of RF data is provided in the FPGA chip, and the direction of the RF data is controlled by setting the selector logic to 0,1 to be directly output or processed into an image signal. The principle of the selector is as follows Figure 2 shown.

[0046] The power supply is responsible for providing the power support required for the operation of each system of the device, and at the same time providing a sufficiently strong working voltage for the high-voltage pulse generator.

[0047] The device provided in this embodiment has the advantages of simple structure, low power consumption and small size, thereby enabling the device to be portable.

[0048] Example 2: Data processing method of portable fatty liver detection device based on ultrasound

[0049] The device used in this embodiment is the portable ultrasonic-based fatty liver detection device described in Embodiment 1. This embodiment provides a method for processing the collected ultrasonic data.

[0050] The specific steps include:

[0051] (1) The ultrasonic probe transmits cylindrical waves in a synthetic transmission mode and collects ultrasonic echo data;

[0052] (2) Aperture synthesis technology is used to reconstruct the data obtained from the transmitted cylindrical wave and the received echo to obtain beam data;

[0053] (3) using beam parallel synthesis technology, changing the array combination of the ultrasound probe for transmission and / or reception, and repeating steps (1) and (2), and superimposing the beam data obtained in each repetition with the previous beam data to obtain new beam data, until all the array combinations for transmission and / or reception are used once, the beam data finally obtained is the ultrasound RF data;

[0054] (4) According to the user's instruction, the ultrasonic RF data is output, or the ultrasonic RF data is post-processed in the FPGA chip of the portable fatty liver detection device.

[0055] Existing low-end and mid-end ultrasound equipment transmits one ultrasonic beam (i.e., the cylindrical wave of this application) in a single channel and receives one echo data at the same time; high-end equipment from GE and Siemens can transmit one ultrasonic beam at a time in a single channel and receive 2-4 echo data. The method proposed in this embodiment enables the ultrasound equipment to transmit one cylindrical wave at a time and receive 64-128 echo data. Since 64-128 transmitted ultrasonic beams can be received at the same time, an ultra-high-speed image refresh rate of more than 3,000 frames per second can be achieved. Through optimization algorithm technology, the 64-128 reflected ultrasonic beams received are superimposed and calculated, and the image quality finally displayed on the ultrasound equipment is significantly improved. This method achieves the purpose of collecting high-quality ultrasonic RF data with a simple, low-power ultrasound probe and processing chip, and is therefore conducive to the miniaturization and portability of the device.

[0056] The grayscale diagram of the ultrasonic RF data detected in this embodiment is as follows: Figure 3 In this figure, both axes can be understood as time axes. The vertical axis is the ultrasonic emission and return process, and the horizontal axis is the duration of RF data acquisition. Figure 3 The brightness represents the amplitude of the echo intensity, with high amplitudes being brighter and low amplitudes being darker. In this embodiment, 2048 points in the vertical direction represent one echo data, and 2560 points in the horizontal direction represent 2560 data. The vertical point spacing is determined by the system sampling frequency, and the horizontal point spacing is determined by the PRF (Pulse Recurrence Frequency). During data analysis, the vertical time axis can be converted into depth, while the horizontal time axis is regarded as the time for collecting data. In this way, the 2048 points in each vertical line can be regarded as one echo data, and the amplitude of the echo signal can be obtained during analysis.

[0057] This embodiment can directly output RF data. At present, the existing technology also has technical solutions that use the processing results of ultrasound data to compare fatty liver and normal liver, but basically all of them use ultrasound images directly output by medical ultrasound equipment. The images are for doctors and patients, so there is no ultrasound RF data acquisition function. However, compared with the most original RF data generated by the equipment, the image information that users usually see on the ultrasound equipment loses a lot of important information. The data on each node has its own value. RF data carries rich acoustic information of the reflection point, including a lot of information that cannot be recognized by the naked eye, such as phase and frequency. If you start collecting from the RF data step, you can make full use of the rich information in the RF data to obtain a more accurate image. The system involved in this embodiment is based on ultrasonic RF data in both hardware acquisition and software analysis. Compared with the traditional ultrasonic equipment detection method, the collected ultrasonic RF data can be directly imported into the PC for display without post-processing such as the envelope and logarithmic compression provided by the FPGA chip. Therefore, the obtained data is relatively complete (the amount of information is about 10-20 times that of the image data), and even includes information such as phase and frequency. The rich amount of information can be used for customized scientific analysis, making it possible to accurately detect fatty liver when the detection device adopts a lower hardware configuration in order to achieve miniaturization.

[0058] The RF data is input to the PC via a USB interface connected to the FPGA chip for further processing. A preferred processing process for the RF data is as follows Figure 4 As shown, the specific steps are as follows:

[0059] (a) The ultrasonic RF data obtained from the ultrasonic signals of the two frequencies are band-pass filtered respectively;

[0060] (b) Performing orthogonal demodulation on the data obtained in step (a) to obtain two orthogonal local oscillator signals of the I channel and the Q channel. The ultrasonic RF data is an ultrasonic echo signal located in the high frequency segment. The orthogonal demodulation process can convert it to the base frequency, thereby making signal processing easier.

[0061] (c) Since the signal obtained by orthogonal demodulation still contains some high-frequency signals, the data obtained in step (b) are subjected to low-pass filtering in sequence. Then the signal is downsampled. Signal downsampling preserves the data of the original signal without losing it, greatly reduces the amount of ultrasonic signal data to be processed, and can improve the running efficiency of the software code.

[0062] (d) performing envelope processing on the data obtained in step (c) to obtain two sets of attenuation curves corresponding to ultrasonic signals of two frequencies;

[0063] (e) After logarithmic compression of the two sets of attenuation curves, the difference obtained by subtraction is linearly fitted to obtain the ultrasonic attenuation coefficient α.

[0064] The amplitude I(f, d) of the signal in the ultrasound image can be expressed as follows:

[0065] I(f,d)=A(f)D(f,d)R(f,d)G(d)exp(-2αdf)

[0066] In order to reduce the impact of gain on the experimental results, we can transform the above formula into:

[0067]

[0068] The attenuation coefficient can be expressed as:

[0069]

[0070] Where I(f,d) is the ultrasonic signal intensity at frequency f, A(f) is the initial pulse waveform amplitude, (f,d) is the transmit receive beam profile value, R(f,d) is the scattering parameter distribution value, G(d) is the receive signal gain compensation value, f1 is the transmit frequency 1, f2 is the transmit frequency 2, I(d) is the ultrasonic signal intensity, and d is the ultrasonic propagation depth.

[0071] From the above formula, it can be seen that the ultrasonic echo intensity of fatty liver at different frequencies shows a significant difference in attenuation trend with depth, and the attenuation coefficient of the liver can be calculated by comparing the intensity values ​​at two frequencies. Therefore, the spectrum difference method provided in this embodiment can be used to calculate the attenuation coefficient.

[0072] The ultrasonic attenuation coefficient (α) can be used to describe the attenuation of ultrasonic signals in human tissue. The attenuation coefficient of a normal human liver is between 0.5dB / cm and 0.6dB / cm, and the attenuation coefficient of patients with fatty liver is around 0.74dB / cm on average, which is significantly different from that of a normal human liver.

[0073] In addition to calculating the ultrasonic attenuation coefficient to determine whether the subject suffers from fatty liver, as an extended application of the technical solution provided by the present invention, ultrasonic signals under different pressures can also be collected by pressing or the like, and attenuation curves under different pressures can be obtained by the data processing method of this embodiment ( Figure 5 The two attenuation curves obtained before and after manual compression of the test site are shown), and the strain is calculated through the change of the attenuation curve to obtain an ultrasound image related to the elasticity of the liver.

[0074] In summary, the present invention provides a portable fatty liver detection device based on ultrasound, which has the advantages of simple structure, low power and small size. In addition, the present invention also provides a data processing method for the device, which enables the portable device to detect fatty liver by obtaining results such as ultrasonic attenuation coefficient or elastic imaging under extremely low requirements on chip structure.

Claims

1. A data processing method for a portable ultrasonic fatty liver detection device, characterized in that: It includes: The ultrasonic probe is made to emit ultrasonic waves of two frequencies and receive ultrasonic echo data, and the ultrasonic echo data is beamformed to obtain ultrasonic RF data; finally, according to the user's instruction, the ultrasonic RF data is output, or the ultrasonic RF data is post-processed in the portable fatty liver detection device; the magnitude relationship between the two frequencies is an integer multiple, and the two frequencies are 3-5MHz and 6-10MHz respectively; It specifically includes the following steps: (1) The ultrasonic probe transmits cylindrical waves in a synthetic transmission mode and receives ultrasonic echo data; (2) Aperture synthesis technology is used to reconstruct the transmitted cylindrical wave and the received echo data to obtain beam data; (3) using beam parallel synthesis technology, changing the array combination of the ultrasound probe for transmission and / or reception, and repeating steps (1) and (2), and superimposing the beam data obtained in each repetition with the previous beam data to obtain new beam data; until all the array combinations for transmission and / or reception are used once, the beam data finally obtained is the ultrasound RF data; (4) outputting the ultrasonic RF data or post-processing the ultrasonic RF data in the portable fatty liver detection device according to a user's instruction; In step (4), the post-processing comprises the following steps: (a) The ultrasonic RF data obtained from the cylindrical wave of each frequency are band-pass filtered separately; (b) performing data demodulation on the data obtained in step (a); (c) performing low-pass filtering and signal downsampling on the data obtained in step (b); (d) performing envelope processing on the data obtained in step (c) to obtain an attenuation curve corresponding to the ultrasonic RF data obtained by the cylindrical wave of each frequency; (e) After logarithmic compression of the attenuation curves, the difference obtained by subtraction is linearly fitted to obtain the ultrasonic attenuation coefficient α.

2. The data processing method according to claim 1, characterized in that: In step (b), the data demodulation method is orthogonal demodulation to obtain data of the I channel and the Q channel; in step (c), the data of the I channel and the Q channel are respectively subjected to low-pass filtering and signal downsampling in sequence; in step (d), the data of the I channel and the Q channel obtained after the processing in step (c) are subjected to envelope processing.

3. The data processing method according to claim 1, characterized in that: In step (e), the ultrasonic attenuation coefficient α is calculated as follows: Among them, f1 and f2 are the probe transmission frequencies, I(d) is the ultrasonic signal intensity, and d is the ultrasonic propagation depth.

4. A portable ultrasonic fatty liver detection device, characterized in that: It includes a front-end ultrasonic transmitting and receiving device, a processing chip and a power supply; the front-end ultrasonic transmitting and receiving device includes an ultrasonic probe, a high-voltage pulse generator and an ultrasonic transmitting and receiving chip; The ultrasonic transmitting and receiving chip and the processing chip are used to perform data processing according to the method described in any one of claims 1 to 3.

5. The detection device according to claim 4, characterized in that: The ultrasound transmitting and receiving chip is an AFE5805 chip, which is used to control the ultrasound probe to transmit cylindrical waves and receive ultrasonic echo data; and / or, the processing chip is an FPGA chip, which is used to implement the data processing process of data reconstruction, superposition and output in the method described in any one of claims 1-3; and / or, the FPGA chip is provided with a selector for controlling the direction of RF data, and the direction of RF data is to directly output or be processed into an image signal.

6. The detection device according to claim 4, characterized in that: The ultrasonic probe is a multi-element ultrasonic probe.

7. The detection device according to claim 4, characterized in that: The ultrasonic transmitting and receiving chip includes an ADC, and the number of the ADC is one.

Images