RU180351U1 - ADAPTIVE DIAGRAM-FORMING DEVICE FOR OBTAINING ULTRASONIC IMAGES - Google Patents

Abstract

The utility model relates to the field of medical instrumentation, in particular to devices for ultrasonic echolocation of internal organs, and can be used in medical diagnostic systems. Provides a reduction in hardware costs for the implementation of adaptive focusing, which affects the quality of the generated ultrasound image, due to the implementation of the phase correction method with the ability to build a small-sized device. The device includes a multi-element ultrasonic sensor 1 connected to the output of the switch 2 of the elements of the multi-element ultrasonic sensor 1 connected to the pulse generating unit 3. An analog receiver 4 is connected to the switch 2, which in turn is connected by its outputs to the corresponding input of one of the receiving channels 5. Each receiving channel contains a digital delay line 6, which includes a series-connected buffer memory (BZU) 7 and a filter an interpolator 8 connected to the multiplier 9. The central receiving channel 5 contains a phase shift former 10, and each of the remaining receiving channels 5 contains a phase shift former 10 and a phase corrector 11 for signal hold Shaper 10 phase shift contains a quadrature demodulator 12, which includes multipliers 13, 14, the outputs of which are connected to the inputs of low-pass filters (LPFs) 15, 16, the outputs of which are connected to the inputs of the calculator 17 of the argument. The phase delay corrector 11 of the signal (see Fig. 3) contains a series-connected subtractor 18, a driver 19 of the time delay of the signal and the adder 20. The device contains common to all receiving channels 5 a generator 21 of the reference frequency, the generator 22 of the signal delay and the adder 23. 2 .P. f-ly, 4 ill.

Description

This technical solution relates to the field of medical instrumentation, in particular to devices for ultrasonic echolocation of internal organs, and can be used in medical diagnostic systems.

From the current level of technology, it is known an ultrasonic chart-forming device (DFU), which includes: a multi-channel transmitter, a beam shaper, a channel switcher, a multi-channel receiver, a multi-element ultrasonic sensor, and the sensor elements are connected to the outputs of the channel switcher, the inputs of which receive signals from the first group of outputs of the beam former, the inputs of which are connected to the outputs of the multi-channel transmitter, and the second group of outputs - to the multi-channel receiver (see, for example, Ultrasound marketing diagnostic devices: A Practical Guide for Users, Osipov LV .; Publ: M .: Vidar-M, 1999, Figure 7).. In this case, DFU performs three main functions:

- switching groups of elements of a multi-element sensor when performing ultrasonic scanning;

- directional radiation of an ultrasonic (ultrasound) signal into biological tissues;

- directed reception of ultrasound echo signals reflected from the boundaries of biological tissues;

It is on the basis of directional radiation and reception of ultrasonic waves (focusing of the ultrasound beam) that echo signals are recorded that simultaneously arrive only from a narrow region along the sounding direction, and high spatial resolution is realized.

The physical meaning of focusing is to ensure the simultaneous arrival of ultrasonic signals from individual elements to a given point (focus). In DFU, delays are calculated due to different ways of propagating the ultrasonic signal from different sections of the aperture, in accordance with the laws of geometric optics. Further, when generating directional radiation in the zone of interest, excitation signals of the active elements of the sensor are generated at different time points, taking into account the calculated signal delays. Accordingly, when receiving an echo signal, a per-channel delay of echo signals (signals from different sensor elements) is performed according to the calculated values. Moreover, channel-by-channel delays of echo signals change directly during reception, thereby providing a mode of dynamic focusing on reception.

The disadvantage of this technical solution is that the focusing of the ultrasound beam, both on radiation and on reception, is calculated under the assumption of tissue uniformity and a constant ultrasound propagation velocity in them equal to 1540 m / s. In fact, the speed of propagation of ultrasound in various tissues can vary significantly: from 1470 m / s to 1600 m / s in soft tissues, and more than 3700 m / s in bone tissues (Ultrasound in medicine: physical principles of use: transl. From English / ed. K. Hill, J. Bamber, G. ter Haar. - Ed. 2, revised and additional - M .: FIZMATLIT, 2008, Fig. 5.3).

Due to the fact that the actual speed of ultrasound in tissues differs from the nominal value of 1540 m / s, respectively, they will differ from the calculated values and the delay in the arrival of echo signals from the focusing area to the sensor elements. This phenomenon is called “phase aberration” (Yue Li, Phase Aberration Correction Using Near-Field Signal Redundancy Principles. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 44, no. 2, March 1997, pp. 355- 371), leads to a decrease in the level of the main lobe of the radiation pattern and to the appearance of additional false side lobes. As a result, the spatial and contrast resolution deteriorates, the signal-to-noise ratio decreases.

Also known from the prior art is a diagram-forming device for acquiring ultrasound images, including a multi-element ultrasonic sensor, an excitation pulse generation unit, an analog receiver and a plurality of receiving channels (see, for example, US 7,744,532, publ. 06/29/2010).

In this technical solution, an adaptive focusing method is proposed that provides compensation for the focusing error due to deviations of the propagation time of the ultrasound from the calculated value. To evaluate the relative signal delay caused by aberration, the cross-correlation function of echoes from neighboring sensor elements is used. After forming an estimate of the relative signal delay caused by aberration, a correction of the signal delays in the receiving channels is performed.

However, the disadvantage of this device is the high demands on the performance of the computing device and the amount of RAM for implementing the adaptive algorithm for focusing the ultrasonic beam in real time. In addition, the disadvantages of the cross-correlation method for measuring aberrations can also be attributed to the insufficient accuracy of finding the maximum of the cross-correlation function even when using high-order interpolation filters.

The task to which the claimed technical solution is directed is to reduce hardware costs for the implementation of adaptive focusing of an ultrasonic beam.

This problem is solved due to the fact that in the inventive adaptive diagram-forming device for obtaining ultrasound images, including a multi-element ultrasonic sensor, a unit for generating excitation pulses, an analog receiver and a plurality of receiving channels, according to a utility model, that a multi-element ultrasonic sensor is connected to the output of the switch connected to the pulse shaping unit and an analog receiver, which is connected by its outputs to the corresponding input of one of the receiving channels s, each of which contains a digital delay line, which includes serially connected buffer memory and a filter interpolator connected to a multiplier, wherein the central receiving channel contains a phase shifter, and each of the other receiving channels contains a phase shifter, including a quadrature demodulator, which includes multipliers, the outputs of which are connected to the inputs of the low-pass filters, the outputs of which are connected to the inputs of the argument calculator, and a phase delay corrector comprising a subtractor, a time delay driver and an adder in series, the device comprising a reference frequency generator, a signal delay generator and an adder common to all receiving channels.

The excitation pulse generation unit 3 is an excitation pulse shaper with a plurality of palsers connected to it.

An analog receiver is an array of analog front-ends, each of which contains a low-noise amplifier, an amplifier with a time-varying gain, an analog-to-digital converter and an interface module for low-voltage differential signal transmission.

The technical result provided by the given set of features is to reduce hardware costs for implementing adaptive focusing, which affects the quality of the generated ultrasound image due to the implementation by the claimed device of a simple, computationally efficient phase correction method. As a result, it becomes possible to build a small-sized adaptive DFU and a portable ultrasound scanner based on it with the ability to compensate for phase aberrations.

The essence of the claimed device is illustrated by drawings, not covering and, moreover, not limiting the scope of claims for this decision, but only being illustrative materials of a particular case of the device, which depicts:

in FIG. 1 is a block diagram of a device; FIG. 2 is a block diagram of a phase shift driver; FIG. 3 is a block diagram of a phase delay corrector of a signal, FIG. 4 is an example of alignment of a point source for estimating phase shifts due to aberrations.

An adaptive chart-forming device for obtaining ultrasound images (see Fig. 1) includes a multi-element ultrasonic (US) sensor 1 containing P elements connected to the output of the switch 2 of the elements of a multi-element ultrasonic sensor 1 connected to a pulse generating unit 3. An analog receiver 4 is connected to the switch 2 channel-by-channel, which in turn is connected by its outputs to the corresponding input of one of the receiving channels 5 with numbers (1), ..., (N). Each receiving channel contains a digital delay line 6, which includes a series-connected buffer memory (CCD) 7 and a filter interpolator 8 connected to a multiplier 9. The central receiving channel 5 with the number N / 2 contains a phase shift former 10, and each of the remaining receiving channels 5 with numbers (1), ..., (N / 2-1), (N / 2 + 1), ..., (N) contains a phase shifter 10 and a phase corrector 11 of the signal delay.

Shaper 10 phase shift (see Fig. 2) contains a quadrature demodulator 12, which includes multipliers 13,14 outputs of which are connected to the inputs of low-pass filters (LPF) 15, 16, the outputs of which are connected to the inputs of the calculator 17 of the argument.

The phase corrector 11 of the signal delay (see Fig. 3) contains a series-connected subtractor 18, a shaper 19 of the time delay of the signal and the adder 20.

The device contains a common for all receiving channels 5 generator 21 of the reference frequency, the generator 22 of the signal delay and the adder 23.

In a preferred embodiment of the device, the excitation pulse generation unit 3 is an excitation pulse generator 24 with a plurality of palsers 25 connected to it. The inputs of the N palsers 25 with numbers (1), ..., (N) are connected to the outputs of the excitation pulse generator 24.

The analog receiver 4 is an array of analog front-end 26 with numbers (1), ..., (N), each of which contains a low-noise amplifier 27, an amplifier 28 with a time-varying gain, an analog-to-digital converter 29, and module 30 LVDS interface - Low Voltage Differential Signaling. The analog receiver 4 is connected to the receiving channels 5 outputs of the front-end 26 with numbers (1), ..., (N), which are connected to the corresponding inputs of the N receiving channels 5 with numbers (1), ..., (N).

The device operates as follows.

The device is based on the principle of ultrasonic echolocation. In a homogeneous medium, ultrasonic waves propagate rectilinearly and at a constant speed. At the boundary of media with unequal acoustic density, part of the energy of the emitted ultrasound signal is reflected, and part is refracted, continuing rectilinear propagation. The higher the gradient of the difference in acoustic density of the boundary media, the greater part of the ultrasonic vibrations is reflected. Received echoes are the basis for obtaining ultrasound images.

A multi-element ultrasonic sensor 1 contains P piezoelectric elements, where the value of P in practical cases usually varies from 64 to 194 elements. Block 3 of the formation of excitation pulses generates sounding ultrasonic signals that are sent to a multi-element ultrasonic sensor 1. The multi-element ultrasonic sensor 1 acts as a transmitter of mechanical vibrations in the tissue when a probe signal is emitted, and, accordingly, a receiver of echo signals from the boundaries of tissues and blood flow in the vessels. The echo signals recorded by the active elements of the multi-element ultrasonic sensor 1, through the switch 2 are sent to the analog receiver 4. Due to the switching device groups of elements of the multi-element ultrasonic sensor 1 performed by the switch 2 (selection of active N elements), the ultrasound beam is moved in the sounding plane, and as a result, a two-dimensional ultrasound image is formed, updated in real time.

The analog receiver 4 performs pre-filtering and amplification of the echo signal, as well as analog-to-digital signal conversion. The sequence of digital samples of the echo signal is transmitted to the digital delay line 11 of each receiving channel 10 with numbers (1), ... (N).

такта работы аналогового приемника с помощью фильтра-интерполятора 8. Для обеспечения высокой пропускной способности приемного тракта фильтр-интерполятор 8 представляет собой набор из четырех параллельных фазосдвигающих фильтров, каждый из которых выполняет дробную задержку сигнала на определенную величину, а коммутатор выходов (на чертежах условно не показано) осуществляет выбор одного из этих фильтров. Тонкая задержка в процессе приема эхо-сигнала выполняется при этом путем выбора номера фазосдвигающиего фильтра.The digital delay line 6 implements the delay of the echo signal by a predetermined amount in accordance with the generated focus on the reception. The digital delay line 6 consists of a buffer memory 7, which performs a coarse delay of the echo signal with an accuracy of 20 ns and a filter-interpolator 8, which implements a thin delay of the echo signal with an accuracy of 5 ns. A rough delay of the echo signal is provided by delaying in the buffer memory 7 the read address relative to the write address by a value proportional to the required number of samples. Thin (fractional) signal delay is accurate to

the operating cycle of the analog receiver using the filter interpolator 8. To ensure high throughput of the receiving path, the filter interpolator 8 is a set of four parallel phase-shifting filters, each of which performs fractional delay of the signal by a certain amount, and the output switch (conditionally not shown) selects one of these filters. A thin delay in the process of receiving the echo signal is performed by selecting the phase-shifting filter number.

Channel attenuation of the signal, implemented using the multiplier 9, is used to apodize the ultrasound beam. The value of the coefficient by which the data is multiplied is determined by the type of the required amplitude distribution of the receiving aperture.

To implement dynamic focusing on reception, the delays are calculated in real time by the signal delay generator 22 using the CORDIC processor included in it, which forms the difference between the distance from this element to the focal point and the distance to the focal point (beam vector length). The CORDIC method (Coordinate Rotation Digital Computer) - the method of iterative vector rotation (IPV) belongs to the class of digit-by-digit algorithms in which the calculation of any elementary function is based on iterative rotation of the vector. The IPV algorithm is effective for implementing computations in a programmable logic integrated circuit (FPGA), since it does not require large amounts of RAM and permanent memory. Namely, operational and read-only memory are the most expensive FPGA resources.

In accordance with the IPV algorithm, at each ith iteration, the rotation angle has a value equal to:

and the new coordinates and the angular position of the vector are given by recurrence formulas:

принимает значение +1 или -1, в зависимости от направления вращения вектора.Where

takes the value +1 or -1, depending on the direction of rotation of the vector.

There are two options for the direction of rotation of the vector. The first option, called VECTORING, corresponds to the case when the coordinates of the original vector (X, Y) are known, and the module and argument of the vector (R, ϕ) are determined during the calculation. In the second option, called ROTATION, the coordinates of the original vector (X, Y) and the necessary rotation angle ϕ are initially known, and in the process of calculation new coordinates of the vector (X ', Y') are determined. In the device, the direction of rotation of VECTORING is used to implement the multiplication of the signal by a complex exponent, and the direction of rotation of ROTATION is used, respectively, to generate the argument (phase) of the echo signal in each nth receiving channel 5, and to form the length of the vector performed by the CORDIC processor in the generator 22 signal delays.

Since the discreteness of the data arriving at the frequency of the analog receiver 4 is not sufficient for the accuracy of setting the delays, and each digital delay line 6 additionally includes a filter interpolator 8, which increases the accuracy of setting the delay by 4 times, then, accordingly, the formation of signal delays by the generator 22 signal delays are made with quadruple accuracy, i.e. with an accuracy of 5 not. In this case, the two least significant binary bits of the output value generated by the signal delay generator 22 control the choice of the phase-shifting filter number in the filter-interpolator 8.

A focused echo signal is generated at the output of the adder 23 after summing the signals of all N receiving channels 5.

The task of the phase aberration compensation algorithm is to provide in-phase addition of the signals of all receiving channels 5 in the adder 23 under conditions of tissue heterogeneity for the propagation of ultrasound. In accordance with the developed algorithm, the compensation of phase aberrations in each receiving channel 5 is performed by measuring the spread of the phase of the ultrasonic signal (phase error) relative to the reference value and its subsequent elimination using phase corrections 11 of the signal delay.

To isolate the phase error due to phase aberrations, the echo signal of each n-th receiving channel 5 after the preliminary filtering, amplification, and digitalization operations performed in the analog receiver 4 is supplied to the phase shifter 14.

The phase correction method is as follows. In phase shaper 10, in-phase I and quadrature Q components of the received echo signal are formed using quadrature demodulator 12 (see Fig. 2). In the quadrature demodulator 12, the echo signal is multiplied with the help of multipliers 13, 14 by trigonometric functions coming from the reference frequency generator 21:

where F _C is the value of the reference frequency, which corresponds to the center frequency of the spectrum of the echo signal.

Further, high-frequency components are suppressed outside the useful signal band using low-pass filters 15, 16. In order to obtain stable, statistically stable phase characteristics of the signal of low-pass filters 15, 16, a narrow part of the spectrum is isolated in the region of the central frequency of the echo signal and can be implemented as a moving average filter. A significant advantage of this filter compared to the standard low-pass filter is that its implementation is associated with a simple summation of a fragment of a sequence of samples. Accordingly, it is not necessary to perform labor-intensive operations of multiplying the signal samples by filter coefficients (all filter coefficients are equal to unity).

The phase difference is measured at the center frequency of the echo spectrum. Since due to the characteristics of the attenuation of ultrasound in tissues with increasing reception depth, the spectrum of the echo signal is shifted to the low frequency region, to compensate for the shift of the spectrum of the signal along the depth of sounding, the reference frequency generator 21 changes the value of the frequency F _C in the process of receiving echoes with different depths.

After low-pass filtering, the calculator 17 of the argument calculates the phase of the signal. To effectively implement this calculation, the CORDIC method is used. Next, the echo signal is supplied to the adaptive phase corrector 11 signal delay, which performs the compensation of phase aberrations by equalizing the delays in all receiving channels relative to the Central receiving channel 5 under the number N / 2.

Corrective delay for any n-th receiving channel 5, where n = 1, ..., N / 2-1, N / 2 + 1. ... N, and N is the number of receiving channels 5 used to form the radiation pattern at a given sounding depth, is determined based on the calculation of the phase difference between the signal of the nth receiving channel 5 and the signal of the central receiving channel 5 with the number N / 2 subtractor 18 included in the composition of the phase corrector 11 signal delay (see Fig 3):

на любой глубине равна полпериода частоты генератора 21 опорной частоты, т.е. ±1/2F_C. Так для значения опорной частоты F_C = 3.5 МГц

Этой величины вполне достаточно, поскольку как показывают многочисленные эксперименты, разброс задержек из-за фазовых аберраций в тканях составляет в пределах от единиц и до нескольких десятков нс.Obviously, in this case, for the central receiving channel 5 under the number N / 2, the correction delay is always equal to zero. Therefore, in the Central receiving channel 5 under the number N / 2 there is no phase corrector 11 of the signal delay (see Fig. 1). Maximum possible amount of delay to compensate for phase aberrations

at any depth is equal to half a period of the frequency of the reference frequency generator 21, i.e. ± 1 / 2F _C. So for the value of the reference frequency F _C = 3.5 MHz

This value is quite sufficient, since numerous experiments show that the spread of delays due to phase aberrations in tissues ranges from units to several tens of ns.

путем вычисления по формуле:Shaper 19 of the time delay of the signal translates the value of the phase difference Φ _n in the value of the time shift

by calculation by the formula:

or by table conversion.

формируемого генератором 22 задержек сигнала, и сформированного формирователем 19 временной задержки сигнала корректирующего значения

Значение задержки с выхода сумматора 20 поступает на цифровую линию 6 задержки.The adder 20 generates the current value of the delay of the echo signal in the receiving path, taking into account the phase spread due to aberrations by adding the theoretically calculated delay value

the signal delay generated by the generator 22 of the signal delay, and the correction value generated by the driver 19 of the time delay of the signal

The delay value from the output of the adder 20 is supplied to the digital delay line 6.

In a preferred embodiment of the device, the formation of sounding ultrasonic signals in the block 3 of the formation of excitation pulses is as follows. To generate probing ultrasonic signals, the parser 25 excitation pulses 25 included in block 3 are generated under the numbers (1), ..., (N), which are excited by a sequence of positive and negative pulses of different durations (PWM method - pulse-width modulation) received from the shaper 24 excitation pulses. By changing the duration of the excitation pulses, the required amplitude of the excitation signal of the multi-element ultrasonic sensor 1 is formed. Changing the excitation amplitude is necessary to perform apodization to radiation. When the bell-shaped apodization function of the emitted aperture is realized, the pulse duration gradually increases from a certain minimum value at the edges of the aperture to a maximum value at the center of the aperture. The distance between pulses is 1 / 2F ₀ , where F ₀ is the fundamental frequency of the radiation.

Transmission focus is static. The position of the focus on radiation along the depth of sounding is controlled by the operator. In this case, the delay pulses of the excitation of the multi-element ultrasonic sensor 1, necessary to create focus at a certain depth of sounding, are calculated for each selected focus position. Further, in accordance with the set delays in the driver 24 of the excitation pulses, the elements of the multi-element ultrasonic sensor 1 are excited at different times. When changing the focus position to radiation, the delay is recalculated.

Analog receiver 4 operates as follows. The echo signals recorded by the active elements of the multi-element ultrasonic sensor 1, through the switch 2 enter the analog receiver 4 at the inputs of the corresponding analog front-end 26 with numbers (1), ..., (N), which are part of the analog receiver 4. In each analog front end 26 pre-filters and amplifies the echo signal using a low-noise amplifier 27 and amplifier 28 with a time-varying gain. Next, analog-to-digital conversion is performed using a high-speed analog-to-digital converter (ADC) 29, which provides the conversion of the echo signal into a digital form with a sampling frequency of the order of 50 MHz and the bit depth of the conversion is not less than 12 bits. The sequence of digital samples of the echo signal through the module 30 of the LVDS interface is transmitted to the digital delay line 6 of each receiving channel 5 with numbers (1), ... (N).

To quantify the phase shifts caused by aberrations, an adjustment point source of coherent radiation is required, which is placed inside the tissues under study, which is not possible.

As such a source, it is proposed in practice to use a bright reflecting point element of tissues located in the region of the focus on radiation (see, for example, US 6,368,279, publ. 09.04.2002). In FIG. 4 illustrates the selection of an alignment point source for estimating phase shifts due to aberrations. In order for only one point reflecting object 33 to enter the focusing zone 32, the ultrasonic beam 31 must be sufficiently narrow and have a sharp focus with a small depth of field.

A bright point-like reflecting object 33 forms a spherical wave 34. And if the propagation medium is homogeneous, then the phase shifts between the echo signals arriving at different elements of the multi-element ultrasonic sensor 1 are fixed and are determined by the distances from the point source to the elements of the multi-element ultrasonic sensor 1. shifts between echoes can be calculated based on the laws of geometric optics. In the presence of tissue inhomogeneities, a phase spread occurs, which is extracted in each receiving channel 5 using a phase shift former 10 and is taken into account further when generating the current signal delay value in the signal delay phase corrector 11.

The choice of a point reflective object 33 can be implemented either manually by an operator or automatically (see, for example, United States Patent 2006/0106309 A1, publ. 05/18/2006). In the case of the manual method, the operator visually determines a bright point object in the ultrasound image and sets the corresponding marker on it. During automatic search, the pixels of the ultrasound image are analyzed in the zone of interest, and the image area is located, which has the brightest peak, and around it the signal level is below a certain threshold.

It should also be noted that a separate sensing mode is not used to evaluate the phase aberrations and their compensation by the device. These operations are performed directly during the formation of ultrasound images of the organs under investigation.

This technical solution significantly reduces the requirements for the performance of the computing device and the amount of RAM for the implementation of the adaptive algorithm for focusing the ultrasonic beam in real time.

Claims (3)

1. An adaptive diagram-forming device for obtaining ultrasound images, including a multi-element ultrasonic sensor, an excitation pulse generation unit, an analog receiver and a plurality of receiving channels, characterized in that the multi-element ultrasonic sensor is connected to an output of a switch connected to a pulse forming unit and an analog receiver which is connected by its outputs to the corresponding input of one of the receiving channels, each of which contains a digital delay line, in becoming a series-connected buffer memory and a filter interpolator connected to a multiplier, the central receiving channel containing a phase shifter and each of the other receiving channels containing a phase shifter including a quadrature demodulator, which includes multipliers whose outputs are connected with inputs of low-pass filters, the outputs of which are connected to the inputs of the calculator of the argument, and a phase delay corrector containing sequentially a single subtractor, a time delay driver and an adder, the device comprising a reference frequency generator, a signal delay generator and an adder common to all receiving channels. 2. The adaptive diagram-forming device according to claim 1, characterized in that the excitation pulse generation unit is an excitation pulse generator with a plurality of palsers connected to it. 3. The adaptive diagram-forming device according to claim 1, characterized in that the analog receiver is an array of analog front-ends, each of which contains a low-noise amplifier, an amplifier with a time-varying gain, an analog-to-digital converter and an interface module low voltage differential signaling.

Images