Magnetic resonance imaging pure hardware magnetic resonance signal reception and feature extractor

Abstract

This invention discloses a pure hardware magnetic resonance imaging (MRI) signal receiver and feature extractor, belonging to the field of medical device technology. The core signal processing function of this feature extractor is implemented by pure analog circuitry, without relying on any digital signal processing unit. It includes an analog receiving front-end that processes the MRI signal according to reference parameters stored in a one-time programmable memory. The pure analog feature extractor directly extracts the physical features of the MRI signal from the processed analog signal. All calibration parameters are permanently stored in the one-time programmable memory at the factory and cannot be modified during operation. This invention can effectively improve the speed and accuracy of MRI signal processing while reducing system complexity and power consumption.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a signal receiving and feature extraction device for a magnetic resonance imaging system. Background Technology

[0002] The magnetic resonance signals received by magnetic resonance imaging (MRI) systems are extremely weak and require a series of processing steps, including amplification, demodulation, and feature extraction, before they can be used for imaging. Current MRI systems typically use analog-to-digital converters (ADCs) to convert analog signals into digital signals, which are then processed by digital signal processors (DSPs). This approach has the following problems: First, the ADC process introduces quantization noise, reducing signal accuracy; second, digital signal processing has latency, which cannot meet the requirements of real-time processing; and finally, digital circuits are complex, consume a lot of power, and are expensive.

[0003] Therefore, there is a need in this field for a high-precision, low-latency, and low-power magnetic resonance signal receiving and feature extraction device. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a pure hardware magnetic resonance signal receiver and feature extractor for magnetic resonance imaging.

[0005] The technical solution of the present invention is as follows: A pure hardware magnetic resonance signal receiver and feature extractor for magnetic resonance imaging (MRI) is disclosed. The core signal processing function is implemented by pure analog circuitry without relying on any digital signal processing unit. It includes an analog receiving front-end that processes the magnetic resonance signal according to reference parameters stored in a one-time programmable memory. The pure analog feature extractor directly extracts the physical features of the magnetic resonance signal from the processed analog signal. All calibration parameters are stored in the one-time programmable memory at the factory and cannot be modified during operation.

[0006] Furthermore, the core functionality of the feature extractor does not rely on analog-to-digital converters and digital signal processors for execution; it does not contain any adaptive signal processing algorithms based on real-time signal quality feedback.

[0007] Furthermore, the analog receiving front end includes a low-noise amplifier and a quadrature demodulator; the quadrature demodulator performs demodulation according to a reference frequency fixed in a one-time programmable memory.

[0008] Furthermore, the physical features extracted by the pure analog feature extractor include at least one of peak amplitude, phase, and relaxation time.

[0009] Furthermore, it also includes a preamplifier noise figure temperature compensation circuit. The noise figure of the preamplifier is measured at different temperatures at the factory, and the temperature compensation curve is stored in a one-time programmable memory. The hardware automatically adjusts the bias current of the preamplifier based on the on-chip temperature sensor data.

[0010] Furthermore, it also includes a real-time motion artifact detection hardware circuit. The hardware analyzes the phase change of the echo signal in real time during data acquisition in the data acquisition space. When the phase jump between adjacent echoes exceeds the threshold of the one-time programmable memory, the data in that row is automatically marked as motion contamination, and key data in the central area of ​​the data acquisition space is automatically reacquired.

[0011] Furthermore, it also includes a hardware-level calibration circuit for relaxation time, a one-time programmable memory to store reference relaxation time values ​​and calibration curves for different tissues, and the hardware automatically looks up a table to output a quantitative relaxation time value after acquiring multi-echo or multi-reversal time data.

[0012] Furthermore, it also includes a multi-channel receiver gain and phase automatic calibration circuit. Before each scan, the hardware sends a standard calibration signal to all channels, measures the gain and phase deviation of each channel, automatically adjusts the gain of each channel to be consistent, and updates the phase correction coefficient.

[0013] Furthermore, it also includes a gradient decoupling feedforward noise cancellation circuit, in which the hardware automatically performs feedforward noise cancellation according to the gradient switching timing to eliminate the coupling interference of the gradient magnetic field to the receiving coil; it also includes an analog circuit temperature compensation circuit, in which the hardware automatically adjusts the reference voltage and bias current of the analog circuit by looking up a table from the temperature compensation curve stored in the one-time programmable memory based on the temperature sensor data on the chip; the analog receiving circuit adopts an independent power domain, with physical isolation and filtering added between it and the RF power domain.

[0014] Furthermore, it also includes a real-time motion detection circuit for brain functional imaging. After each whole-brain acquisition, the hardware automatically compares the displacement of the current image with that of the reference image. When the displacement exceeds the threshold of the one-time programmable memory, it automatically updates the scanning position parameters for the next acquisition cycle. It also includes a dedicated calibration circuit for temperature imaging. The temperature coefficient of the proton resonance frequency is fixed in the one-time programmable memory. The hardware automatically calculates the temperature distribution map during temperature imaging scanning. When the system is idle, it automatically verifies the accuracy of temperature measurement using a calibration phantom with a known temperature. Beneficial effects

[0015] This invention employs purely analog circuitry to receive and extract features from magnetic resonance signals, eliminating the need for analog-to-digital conversion and digital signal processing. This effectively avoids the introduction of quantization noise and improves signal processing accuracy. The purely analog architecture enables real-time signal processing, meeting the stringent speed requirements of magnetic resonance imaging. Furthermore, this invention features a simple structure, low power consumption, low cost, and high reliability. Detailed Implementation

[0016] The present invention will be further described in detail below with reference to specific embodiments.

[0017] The pure hardware magnetic resonance signal receiver and feature extractor for magnetic resonance imaging of the present invention mainly consists of an analog receiving front end, a pure analog feature extractor, a one-time programmable memory, and a calibration circuit.

[0018] The analog receiver front end is used to receive magnetic resonance signals and perform amplification and demodulation. It includes a low-noise amplifier and a quadrature demodulator. The low-noise amplifier amplifies the weak magnetic resonance signal, and the quadrature demodulator demodulates the radio frequency signal into a baseband signal.

[0019] A pure analog feature extractor directly extracts the physical features of the magnetic resonance signal, such as peak amplitude, phase, and relaxation time, from the demodulated analog signal. A pure analog feature extractor consists of analog computing circuits and can perform various signal processing functions.

[0020] One-time programmable memory is used to store reference and calibration parameters. These parameters are written by professionals using specialized equipment at the factory and are permanently fixed after writing, and cannot be modified during operation.

[0021] The calibration circuit includes a preamplifier noise figure temperature compensation circuit, a multi-channel receiver gain and phase automatic calibration circuit, and an analog circuit temperature compensation circuit, which can automatically calibrate various parameters of the system and ensure the stability of the system performance.

[0022] In practical applications, after the magnetic resonance signal is received by the receiving coil, it is first amplified by a low-noise amplifier and then demodulated by a quadrature demodulator. The demodulated baseband signal is then sent to a pure analog feature extractor to extract the physical features of the magnetic resonance signal. The extracted features can be used directly for imaging or output to external devices via an interface. Throughout the signal processing, the calibration circuit monitors the system's operating status in real time and automatically calibrates various parameters to ensure stable system performance.

Claims

1. A pure hardware magnetic resonance signal receiver and feature extractor for magnetic resonance imaging, characterized in that, The core signal processing function is implemented by pure analog circuits without relying on any digital signal processing unit; it includes an analog receiving front-end, which processes the magnetic resonance signal according to reference parameters fixed in a one-time programmable memory; and a pure analog feature extractor directly extracts the physical features of the magnetic resonance signal from the processed analog signal. All calibration parameters are permanently stored in the one-time programmable memory at the factory and cannot be modified during operation.

2. The feature extractor according to claim 1, characterized in that, The core functionality of the feature extractor is not dependent on analog-to-digital converters and digital signal processors; it does not contain any adaptive signal processing algorithms based on real-time signal quality feedback.

3. The feature extractor according to claim 1, characterized in that, The analog receiving front end includes a low-noise amplifier and a quadrature demodulator; the quadrature demodulator performs demodulation based on a reference frequency fixed in a one-time programmable memory.

4. The feature extractor according to claim 1, characterized in that, The physical features extracted by the pure analog feature extractor include at least one of peak amplitude, phase, and relaxation time.

5. The feature extractor according to claim 1, characterized in that, It also includes a preamplifier noise figure temperature compensation circuit. The noise figure of the preamplifier is measured at different temperatures at the factory, and the temperature compensation curve is stored in a one-time programmable memory. The hardware automatically adjusts the bias current of the preamplifier based on the on-chip temperature sensor data.

6. The feature extractor according to claim 1, characterized in that, It also includes a real-time motion artifact detection hardware circuit. The hardware analyzes the phase change of the echo signal in real time when the data is acquired in the data acquisition space. When the phase jump between adjacent echoes exceeds the threshold of the one-time programmable memory, the data line is automatically marked as motion contamination, and key data in the central area of ​​the data acquisition space is automatically reacquired.

7. The feature extractor according to claim 1, characterized in that, It also includes a hardware-level calibration circuit for relaxation time, a one-time programmable memory to store reference relaxation time values ​​and calibration curves for different tissues, and the hardware automatically looks up a table to output a quantitative relaxation time value after acquiring multi-echo or multi-reversal time data.

8. The feature extractor according to claim 1, characterized in that, It also includes a multi-channel receiver gain and phase automatic calibration circuit. Before each scan, the hardware sends a standard calibration signal to all channels, measures the gain and phase deviation of each channel, automatically adjusts the gain of each channel to be consistent, and updates the phase correction coefficient.

9. The feature extractor according to claim 1, characterized in that, It also includes a gradient decoupling feedforward noise cancellation circuit, in which the hardware automatically performs feedforward noise cancellation according to the gradient switching timing to eliminate the coupling interference of the gradient magnetic field to the receiving coil; it also includes an analog circuit temperature compensation circuit, in which the hardware automatically adjusts the reference voltage and bias current of the analog circuit by looking up a table from the temperature compensation curve stored in the one-time programmable memory based on the temperature sensor data on the chip; the analog receiving circuit adopts an independent power domain, with physical isolation and filtering added between it and the RF power domain.

10. The feature extractor according to claim 1, characterized in that, It also includes a real-time motion detection circuit for brain functional imaging. After each whole-brain acquisition, the hardware automatically compares the displacement of the current image with that of the reference image. When the displacement exceeds the threshold of the one-time programmable memory, the scanning position parameters for the next acquisition cycle are automatically updated. It also includes a dedicated calibration circuit for temperature imaging. The temperature coefficient of the proton resonance frequency is fixed in the one-time programmable memory. The hardware automatically calculates the temperature distribution map during temperature imaging scanning. When the system is idle, the accuracy of the temperature measurement is automatically verified using a calibration phantom with a known temperature.