SU1100547A1 - Method and device for extracting weak signals of magnetic resonance - Google Patents

Abstract

1. A method for isolating weak magnetic resonance signals, including coherent summation of a recorded signal during repeated passage of a resonance region of the Spectrum, characterized in that, in order to increase sensitivity, and accuracy of reproduction of the magnetic resonance spectrum while reducing the memory volume of the recording device by averaging only noise component of the recorded signal, with each subsequent passage of the resonant region of the spectrum the recorded signal is compared with the accumulated signal m, the difference between them is measured and the difference signal is coherently summed, with each difference passing (the first signal is added to the accumulated signal with a weighting factor inversely proportional to the number of full passages of the resonant region of the spectrum.

Description

Sc

-

cl

four

T

: D 4Z} -

 Signal breakdown

./

2. A device for isolating weak magnetic resonance signals, comprising analog-to-digital and digital-to-analog converters and a memory device, distinguished by the fact that it further comprises detracting and summing amplifiers, a controlled voltage divider and a control unit, the subtracting amplifier, A controlled voltage divider that summarizes the invention relates to technical physics and can be used to record weak magnetic resonance signals: EPR, NMR, FMR, DEAR, and others. A method is known for detecting weak signals (signals with a low signal-to-noise ratio) of magnetic resonance and a device for making it based on modulating resonant magnetic resonance conditions (for example, modulating a magnetic field into recording an EPR signal). In this case, the magnetic resonance is detected at the modulation frequency, which makes it possible to use narrow-band gain and synchronous detection to increase the sensitivity. An improvement in the signal-to-noise ratio of the detected magnetic resonance signal is achieved by narrowing the band. bandwidth of the recording system (the bandwidth is determined by the constant time of the low-frequency RC filter switched on at the output of the synchronous detector), which is associated with an increase in the recording time ij. A disadvantage of the known recording method is the limited sensitivity due to the low-frequency magnetic resonance input to the recorded signal. Drifts (noise type 1 / f) of the output stages of the synchronous detector. Closest to the proposed by the technical nature of the method of extracting magnetic resonance signals, including coherent summation of the recorded signal at

A converter, analog-to-digital converter, memory and digital-to-analog converter are connected in series, the second inputs of the subtractor and summing amplifiers are connected to the output of the digital-analog converter, and the outputs of the control unit are connected to the control inputs of the voltage divider and memory respectively. repeated passage of the resonant region of the spectrum. A device for implementing the method. The spectra accumulator includes analog-digital and digital-to-analog converters, a digital adder and a memory device. With coherent summation of signals recorded over N passages of the resonance region of the spectrum, the amplitude of the magnetic resonance signal increases N times, while the noise intensity increases only - {раз times, because, unlike the magnetic resonance signal, its noise components are not correlated with each other. Thus, the known method provides an improvement in the signal-to-noise ratio of the recorded signal by a factor of H. The memory size P of the storage device of the storage device is determined by the size and frequency of the analog-to-digital converter (ADC), the number of accumulations, the number of accumulation channels K and. is 1 i p; 0 where irn is the number of binary bits, the ADC, which determines the reproduction accuracy of the magnetic resonance signal (TO4HocTb / 2w). With a given amount of memory P, an increase in sensitivity (i.e., an increase in the number of accumulations of N) is possible only due to a decrease in the ADC m resolution (i.e. a decrease in the reproduction accuracy of the recorded magnetic resonance spectrum). Conversely, an increase in the accuracy of the reproduction of the spectrum by increasing the bit width of the frt ADC is possible only to the detriment of the maximum achievable sensitivity 2. . The disadvantages of these methods and devices are the low reproduction accuracy of the accumulated magnetic resonance spectra and the limited sensitivity. The purpose of the invention is to increase the transience and accuracy of the reproduction of the magnetic resonance spectrum while reducing the memory capacity of the storage device, by averaging only the noise components of the recorded signal. The goal is achieved by the method of allocating weak magnetic resonance signals. including a coherent summation of the recorded signal with a multiple passage of the resonance region of the spectrum, with each passage of the resonance region of the spectrum, the recording signal is compared with the accumulated signal, the difference between them is measured and the difference signal is coherently measured, and each time the differential passes, the differential signal reaches the accumulated signal with a weighting factor , inversely proportional to the number of completed passages of the resonance region of the spectrum. A device for isolating weak magnetic resonance signals, comprising analog-to-digital and digital-to-analogue converters and a memory device, further comprises subtracting and summing amplifiers, a controlled voltage and control divider, and a subtracting amplifier, a controlled voltage divider, a summing amplifier, analog-digital The npe generator, memory device and digital-to-analog converter are connected in series, the second inputs of the subtractor and summing amplifiers are connected to the output of the analogue converter digits, and the outputs of the control unit are connected respectively to the control inputs of the voltage divider and the memory device. FIG. 1 shows a block diagram of a device for extracting weak magnetic resonance signals implementing the proposed method, FIG. 2 - EPR spectra of the sample (diphenyl picrilhydrazine (DPPH). The device contains analog-digital 1, digital and analog 2 converters, memory 3, subtracting 4 and summing 5 amplifiers, controlled voltage divider 6 and control block 7, and subtracting amplifier 4, voltage divider 6, summing amplifier 5, ADC 1, memory 3 and DAC 2 are connected in series, the second inputs of the subtractive 4 and the summing 5 amplifiers are connected to the output of the DAC 2, and the outputs of the control unit 7 are connected respectively to to input voltage divider 6, and a memory 3. Presented in FIG. 2 ESR spectra of the sample of DPPH vscheleny known noise spectrum {8-10) and proposed (spectra 11-13) at various ways including savings. The device enters the input of the si-nal, registrations during the passage, the resonance region of the spectrum. , (1) where Si is the signal recorded during the 1st passage of the resonant spectral region, 5c is the magnetic resonance signal (informative component of the recorded sigal, 5ш1 noise component of the recorded signal. Unlike the well-known proposed magnetic resonance signal jj from noise is not based on the coherent summation of the recorded signals I1. (i.e.,. on the simultaneous summation of informative 5c. and sound; the component of the recorded signal a), but on the coherent summation of only noise components Bent .E, 5c; of the recorded signal, which is provided by performing the following operations: From the recorded signal 5 is subtracted, the signal recorded in the memory of the storage device (memory) 3, 5-Sm-i (2) is performed by the subtracting amplifier 4, one of the inputs of which receives a recorded signal 5,, 5 and on the other - a .5ni-o signal recorded in the memory of the memory during the previous one (1 pass of the resonance region of the spectrum. The difference signal is normalized by a weighting factor inverse to the number of passes made | (5; -S,.,); (3) what is being done by controlled voltage divider 6. The normal extended signal is added to the signal stored in the memory of memory 3. . ., t (3; -5п,., () that is done with the help of summing amplifier 5, one of the inputs of which receives the normalized difference signal, and on the other - the signal from the memory of memory 3. Thus, as a result of performing the above operations, when the resonance region of the memory passes through the memory of memory 3, the recorded signal is 5 "-5" -., 4 5-5p; mU (5) Digital conversion of the signal for recording into the memory is using the A / D converter 1, and the inverse conversion (i.e., conversion from c 1 of the digital form to analog) when outputting from memory 3 of the memory using the DAC 2 The reproduction accuracy of the recorded signal is determined by the conversion accuracy of the ADC and DAC and is 2, where rp is the ADC and DAC resolution (the ADC and DAC have the same bit width). The sequence of operations performed is determined by control unit 7, which In addition, it generates a sweep signal for the passage of the resonant region of the spectrum. Consider the levels of the signals with different passages of the resonance region of the spectrum. During the first passage of the resonant region of the spectrum, the recorded signal is recorded in the memory of the accumulator 5 p, 5, because before starting the measurement, the information recorded in the memory of memory 3 is erased (5p (5 0). 476 During the second passage of the resonant region of the spectrum. performing the operation described by the expression (2), the informative component 5 is compensated in the recorded signal and only the noise component S2-5nrS - 5, Sj-Sy, is shown. (6) After performing the operations described by expressions (3) and (4) ), the memory of memory 3 records the signal: 5, 6c (,, (7) Similarly, after N passages of the resonance region of the spectrum, the signal a — C. 4 1 V a n with N —cf, will be recorded in the memory of memory 3. Since the noise components of Si are not correlated, after each N, summations their intensity increases only once with respect to the first component 5c ,; and as a result of Sv, .Sc-qV5 Thus, after N passes of the resonance region (and, accordingly, N accumulations), the signal-to-noise ratio of the recorded magnetic resonance signal at the output of devices increases once, but unlike the known way the signal is weak in the signals without increasing the intensity of the magnetic resonance signal by decreasing the level of noise by a factor of 41. The memory size of the charger is determined only by the size of the ADC W and the number of channels K. It does not depend, unlike the known device, on the number of accumulations N 2 Thus, with the same memory capacity of the memories of the devices implementing the known and proposed methods, the proposed method weak magnetic resonance signals provide higher accuracy of reproduction of magnetic resonance spectra due to higher ADC resolution (which can be used for npezwnraeMoro device

Claims (2)

1. A method for isolating weak magnetic resonance signals, including coherent summation of the recorded signal during multiple passage of the resonance region of the spectrum, characterized in that, in order to increase the sensitivity and accuracy of the reproduction of the magnetic resonance spectrum while reducing the memory capacity of the storage device by averaging only the noise components of the recorded signal, with each subsequent passage of the resonance region of the spectrum, the recorded signal is compared with the accumulated signal, and measures the difference between them and coherently dosummiruyut difference signal, wherein during each passage of the difference signal dosummiruyut to the accumulated signal with a weighting factor inversely proportional to the number of passages made spectrum resonance region. Sc 2. A device for extracting weak magnetic resonance signals containing analog-to-digital and digital-to-analog converters and a storage device, characterized in that it further comprises subtracting and summing amplifiers, a controlled voltage divider and a control unit, wherein a subtracting amplifier, a controlled voltage divider, summing the amplifiers an amplifier, an analog-to-digital converter, a storage device and a digital-to-analog converter are switched on sequentially, the second inputs of subtracting and summing of the amplifiers are connected to the output digital to analog converter, and the control unit outputs are connected respectively to the control inputs of the voltage divider and a memory.