CN1242079A - Ultrasound scan conversion with spatial dithering - Google Patents

Abstract

An ultrasound imaging system includes a scan conversion process for converting ultrasound data into a standard display format conversion and can be performed on a personal computer by programming the computer to convert data from polar coordinates to cartesian coordinates suitable for display on a computer monitor. The data is provided from scan head enclosure that houses an array of ultrasonic transducers and the circuitry associated therewith, including pulse synchronizer circuitry used in the transmit mode for transmission of ultrasonic pulses and beam forming circuitry used in the receive mode to dynamically focus reflected ultrasonic signals returning from the region of interest being imaged.

Description

The ultrasound scan conversion that carrying space is shivered

Relevant patented claim

This is the part continuation application of the International Patent Application PCT/US96/11166 of application on June 28th, 1996, wherein said International Patent Application PCT/US96/11166 is the U.S. Patent application the 08/599th of application on February 12nd, 1996, No. 816 part continuation application, and this part U.S. Patent application is the U.S. Patent application the 08/496th of application on June 29 nineteen ninety-five, No. 804 and the 08/496th, No. 805 part continuation application is incorporated the full content of above-mentioned patented claim into this paper at this by quoting as proof.

Prior art of the present invention

Conventional ultrasonography system generally includes by cable and the hand-hold scanning head that control desk is handled and display unit is coupled that is installed on the big support.This scanner head generally includes ultrasound transducer array, this transducer array with ultrasonic energy emissions to just in the zone of imaging and receive the ultrasonic energy returning from this zone.These transducers convert the ultrasonic energy of receiving to low level electric signal, and this electric signal sends processing unit to by cable.The suitable beam forming technique (as dynamic focusing) of processing unit utilization will merge from the signal of transducer, so that produce the image in perception interest zone.

Typical conventional ultrasonic system comprises having for example 128 transducer array of a large amount of ultrasonic transducers.Each transducer all links to each other with its treatment circuit system, and this Circuits System is positioned in the control desk processing unit.This treatment circuit system generally includes driving circuit, and this driving circuit sends to transducer so that start the emission of ultrasonic signal with the driving pulse of accurate timing in emission mode.These timing transponder pulses send to scanner head from the control desk processing unit along cable.In receiving mode, the beam shaping circuit is introduced each low level electric signal from transducer with suitable delay, and these signals are carried out dynamic focusing, so that can produce exact image subsequently.

For phase-array scanning head or crooked linear sweep head, (r θ) is received and digitizing ultrasonic signal with its intrinsic polar form.With regard to demonstration, this representation is inconvenient, thus for further handle with it be transformed into the rectangular coordinate representation (x, y).This rectangular coordinate representation is to carry out figure adjustment at the dynamic range of various displays and hard copy device and brightness.Can also store and recover these data, so that show once more.Polar coordinates (r, θ) with rectangular coordinate (x, when carrying out conversion between y), must be by (r, θ) value calculates that (x, y) value is because in that (r, θ) point in the array is with (x, y) point on is inconsistent in normal grid.

In early days in the scan conversion system, visit (x, y) point of each on the grid square and utilize corresponding to the numerical value of two θ values of the most contiguous this point by linear interpolation calculate the numerical value of this point or utilization (r, θ) in the array numerical value of four the most contiguous points by secondary linear interpolation calculate the numerical value of this point.The method of finishing this work is to utilize finite state machine to produce (x, y) traversing graph, utilize bidirectional shift register with (r, θ) the data increment is kept in a large amount of Digital Logic and memory cell, so that control the correct (r of this process and the asynchronous reception of assurance, θ) (x, orthochronous y) arrives so that interpolation the data increment being fit to each.This early stage instrument may be inflexible and be unnecessarily complicated.Although a large amount of control hardwares is arranged, only has one by (x, y) path of array.

General introduction of the present invention

In a preferred embodiment of the invention, the scan-data that will be transformed into display format after finishing beam shaping and scan conversion directly sends computing machine to.In preferred embodiments, scan conversion can utilize the software on the personal computer to finish fully.In addition, can insert the circuit board of band additional firmware, so that selected scan conversion function is provided or finishes whole scan conversion processing.With regard to many application, software systems are preferred, because additional firmware can be minimized like this, so personal computer can be a portable stage, for example laptop computer or palmtop computer.

Scan conversion preferentially adopts the spatial dithering treatment technology to finish, below with more detailed introduction.Spatial dithering will be simplified the calculation requirement to scan conversion, simultaneously resolution of Bao Chiing and quality.Therefore, scan conversion does not need more complicated interpositioning just can finish on personal computer, and still can finish conversion with the frame speed that is fit to the real-time ultrasound imaging.

Scan conversion program preferentially comprise the input array, again shine upon array and output array.Shining upon array again is that an array of indexes is an array of pointers, and this array is the size of output image, and it is used for determining where to obtain each pixel from the input array.Shining upon the extracting position of each pixel in the input array that each locational numeral in the array points out to enter with same position the output array again.Therefore, can think that shining upon array again has identical geometric configuration with the output array, and the input array has identical data type with the output array, i.e. Shi Ji pictorial data.

The input array has the new data of every frame ultrasonic image, this means that it handles these data and these data are put in the output array frame by frame.According to the preferred embodiments of the invention, the new ultrasonic image of a frame was arranged approximately in per 1/30 second.So as long as the customary computing of calculating new output image with new input data set can finish with the frame speed of about 30 frames of per second, the data of shining upon array more just can generate (still still should about below 1 second) more lentamente.To allow to use conventional personal computer to finish the task that the data of array are shone upon in generation more like this, and can not sacrifice combination property, also needn't finish this task with additional firmware.In the computing system with digital signal processor (DSP), DSP can finish the calculating of shining upon array again.

In addition, some scan conversion function can be finished with hardware, and this hardware is installed on the circuit board and is inserted in the personal computer.The sort circuit plate be plug-in unit be can plug and be used as interface so that send data to the PC bus controller by rights.

Brief description of drawings

As what be described in the drawings, by following the preferred embodiments of the invention are introduced more specifically, clear and definite more above-mentioned target of the present invention and other target and the features and advantages of the present invention.Identical in the accompanying drawings reference symbol refers to same parts all the time in different accompanying drawings.These accompanying drawings not necessarily draw to scale, but it is in principle of the present invention to focus on explanation.

Fig. 1 is the block scheme of the imaging array of the routine used in the ultrasonography system.

Fig. 2 A illustrates the linear ultrasound transducer array and the synoptic diagram of the relation between the rectangular scanning district according to the present invention.

Fig. 2 B is the synoptic diagram that the relation between crooked linear ultrasonic transducer array and the crooked scanning area is described according to the present invention.

Fig. 2 C illustrates the linear ultrasound transducer array and the synoptic diagram of the relation between the keystone scanning district according to the present invention.

Fig. 2 D is the synoptic diagram in explanation phase-array scanning district.

Fig. 3 is the synoptic diagram of ultrasonography optimum system choosing embodiment of the present invention.

Fig. 4 A is the functional-block diagram of ultrasonography optimum system choosing embodiment of the present invention.

Fig. 4 B is the functional-block diagram of another preferred embodiment of ultrasonography of the present invention system.

Fig. 5 A is the synoptic diagram according to beam shaping of the present invention and filtering circuit.

Fig. 5 B is the synoptic diagram according to another preferred embodiment of beam shaping of the present invention and filtering circuit.

Fig. 5 C is the synoptic diagram according to another preferred embodiment of beam shaping of the present invention and filtering circuit.

Fig. 5 D is the synoptic diagram according to low-pass filter of the present invention.

Fig. 5 E is an example according to interface card of the present invention.

Fig. 5 F is the preferred embodiment according to beam shaping integrated circuit of the present invention.

Fig. 6 is the graphic extension according to the passband of wave filter of the present invention.

Fig. 7 A is the synoptic diagram that covers the input point in the demonstration.

Fig. 7 B has the synoptic diagram that shows shown in Fig. 7 A of the input data that are transformed into pixel.

Fig. 8 is the general image preferred embodiment synoptic diagram of mapping architecture again.

Fig. 9 A and Fig. 9 B shine upon the process flow diagram of array computing technique again according to the present invention's explanation.

Figure 10 is the process flow diagram of output frame computing engines.

Figure 11 A and Figure 11 B are the synoptic diagram of use in ultrasonography of the present invention system the two kinds demonstration expression formula forms that can be selected by the user.

Figure 12 is the functional-block diagram of preferred graphical interface of user.

Figure 13 explanation is used for the dialog box of ultrasonography control.

The display box of Figure 14 A to Figure 14 D explanation input system information.

The additional dialog box of Figure 15 A to Figure 15 C explanation input probe or visual field data.

Figure 15 D to Figure 15 J explanation is used for the additional demonstration and the dialog box of the preferred embodiments of the invention.

Figure 16 illustrates the imaging and the display operation of the preferred embodiment of the invention.

The preferred embodiment of Figure 17 A to Figure 17 C explanation foundation integration probe of the present invention system.

Figure 18 illustrates the integrated manipulator of 64 channels of the emission/receiving circuit that is used for ultrasonic system.

Figure 19 explanation transmits and receives the another kind of preferred embodiment of circuit.

Figure 20 illustrates according to Doppler sonar system of the present invention.

Figure 21 explanation is based on the colored flow pattern according to Fast Fourier Transform (FFT) pulse Doppler disposal system of the present invention.

Figure 22 illustrates according to waveform of the present invention and generates disposal system.

Figure 23 is the system that is used for producing according to the present invention colored flow pattern.

Figure 24 is the processing flow chart that is used for calculating according to the present invention colored flow pattern.

Figure 25 adopts cross-correlation technique to produce the processing flow chart of colored flow pattern.

Detailed description of the present invention

The schematic block diagram of the imaging array 18 that the N that uses in a ultrasonography system piezoelectric type ultrasonic transducer 18 (1)-18 (N) is formed is shown in Fig. 1.This piezoelectric transducer element array 18 (1)-18 (N) produces the sound pulse of propagating with narrow beam 180 in imaging target (normally certain zone of tissue) or propagation medium.These pulses are to propagate with the speed of constant as spherical wave 185.Come the echo of selfimaging point Ip (being reflecting body) to be detected and can show by same transducer element array 18 or another receiving array in a kind of mode of reflection configuration position of indicating with the return signal form.

In propagation medium, come the echo of selfimaging point Ip after the different travel-times, to arrive each element of transducer 18 (1)-18 (N) of receiving array.The travel-time that arrives each element of transducer is different, and this travel-time is depended on the distance between each element of transducer and the one-tenth image point Ip.This is that soft tissue is correct for typical transonic medium, is more constant at least in this occasion velocity of sound.After this, the information of receiving will indicate the mode of reflection configuration position to show with certain.

In the B-of two dimension mode scanning, these pulses can be launched along a large amount of sight lines, as shown in Figure 1.If echo is sampled and their amplitude is encoded as brightness, can on cathode ray tube (CRT) or monitor, gray scale image be shown so.One frame image comprises 128 this sweep traces that form 90 ° of fan-shaped images with 0.75 ° of angular distance usually.Because the velocity of sound in the water is 1.54 * 10 ⁵Cm/sec, for time of the degree of depth round trip of 16cm will be 208 μ s.Therefore, the T.T. along 128 sight lines (frame image) acquisition data is 26.6ms.If the data that the speed of other signal processor is enough to keep such in the system are obtained rate, can produce two-dimensional image with the speed that is equivalent to the standard television image so.For example, if USI is used for observing sound wave reflection or backward scattering by the wall of the chest between a pair of rib, heartthrob can real time imaging so.

With regard to height pattern fix and horizontal directivity pattern mainly by postponing to turn to the turning to for the array of control, ultrasonic transmitter is the linear array of a piezoelectric transducer 18 (1)-18 (N) (separating half wavelength usually) normally.It mainly is to apply the delay transponder pulse for each element of transducer 18 (1)-18 (N) by such mode to be controlled that the beam radiation of conventional arrays (orientation) is schemed, so that is becoming image point Ip to be added to the beam shape that generation together needs from the energy of all transmitters.So, need a delay circuit to link to each other, so that produce required radiation diagrams along predetermined direction with each element of transducer 18 (1)-18 (N).

As mentioned above, same element of transducer 18 (1)-18 (N) array 18 can be used to receive return signal.The waveform that comes from into reflection (the being echo) beam energy of image point arrives each element of transducer after time-delay, wherein said time-delay equal from become image point to the distance of element of transducer divided by supposing acoustic wave propagation velocity constant medium.Similar with emission mode, this time-delay is different to each element of transducer.Should compensate these difference of path at each receiving transducer element place, its method is that the reflected energy from the concrete one-tenth image point of any given depth that each receiver is received is focused.The delay of each receiving element all be from this element to array center measured distance and perpendicular to the function of the viewing angle direction of this array actual measurement.

Beam shaping operation and focusing operation all relate to formation by the observed scattering waveform of all transducers and, still, when carrying out this stack, these waveforms must be by differential delay so that they all arrive at and must weighting when superposeing by phase place.Therefore, need different delays to be added to the beam shaping circuit that also can change that delay on each channel at any time.Echo along assigned direction when darker tissue returns, receiving array will continuously change its focus along with change in depth.This method is called as dynamic focusing.

After forming received beam, in a usual manner with its digitizing.The numeral expression formula of each pulse of receiving all is a time series, and this time series and the position angle that forms with wave beam as distance function are corresponding from the backward scattering xsect of the ultrasonic energy that the field measurement point returns.Continuous impulse points to different directions, covers the visual field from-45 ° to+45 °.In some system, improve image quality with the time average (being referred to as the persistence weighting) of calculating the continuous observed data that obtains in same point.

Fig. 2 A to Fig. 2 D illustrates the various transducer arrays configurations of use in the present invention and the synoptic diagram of the relation between their corresponding scanned imagery zones.Fig. 2 A explanation produces the linear array 18A of rectangular scanning imaging district 180A.Such array generally includes 128 transducers.

Fig. 2 B is the synoptic diagram of the relation between the scanned imagery district 180B of crooked linear transducer arrays 18B of explanation and the bending of resulting cross section.Array 18B also comprises 128 transducers usually.

Fig. 2 C represents the relation between linear transducer arrays 18C and the keystone scanning imaging district 180C.The transducer that normally adjoined by 192 (rather than 128) of array 18C constitutes in this embodiment.This linear array is used to produce keystone scanning district 180C, and its method is that linear sweep and the phase-array scanning shown in Fig. 2 A combined.In one embodiment, 64 transducers on array 18C opposite end are used in the phased array configuration, so that form the arc triangle edge part at 180C area edge place.64 middle transducers use in linear scan pattern, so that finish the rectangle part in 180C district.Therefore, step sector 180C utilizes little aperture scanning method (sub-aperture scaning approach) to realize it is effective whenever all having only 64 transducers in this method.In one embodiment, alternately activate by what 64 transducers were formed and adjoin transducer group.In other words, at first 1 to No. 64 transducer is activated.Next, 64 to No. 128 transducers are activated.Next step activates 2 to No. 65 transducers, activates transducer then 65 to No. 129.This scheme is proceeded, till 128 to No. 192 transducers are activated.Next, since 1 to No. 64 transducer next round scanning process again and again.

The short battle array of the linearity 18D of Fig. 2 D explanation transducer should the short battle array of linearity will be used to finish the phased array imaging according to the present invention.Handle use in order to produce fan section 180D linear array 18D through beam steering.

Fig. 3 is the synoptic diagram of ultrasonography of the present invention system 10.The hand-hold scanning head 12 that this system comprises with portable data is handled and display unit 14 couplings join, wherein portable data processing and display unit can be portable computers.In addition, data processing and display unit 14 can comprise personal computer or other computing machine that docks with the CRT that is fit to the demonstration ultrasonic image.Data processor display unit 14 can also be the monolithic cell of small volume and less weight, this unit be small enough to allow the user hand-held or be worn on one's body.Though what Fig. 3 showed is the external scanner head, scanner head of the present invention also can be to be fit to insert the built-in scan head of health in the inside of human body imaging by body cavity.For example, this scanner head can be the esophagus probe that is used for the heart imaging.

Scanner head 12 links to each other with data processor 14 by cable 16.In alternate embodiment, system 10 comprises an interface unit 13 (dotting) that plays coupling between scanner head 12 and data processing and display unit 14.Interface unit 13 preferably comprises controller and comprises the treatment circuit system of digital signal processor (DSP).Interface unit 13 can be finished the signal processing tasks that needs and signal output can be offered data processing unit 14 and/or scanner head 12.For user with palmtop computer, interface unit 13 preferred inner card or chipsets.When using desk-top computer or laptop computer, interface unit 13 can be used as external unit.

Hand-held housing 12 comprises transducer portion 15A and handle portion 15B.Transducer portion 15A remains below under 41 ℃ the temperature, so that the temperature of the housing parts that contacts with patient skin is no more than this temperature.The temperature of handle portion 15B is no more than second higher temperature, preferred 50 ℃.

Fig. 4 A is the functional-block diagram of an embodiment of ultrasonography of the present invention system 10.As shown in the figure, scanner head 12 comprises ultrasound transducer array 18, and to be transmitted into perceptual interesting zone be in the imaging target 11 (as certain zone in the tissue) to this array and receive the reflected ultrasonic of returning from the imaging target with ultrasonic signal.Scanner head 12 also comprises transducer driving circuit system 20 and pulse synchronization circuit system 22.Pulse synchronization circuit 22 sends to high-voltage driving circuit 20 with a series of accurate timing pips and delayed pulse.When driving circuit 20 was received each pulse, this high-voltage driving circuit was activated, and high-voltage driven signal is sent to each transducer in the transducer array 18, activates this transducer ultrasonic signal is transmitted in the imaging target 11.

The ultrasonic echo of imaging target 11 reflections is surveyed by the ultrasonic transducer in the array 18.Each transducer is transformed into representational electric signal with the ultrasonic signal of receiving, this electric signal is sent to pre-amplification circuit 24 and time-varied gain control (TGC) Circuits System 25.Pre-amplification circuit system 24 will be arranged on the level that is fit to aftertreatment from the level of the electric signal of transducer array 18, and TGC Circuits System 25 is used to compensate the decay of sound pulse when passing tissue, and this Circuits System also drives the beam shaping circuit 26 (below will be introduced) that produces the line map.Electric signal through overregulating is sent to beam shaping Circuits System 26, and difference postpones to introduce each signal of receiving this Circuits System with suitable distinguishing, these signals is dynamically focused, so that can produce accurate image.In the International Patent Application PCT/US96/11166 that incorporates into, introduced in more detail and will distinguish that difference postpone to be introduced the signal received and used beam shaping Circuits System 26 and all delay circuits during the pulse signal that produces by pulse synchronization circuit system 22.

In a preferred embodiment, through dynamic focusing and add and signal be sent to A/D converter 27, this A/D converter will through adding and signal digitalized.Then, digitized signal data is delivered to color Doppler treatment circuit 36 from A/D converter 27 through cable 16.Should be noted that directly transmit by cable 16 through adding and the alternate embodiment of simulating signal in, do not use A/D converter 27.In data processing and display unit 14, digital signal through after the demodulation, is sent to the scanning conversion circuit 37 in data processing and the display unit 14 in demodulator circuit 28.

As shown in the figure, storer 29 storages of scanner head are from the data of control circuit 21 and data processing and display unit 14.Scanner head storer 29 offers pulse synchronization circuit system 22, TGC 25 and beam shaping Circuits System 26 with storage data.

Scanning conversion circuit system 37 will from the digitized signal data of beam shaping Circuits System 26 from polar coordinates (r, θ) be transformed into rectangular coordinate (x, y).After conversion, the rectangular coordinate data can be delivered to non-essential signal post-processing platform 30, there with its format, so that showing on the display 32 or compression in video compression circuit 34.Aftertreatment 30 also can utilize the scan conversion software of introducing below to finish.

The digital signal data that pulse Doppler processor in data processing unit 14 or continuous-wave doppler processor 36 receive from A/D transducer 27.Pulse Doppler processor or continuous-wave doppler processor 36 produce moving target tissue 11 (as the blood that flows) the used data of imaging.In preferred embodiments, handle the generation flow pattern with pulse Doppler.Pulse Doppler processor 36 will send scanning conversion circuit system 28 to through the data that it is handled, and the polar coordinates with data are translated into the rectangular coordinate that is fit to demonstration or video compress there.

The higher level operation of control circuit [inner microprocessor 38 forms of preferred personal computer (as desktop computer, laptop or palm machine)] control ultrasonography system 10.Microprocessor 38 or DSP will postpone and the scan conversion initialize memory.Control circuit 38 postpones by the difference of distinguishing of scanner head storer 27 control introducing pulse synchronization circuit systems 22 and beam shaping Circuits System 26.

Microprocessor 38 is gone back control store 40, the data that this memory storage scanning conversion circuit system 28 uses.Should be appreciated that storer 40 can be that single storer also can be the multi-memory circuit.Microprocessor 38 also has the interface that links to each other with video compression circuit system 34 with signal post-processing Circuits System 30, so that control their functions separately.Video compression circuit system 34 is with data compression, so that pictorial data is arrived distant station for showing and analyzing through transmission channel.Transmission channel can be cellular communications channel modulation or wireless or other known communication means.

Portable ultraphonic imaging system 10 preferred battery of the present invention 44 power supplies.The primary element driven stabilized voltage supply 46 of battery 44 outputs, this stabilized voltage supply will offer each subsystem of imaging system 10 through the power of adjusting, and be included in those subsystems in the scanner head 12.Therefore, the power of supply scanner head can be provided through cable 16 by data processing and display unit 14.

Fig. 4 B is the functional-block diagram of another preferred embodiment of ultrasonography of the present invention system.In improved scanner head 12 ', the demodulator circuit system is replaced by software, and this software will be carried out by the microprocessor in improved data processing and the display unit 14 ' 38.Specifically, will be from the digitized data stream of A/D converter 27 by FIFO storer 37 allocation buffers.The instruction of microprocessor executive software is so that finish demodulation, scan conversion, color Doppler processing, signal post-processing and video compress.Therefore the many hardware capabilitys among Fig. 4 A can be replaced by the software in the storer 40 that is stored in Fig. 4 B, thereby have reduced the requirement to the hardware size and the weight of system 10 '.

Fig. 5 A, 5B and 5C describe the additional preferred embodiment of the beam shaping Circuits System of ultrasonic system.Wherein each embodiment simulated data of all requiring to gather mixes down to baseband frequency or frequency conversion from intermediate frequency (IF).Frequency reducing or frequency conversion are finished by following method: at first sampled data be multiply by a complex value (with the complex value exponential representation of input multiplier), be converted to the image of near by frequency then through filtering filtering from data.The output of this processing is applicable to minimum output sampling rate and is applicable to that subsequently demonstration or Doppler handle.

In Fig. 5 A, catch data 54 with one group of sample circuit 56, this data increment on being produced on integrated circuit 50 based on representing with one group of electric charge in the treatment circuit of CCD.Data are put into one or more lag line and utilize storage and control circuit system 62 and programmable delay circuit 58 to output to non-essential interpolation filter 60 in due course.Interpolation filter can be used for provided accurate judgement and provide focusing preferably to the return signal from the sensitive element array whereby the two-way time of sound wave.Fig. 5 A only depicts two processing channel 52 in the processor array.Output from interpolation filter adds the merging with node 66 places in simulated data, so that the output data through beam shaping from this processor array is provided.

The data class that utilizes ultrasonic transducer to obtain is similar to the signal output with the bandwidth appropriateness of transducer centre frequency modulation.The centre frequency of transducer is that characteristic frequency and IF are of equal value.In the simulating signal sampling system, (for example utilize CCDs), Ω=2 π f _I/ f _S, f wherein _IBe intermediate frequency, f _SIt is sample frequency.Numerical value n corresponding to the increment sequence number (be n=0,1,2,3,4 ...).The output of multiplier 68 is called as homophase increment (I) or quadrature increment (Q).In general, the value of I and Q will not be zero.But, select IF to equal f _SThe output of/4 o'clock multipliers will by repetitive sequence I, Q ,-I ,-Q, I, Q ,-I ... only produce the value of I or the value of Q.In fact, the input data only convert by 1 and-1.Therefore, if the input data are sampled i.e. a[0 in order], a[1], a[2], a[3], a[4] ... a[n], output data is a[0 so], j ^*A[1] ,-a[2] ,-j ^*A[3], a[4] ..., a[n], output data is a[0], j ^*A[1] ,-a[2] ,-j ^*[3], a[4] ...

Each all passes through low- pass filtering 70,72 output 74,76 of I and Q, so that remove the signal pattern of sneaking into base band.The coefficient of low-pass filter can utilize lowest mean square criterion (LMS or L2-norm criterion) or Chebyshev's criterion (L-infinity norm criterion) to design.What suit the requirements in practice, is to reduce the number that obtains the predetermined necessary coefficient of filter characteristic as far as possible.

The example of the low-pass filter that provides with CCD is illustrated with Fig. 5 D.Device 90 is made up of the tapped delay line of 13 attitudes, implements filter factor by 5 fixed weighting multipliers 94.As what in the explanation of Fig. 6, can see, the fluctuation in the passband below 0.5dB, and the decay of inhibition zone less than full scale-30dB.

Then, divide sample by the factor 2 for the output of low-pass filter at least.Branch sample greater than 2 can prove that whether the bandwidth of ultrasonic signal is the limit bandwidth, whether is limited to half the degree that is far smaller than sample frequency.For most of ultrasonic signals, often adopt branch like factor greater than 2, be narrow because compare the bandwidth of ultrasonic signal with sample frequency.

In order to reduce the clock frequency of low-pass filter, can exchange the order of branch sample and low-pass filtering (device).By using bank of filters, can select the low-pass filtering coefficient of I and Q, so that each wave filter is only accepted the data at its input end every one.Selecting to divide the sample rate is 2 o'clock, and this " alternative clock " pattern allows to relax design limit.If divide like factor greater than 2 (be signal bandwidth≤≤ f _S/ 2 o'clock) these restrictions can further relax.

Output data through frequency reducing is proceeded further processing, and these processing can comprise that the detection of signal envelope or Doppler handle.In order to show, signal envelope (being also referred to as signal amplitude) is calculated as square root sum square of I and Q output.For IF=f _S/ 4 o'clock situation (promptly or the situation of I=0 or Q=0), the detection of envelope becomes insignificant.The input handled of Doppler often of the data of I and Q, Doppler handle also use signal envelope in case signal just and/or information extraction in the marginal frequency band.In Fig. 5 A, after the ultrasonic signal beam shaping, only need a frequency reducing level.

In Fig. 5 B, the frequency reducing level is placed in each processing channel 52 of sample circuit 56 back.As previously mentioned, can accurately finish the generation of I and Q data 86,88 in this case, and system speed is faster.The major advantage of this method is based on the bandwidth of ultrasonic signal and selects low-pass filtering and divide like factor the data transfer rate in every processing channel can be reduced to minimum.In this implementation, all processing channel 52 all will be used identical complex value multiplier and consistent coefficient and branch like factor in filtering stage.As above-mentioned implementation,, complex-valued data is postponed to handle and interpolation processing for the output of beam shaping is provided.

It is almost consistent that hyperacoustic processing in early stage and Fig. 5 B that describes in Fig. 5 C described.Difference is that interpolation level 85,87 has been cancelled, but selects unique value to replace with the complex value multiplier, so that the delay of computing channel more accurately.The deficiency of this method is I value and the Q value that the output of multiplier will always present non-zero.This is the result who changes sampling rate in complex plane figure round the unit circle of multiplier input.Therefore, this method can be calculated more accurately to the sampling delay in every channel, but all produce complex-valued data with the output terminal every processing channel is cost.This improvement is compared with above-mentioned implementation may need more aftertreatment to finish envelope and Doppler's detection.

Between the output terminal of beam shaping circuit or filtering circuit and computing machine, provide the preferred embodiment of the system of interface to should be computing machine plug-in unit (PCMCIA) is provided.

A kind of embodiment of circuit board 700 explanations shown in Fig. 5 E, in this embodiment, differential receiver 702 receives 16 numerical data from scanner head and the processing of process beam shaping on cable.Clock signal also is received at register 704 places with the differential data through conversion.First logic array 708 becomes 32 with half data transfer rate with 16 bit maps.32 bit data enter FIFO 712 through the clock gating, this FIFO stack output supplemental data (add-on data) 716.Second logic array 710 has the access port of full control signal and to the output terminal 714 of pci bus controller.This concrete example utilizes 16 data, but this design also is applicable to 32 or bits of data more.

In addition, can also use the groove that is fit to be inserted in personal computer, laptop computer or palmtop computer or mouthful in plug-in unit.In this embodiment, differential receiver is data input register, and this register sends data to FIFO and then sends bus controller on the plug-in unit to.Output from bus controller directly links to each other with the pci bus of computing machine.Using differential driver and receiver to make the alternative method of scanner head and interface board (being plug-in unit) interconnection is to utilize IEEE 1394 standard cables that are called as " firewire ".

The example of the preferred embodiment of integrated beam shaping circuit 740 is illustrated with Fig. 5 F.Circuit 740 comprises timing circuit 742 and attached to 5 delay circuits 760 that add with circuit 754 every sides.Each circuit 760 comprises sample circuit 746, CCD lag line 752, control and memory circuit 750, demoder 748 and clock driver circuit 744.This Circuits System is touched sheet 756 and surrounds, so that the inlet towards this integrated circuitry is provided.As introducing by quoting as proof in the various embodiments that the patented claim incorporated into states in front, the preferred area of this integrated circuit is less than 20mm ²And can be installed on the veneer in the scanner head.Utilize similar structure can realize the integrated circuit of 16,32 or 64 lag lines.

Fig. 7 A is the synoptic diagram that is superimposed upon the input point in the demonstration.As shown in the figure, the input point Ip that receives from ultrasonic beam 180 does not accurately align with rectangular arranged picture point P on conventional display 32.Because the data of 32 display elementizations of display, so input point Ip must be transformed into the rectangular coordinate form.

Fig. 7 B is the synoptic diagram of the demonstration of Fig. 6, and its input data have been transformed into pixel.As shown in the figure, in order to form image, each becomes image point Ip all to be assigned to a corresponding picture element P on display 32.

A purpose of scan conversion is to finish the spatial alternation of coordinate, needs this conversion when using uneven linear sweep head (as phase-array scanning head, keystone scanning head or crooked linear sweep head).In order to carry out this conversion, must press a kind of sequential read data, and write output data by another sequence.Many existing systems generation order change that must be in operation will reduce dirigibility like this and make the keystone scanning pattern become more difficult.

Because scan conversion will be resequenced for data, so also can be rotated, the face position shows (pan) and zooming (zoom) processing to data.In order to watch scanner head in top, left side, right side or the bottom of image or become the arbitrarily angled image of describing, rotation is useful.Zooming (zooming) and face position show that (panning) is generally used for closely studying the various piece of image.

Except going deep into certain regional zooming of target, can see that simultaneously a plurality of pictures also are useful in the zones of different of fluorescent screen.Often show complete image on the fluorescent screen, the image of (zoomed-in-view) replaces but find a view with trucking up in some zone.This feature is commonly called " picture-in-picture (window-in-a window) ".Current AS provides a window with this ability, but preferably imaging system provides the zooming district of any amount, and wherein each zone all has size and dimension arbitrarily.

Use irregular scan pattern simplified system to design and to allow to utilize more fully scanner head.Specifically, this allows to reduce or hide the dead time relevant with the deep regional imaging.Under the situation of deep regional imaging, wave beam is launched away, but will just can be received in certain slower moment, and this wave beam must advance to depth capacity if having time and returns then before this.If in this internal radiation dead time and rebuild other zone, just can this system of more efficient use and can obtain higher frame speed or more laterally sampling.This may make scan pattern become irregular (though be fix and obviously can calculate).The scan conversion of introducing below flexibly will be proofreaied and correct automatically at this situation.

Fig. 8 is the synoptic diagram of the preferred embodiment of general image reconstruction technology.According to the preferred embodiments of the invention, preferably after beam shaping, directly data are delivered in the PC, finish remaining operation with software.So, additional firmware is reduced to bottom line, so computing machine can be small-sized portable stage, as laptop computer or palmtop computer.

Preferably, there are an input array 142, one to shine upon array 144 and an output array 146 again.Shining upon array 144 again is an array of indexes or array of pointers, and this array is the size of output image, and it is used to definite each pixel that where obtains from input array 142.Each locational numeral is pointed out and will be entered the extracting position of each pixel of output array 146 in the input array with same position in shining upon array 144 again.Therefore, can think that shining upon array 144 again has identical geometric relationship with output array 146, and input array 142 has identical data type with output array 146, i.e. Shi Ji pictorial data.

Input array 142 has the new data that is fit to every frame ultrasonic image, this means that it handles these data and they are put in the output array 146 frame by frame.According to the present invention, new ultrasonic image frame forms about 1/30 second of preferably every frame with the speed of per second at least 20 frames.But, shine upon array 144 again and only just be updated in the type of upgrading scanner head or the parameter of finding a view (being that zooming and face position show) time.Therefore, as long as can finish with the frame speed of about 30 frames of per second from the routine operation of the new output image of one group of new input data computation, the data of shining upon array 144 again can generate more lentamente, (but still about below one second, the trouble otherwise it may become).This means under the situation of not damaging overall performance and can finish the data computing task of shining upon array 144 again, and will additional firmware not be used for this task with general personal computer.In the computing system with digital signal processor (DSP), DSP can finish the calculating of shining upon array 144 again.

In a preferred embodiment of the invention, the input store that is used to import array 142 can be two groups of static RAMs (SRAM) or one group of video RAM (VRAM), and wherein input is serial access and to export be random access.But VRAM group possible speed is too slow and to refresh cost too high.Though random-access SRAM also will move, the mapping memory again that is used for shining upon again array 144 preferably is embodied in sequential access memory or the dynamic RAM (DRAM) of VRAM.The output storage that is used to export array 146 can be frame buffer or first in first out (FIFO) impact damper.The scan conversion that is in operation is promptly carried out once request basically.The preferential selection of scan conversion finished on PC by software.But if carry out scan conversion with hardware, therefore PC storage data has only so reduced the complicacy of system.Therefore, preferably have only shining upon impact damper again and replacing small-sized FIFO (if there is) or the position that output buffer uses for pipelining of two random-access input buffers, a sequential access according to architecture of the present invention.This means that the frame output buffer is in the storer of PC.

According to the preferred embodiments of the invention, in ultrasound scan conversion, use the spatial dithering technology that adopts error diffusion.Typically shiver and in the pixel brightness territory, carry out.But shivering according to the present invention is used in the ultrasound scan conversion, so that aspect spatial domain rather than approaching pixel aspect the pixel brightness territory.Spatial dithering is used to approach and drops on the values of two inputs between the data point.This thing happens only is that pixel on the display screen may drop on and needs between two radiuses through filtering because sample radius is discontinuous.For interpolation between vertical sampled point must usage space be shivered.

Please recall, shine upon array 144 again and storing of the mapping of each output point input point.The input data point is normally in polar coordinate system, and output point is then in rectangular coordinate system.Only comprise the index that enters input array 142 though shine upon array 144 again, can regard them as and comprise radius value (r) and angle value (θ).In the imagination, these numerical value have arbitrary accuracy and needn't be corresponding with actual samples point.Consideration now must become round values with the digital conversion of these arbitrary accuracies.The integer radius value is corresponding to discrete increment, and these increments have been removed and have been subjected to the radius sampling density of system to limit.The integer angle value is corresponding to discrete radius, these radius be scanned and therefore limited by scanning angle.If shiver in the application space, these floating point values can be transformed into fixing round values and not have non-natural sign, and sort signal is not occur when having the discrete rounding of error diffusion.

Fig. 9 A to Fig. 9 B is a process flow diagram, and the computing technique of shining upon array according to the present invention again is described.In step 205, check scanner head looks at whether change.If scanner head changes, continue execution in step 210 so, in this step, give the type configuration of the scanner head that makes new advances.After completing steps 210, perhaps the scanner head configuration does not change (step 205), continues execution in step 215 so.In step 215, the inspection display window looks at whether zooming is arranged, the face position shows or new picture-in-picture feature.If have, continue execution in step 220, import the new parameter of finding a view this step user.After completing steps 220, perhaps display window does not change in step 215, continues execution in step 225, and array is shone upon in removing again in this step, so that point out to import the new relation between array and the output array.

In step 230, the window W that this procedure Selection is pending.In step 235, with all capable error amount L _EWith all sampling error value S _EAll be initialised to zero.In step 240, some counter P is initialized to the picture element in the window W upper left corner.

In step 245, the floating-point row L of each point among the V of the application computes visual field _FPWith sample-offset S _FPFor phased array, this will be radius r and angle θ.In step 250, the propagated error item L in early stage _E, S _E(below discussed) is added to the floating point values L of a P _FP, S _FPOn, arrived immediate integer L in step 255 by rounding _R, S _R, this to integer corresponding to the actual samples point.In step 260, application computes rounding error:

L _RE＝L _FP-L _R

S _RE＝S _FP-S _R。

Propagated error

L _{E (right side)}=L _{E (right side)}+ L _RE ^*7/16

L _{E (bottom left)}=L _{E (bottom left)}+ L _RE ^*3/16

L _{E (descending)}=L _{E (descending)}+ L _RE ^*5/16

L _{E (bottom right)}=L _{E (bottom right)}+ L _RE ^*1/16

S _{E (right side)}=S _{E (right side)}+ S _RE ^*7/16

S _{E (bottom left)}=S _{E (bottom left)}+ S _RE ^*3/16

S _{E (descending)}=S _{E (descending)}+ S _RE ^*5/16

S _{E (bottom right)}=S _{E (bottom right)}+ S _RE ^*1/16

In step 265, error is extended towards the right side, lower-left, the following and bottom-right picture element of current some P.

In step 270, application program is the computational data index on the basis of scan-data ranking index:

REMAP(P)＝INDEX(L _R，S _R)。

In step 275, check implementation status, look at whether more point is arranged in this window.If there is more point pending, in step 280 pointer P is added 1, to next point.Turn back to step 245 then.In case points all in the window continue execution in step 285 all through handling.

In step 285, check implementation status, look at whether to have more window pending.If have, turn back to step 230.Otherwise, continue down to carry out.

Because shiver a source (one source) is transformed into each output pixel, becomes possibility so same mapping architecture again can be used to make real-time scan conversion to finish with software, on portable computer even also be like this.Therefore, complicated shiver operation just during the initialization or the parameter of finding a view just be performed when changing.But the benefit of shivering but appears in all images.

Figure 10 is the process flow diagram of output frame computing engines.In step 305, the input data of process beam shaping and demodulation process are read into storer.In step 310, output pixel index P is initialised.In step 315, according to following formula the output array is set, make it equal the input array of shining upon through again,

OUTPUT(P)＝INPUT[REMAP(P)]。

In step 320, output pixel index P adds 1.In step 325, check pixel index P, look at whether this image forms.If do not form, return step 315.In case all pixels all through calculating, continue execution in step 330 in the image, output image carries out non-essential smoothing processing in this step.At last, show output image in step 335.

Really removed those non-natural signs (machine striped (mach-banding) and cross-hatched pattern) of following the increment rounding to take place though shiver, shivered and to introduce high frequency noise.The mean value of this high frequency noise can provide and seamlessly transit effect just.For untrained eyes, it is so tedious that these non-natural signs can not show a candle to those non-natural signs that obtain with simple rounding or closest approach method, may be exceptionable for the ultrasonic technique professional still.

By the low pass spatial filtering image smooth can will be significantly reduced these non-natural signs after the mapping processing again, even may eliminate.This wave filter can be a kind of tank filters or the asymmetric filters that can mate with the input resolution character of needs.This class wave filter can with the straight line territory of the direction of the point coordinate of specific position or angle coupling in use.

Basically, the wave filter with a coupling is satisfactory, and the filter range of this wave filter is equivalent to or is proportional to distance between the point of shivering.Therefore, high power preferably has sizable smoothing filter and accompanies, otherwise, in the occasion of the interval of sample radius r or angle θ all quite little (on the order of magnitude of a pixel), may just not need filtering.

Because map operation is twice loading basically and once stores again, so it can be finished with the personal computer of standard.During with assembly language coding again mapping algorithm proved already that on based on the PC of 166Mhz Pentium chip operation will obtain very the result near true-time operation.In addition, on PC, finished demodulation with the demodulation process of compilation language compilation, and near true-time operation.Text and icon are subjected to following operating influence: with some fixed numbers is the top that color is stored in input buffer, is mapped to those places that are about to use these colors then.If come into force, different shape is that text is plotted in and shines upon in the array again, and this array will be opened and just need not to calculate and can automatically be superimposed upon on all images.

Figure 11 A and Figure 11 B are the synoptic diagram of the display format that may occur on display 32 of the present invention.System of the present invention has the many windows display format that can be selected by the user, rather than resembles and only show a data window the early stage ultrasonography system.Figure 11 A represents that selectable many windows show that wherein three information windows are simultaneously displayed on the display.The B-scanning image of window A display standard, and window B shows the M-scanning image of the colored flow pattern of Doppler of two dimension.Window C is the user profile window, and it passes to the user with command selection and is convenient to the manual selection of user.The essential single window of Figure 11 B right and wrong shows that wherein entire display screen only is used to submit to the B-scanning image.B mode scanning image and color Doppler scanning image can be optionally showed in this demonstration simultaneously, and its method is to be piled up or to utilize the fluorescent screen to cut apart feature two kinds of demonstrations to show side by side.

Figure 12 is the functional-block diagram of preferred graphical interface of user.Virtual controlling 400 comprises that ultrasonic image control demonstration 410, the demonstration 420 of probe model performance and probe property show 500.Virtual controlling shows that the dialog box of 400 preferential conducts in Windows is encoded.

Figure 13 explanation is used for the dialog box of ultrasonic image control 410.Show 410 by ultrasonic image control, the user can select probe type 412, subregion demonstration 414, demodulator filter 416 and Algorithm Options 418.Can also finish the ultrasonic scanning initialization by this dialog box.

Probe model performance shows geometric configuration 450, area of image data 455, Doppler district data 460, color region data 465, visual apodization 470, Doppler's apodization 475 and the colored apodization 480 of 420 integrated pulse amplitude (IPA) data 435 that comprise model 425, security information 430, image, Doppler IPA data 440, colored IPA data 445, probe.These model performances preferentially are compiled into code as dialog box.By model performance dialog box 425 users can the input probe model general adjustment.

Figure 14 A explanation is used to import the dialog box of probe model performance of finding a view.The parameter of input will be downloaded to ultrasonic probe.

Figure 14 B explanation is used to import and check the dialog box of security information 430.As shown in the figure, the user can import the beam angle tables of data 434 of general adjustment 432 and every kind of control criterion.

Figure 14 C explanation is used to import and check the dialog box of visual IPA data 435.The output valve of this dialog box disclosing solution beam shaping is listed at different driving voltages as the function of image displaying area, and unit is a volt.Similarly dialog box is used to import Doppler IPA data 440 and colored IPA data 445.

Figure 14 D explanation is used to influence the dialog box that image is cut apodization function 470.As shown in the figure, the user can import and check general adjustment 472 and Vector Message 474.The user can select to be used to open effective unit of array window (being apodization).

The probe property shows that 500 comprise the dialog box 510 that is used for the input probe property, visual field of view (FOV) data 520, Doppler FOV data 530 and colored FOV data 540.Can import general adjustment 512, imaging static information 514, Doppler's static information 516 and FOV by probe characteristic dialog box 510 users and adjust 518.

The dialog box that Figure 15 A explanation is used to import and check the probe specific information.Can support the probe of any amount.

Figure 15 B to Figure 15 C explanation is used for the dialog box of input imagery FOV data.As shown in the figure, the user can import general adjustment 522, breakpoint PGC data 524, partition boundaries 526 and subregion duration 528.Be used to show the dialog box the 530, the 540th of Doppler and colored FOV data, similarly and also the both be generally adjust 532 and 542, breakpoint TGC data 534 and 544 and the inlet of PRF data 536 and 546.

Figure 15 D to Figure 15 J explanation is according to additional window of the present invention and the control panel that is used to control the ultrasonography system.Figure 15 D show perception interest zone view window and with scanning image control panel side by side.Figure 15 E shows can select the various controls set to Doppler visual field and other.Figure 15 F shows the control of colored visual field.Figure 15 G shows the performance of probe.Figure 15 H shows the colored IPA data of probe.Figure 15 I shows the probe geometry setting that is used for linear array.Figure 15 J shows the adjustment that is used for Doppler's apodization.

Figure 16 illustrates the zooming feature according to the preferred embodiment of imaging system of the present invention.In this concrete illustrating, the detailed features of mirage (promptly being presented at the anatomical features 600 in the patient's body on the fluorescent screen 32) can be selected in display window or on the display window and amplify.In this instantiation, user-selected area 602 and it is amplified in window 609.Can there be large quantities of such zones to be exaggerated simultaneously and to be simultaneously displayed on the fluorescent screen 32 with that separate or overlapping window.If use two scanner heads simultaneously, the different visuals field can show simultaneously that perhaps the image that will record in the past accesses and be presented at the next door of the image of real-time submission from storer.

For small size, low-power consumption being provided and scanning dirigibility to greatest extent, designed the architecture of integrated front-end probe method, comprising: subregion focuses on when 1) launching; 2) drive various probes (such as linearity/flexural linear, linear/trapezoidal) and the ability of sector scanning; 3) provide the ability that the B-pattern shows, the M-pattern shows, colored flow pattern (color flow map) shows and Doppler sonar figure (Doppler Sonogram) shows; 4) multiple selective pulse shape and frequency; And 5) different shooting sequence.The various embodiments that are used for front end integrated system 700 are illustrated in Figure 17 A, 17B and 17C.The exclusive module of the present invention be respectively with beam shaping chip 702, emission/receiving chip 704, prime amplifier/TGC chip 706 corresponding components.

The parts of indicating " front-end probe " (front controller) offer the clock signal of module 702,704,706 and memory cell 708 and the routine operation that control signal is directly controlled ultrasound scanhead by generation.These signals are used to guarantee continuous data output and point out the data that appear at the memory cell output terminal are scheduled to for which module use.The Advanced Control of scanner head 710 and initialization, data processing function and Presentation Function are provided by universal host machine (such as desktop PC, laptop and palm machine) 720.Therefore, front controller also with host interface (for example by PCT bus or FireWire714) so that allow main frame that control data is write the memory cell of scanner head and receives return data.This is to select different sweeping schemes when initialization and the user, need to change parameter (such as quantity and/or the position or the scanner head type of subregion) time and finish.In order when the data from beam-shaper must send main frame to by the link of limiting bandwidth, to prevent that loss of data, front controller from also providing pooling feature and data flow con-trol function.

Described system allows to realize two kinds of different functions, colored flow pattern (CFM) function and Doppler sonar figure (DS) function.Figure 17 A shows hardware based configuration 722, and wherein Zhuan Yong Doppler's process chip is installed on the rear end plug-in unit 724 and is used as the coprocessor of main frame 720, and CFM calculates and DS calculating so that finish.Figure 17 B shows software arrangements, and wherein CFM calculating and DS calculate and finished by main frame.

Figure 17 C shows another kind of integrated system, and wherein transducer array and each front-end processing unit are not integrated in the housing, but couple together by concentric cable.As shown in the figure, front end unit comprises front controller, storer and three modules 704 (emission/receiving chip), 706 (prime amplifier/TGC chip) and 702 (beam shaping chips).

" FireWire " refers to the IESE of high speed data transfer standard 1394 is provided on serial link.This allows the commercially available part that the big cost of volume is low to be used for interface.This standard is supported asynchronous data-transmission mode.This pattern can be used for sending instruction and system configuration data to probe memory.Side information it can also be used for the state of inquiry scan head and obtain to be activated about the button on scanner head or other input equipment.In addition, the asynchronous data transfer pattern can be used to detect the type of used probe.The data transmission that isochronous transfer mode can be used for returning from beam-shaper is given main frame.If doppler processor is placed in the probe, the data of handling through Doppler can transmit with FireWie so.In addition, can carry out Doppler's processing to data by software in the main frame or hardware.The FireWire standard also has wireless version, and it allows to carry out wireless communication by optical link.When utilizing wireless FireWire to be connected to probe on the main frame, this can be used to the freedom that provides bigger.

Prime amplifier/TGC chip is the integrated circuit that is made of 32 lower powered prime amplifiers of parallel low noise/TGC unit when realizing.Each unit have 60dB programmable-gain, be lower than 1.5nVI

Noise voltage and be lower than the dissipation of each receiver channels 11mW.

As shown in figure 18, multichannel emission/receiving chip is made up of the parallel emission/reception controller of global counter, global storage and double-channel.Local storage 745, delay comparator, Pin Shuaijishuoqi ﹠amp are arranged in each controller 740; Comparer, Mai Chongjishuoqi ﹠amp; Comparer, phase selector, emission/reception multidiameter option switch (T/R switch) and level inverter cell.

Global counter 742 sends major clock and place value to each channel processor 740 simultaneously.Global storage 744 control transmission frequencies, umber of pulse, pulse sequence and emission/reception are selected.Local delay comparer 746 will postpone to select to offer every channel.For example, adopt the clock of 60MHz and 10 global counter, the delay that is no more than 17 μ s can be offered every channel.Local frequencies counter 748 provides programmable transmission frequency.4 digit counters of a tape comparator can provide 16 kinds of different frequencies to select at most.For example, adopt the 60MHz major clock, one 4 counter can provide different transmission frequencies through programming, as 60/2=30MHz, 60/3=20MHz, 60/4=15MHz, 60/5=12MHz, 60/6=10MHz etc.Local pulse counter 750 provides different pulse sequences.For example, 6 digit counters of a tape comparator can provide programmable fire pulse width, from pulse to 64 pulse.The local phase selector switch provides period of the day from 11 p.m. to 1 a.m clock delay resolution.

Although determine delay resolution by the clock period of transmitting chip usually, a kind of technology that is called as programmable period of the day from 11 p.m. to 1 a.m clock delay resolution allows delay resolution more accurate than the clock period.Adopt programmable period of the day from 11 p.m. to 1 a.m clock to postpone resolution techniques, use output based on the clock phase gating frequency counter of every channel programming.In the simplest form, adopt two phase clock, and the output of frequency counter is with asserting clock (assenedclock) gating or with non-clock (deasserted clock) gating of asserting.Another kind of way is to adopt a plurality of clocks (skewed clocks) that tiltedly move.Can select one of every channel, and with the coarse timing signal of its gating from frequency counter.For example, for the major clock of 60Mhz, the delay resolution of 8ns is provided two-to-one phase selector and four to one phase selector provides the delay resolution of 4ns.

Being used for exomonental integrated emission/reception selector switch 754, T/R switch and high level phase inverter 750 also is demonstrated out.Can adopt a slice can handle the emission/receiving chip of 64 channel drivers and 32 channel receivers, every channel has a controller shown in Figure 180.

In the another kind configuration (as shown in figure 19), T/R multidiameter option switch and high pressure phase inverter separate with other components and parts on other chip, so that allow to adopt high-voltage semi-conductor technology (silicon CMOS/JFET or the GaAs technology high as voltage breakdown) to produce these components and parts.

The basic skills of pulse Doppler ultrasonography is illustrated with Figure 20.Waveform is made up of the train of impulses that comprises N pulse 770.After each pulse, gather the increment of different distance (degree of depth) on demand.Velocity distribution in this range gate in the material was launched to be revealed as the sonar chart 772 of two dimension by the time, and wherein transverse axis is represented the time and Z-axis representation speed (this speed is by the Doppler shift estimation).Can query to different zones with its size of change by displacement ripple door.As shown in figure 20, adopt Doppler's processing of a range gate just can produce Doppler sonar figure.With the sequence of N ultrasonic pulse with pulse repetition rate f _PrfLaunch along given viewing angle.The echo that returns is behind the range gate gating, only use, this means in order to extract doppler information and only handle the signal that the viewing angle along select corresponding with the zone of selected distance (for example from depth d to d+ δ d) returns from transducer array from the echo between a distance regions.By the Doppler shift that calculating is received, can obtain the velocity distribution of selection area inscattering body from the echo of scatterer.In other words, the time-domain signal of receiving is carried out Fourier transform 776 will provide frequency information, comprise required Doppler shift f _dVelocity distribution at the interesting regional inscattering body of perception can obtain from following relational expression:

f _d＝2(v/c)f _c

Wherein c is the velocity of sound in the transmission medium, f _cIt is the centre frequency of transducer.For example, if N=16, f _Prf=1kHz uses following formula will produce the sonar chart 772 that shows the 16ms doppler data.If every N/f _PrfRepeat once this process second, and so continuous Doppler sonar figure can produce.

Another embodiment relates to and is fit to the pulse Doppler processing that colored flow pattern is used.What meet clinical requirement is to show large-scale flow velocity and flow pattern in real time.Be fit to be called as look stream mapping (CFM) with the method for this task of ultrasonic Treatment.Look stream mapping techniques is the extension of above-mentioned single gating system.In CFM, the speed assessment is not only carried out along a direction or line segment, and carries out many covering on the regional direction (multi-strip scanning line) of perception interest.Velocity information normally coloud coding (bulc away is flowed in for example red expression towards transducer) and overlap on the B-pattern image that shows bottom layer tissue.

The colored flow pattern of handling based on pulse Doppler 780 is shown in Figure 21.By be kept in different depth sampling and with increment the storage space that is used for additional treatments can be with single distance regions shown in Figure 20 between the ultimate system popularization so that a large amount of range gates is measured.Notice that this does not increase detection time, because data are gathered from same RF line.Allow inswept certain zone of wave beam just might in the 2 dimensional region of perception interest, be combined into the speed image.In when operation, parallel processing along a direction from 782 data between J distance regions.After handling N pulse echo, J * N of output expression is to the distribution to Doppler of the distance of data, and this distribution can be used to produce the velocity distribution profile of a J * N to data successively.Each depth d _k(k=1,2 ... J) average velocity is used for producing the i.e. grid of a point on colored flow pattern; In each grid, standard deviation is used to assess turbulivity.If the every N/f of this program _PrfRepeat once to (for example, between a spaced apart J/2 distance regions) between every J distance regions with to every sweep trace in the interesting zone of perception second, can produce the color flow spectrogram of two dimension so.

It should be noted that, just as Jorgen A.Jensen in publication " Estimation ofBlood Velocities Using Ultrasound " (University Press 1996), introduce like that, cross-correlation techniques can replace being used to produce similar colored flow pattern based on the calculating of FFT, this article content is incorporated into by quoting as proof at this.

The size of range gate and position can be determined by the user.The pulse width and the pulse repetition rate of emission determined in this selection.The size of range gate is determined by pulse width.The duration of pulse is:

T _p＝2I _g/C＝M _fc

I wherein _gBe gatewidth, M is the sinusoidal period number.The degree of depth of ripple door determines to obtain the pulse echo line with speed how soon.Maximal rate is:

f _Prf=c/2d ₀D wherein ₀It is distance to the ripple door.

The waveform that belongs to together that is used for the pulse Doppler ultrasonography represents that with Figure 22 wherein waveform comprises the train of impulses 800 of N pulse.Foundation needs to gather and the as many degree of depth increment of distance after each pulse in train of impulses.Figure 22 also shows the block scheme 810 of the normal signal processor that is fit to this imaging technique, wherein the back echo that receives of each transducer before homophase and quadrature demodulation, sample and coherently add and.Frequency reducing/base band echo is transformed into the numeral expression formula, is stored in then in the memory buffer, till the whole pulse echos that comprise coherence interval all are received.Then, from storer, read N the pulse echo that each degree of depth is gathered, apply weighting sequence v (n), so that the control Doppler sidelobne, and then calculate N-point FFT.During handling by Doppler filter from the degree of depth increment of a coherence interval, handle from the echo of next coherence interval by wave filter, arrive at and be stored in second input store from the echo of next coherence interval.The output of FFT818 is transferred to display unit, or calculates the time average of Doppler's increment for demonstration subsequently.

Except the repertoire of pointing out in the CDP of this introduction device will be finished frame of broken lines at Figure 22, A/D mapping function, this function is unnecessary, because the CDP device provides the function of analog sampled data.This CDP pulse Doppler processor (PDP) device has the ability of compute matrix multiply matrix, so the needed ability of function is much bigger shown in the energy force rate that it the had realization frame of broken lines.

The PDP device calculates the product of two real number value matrixes, its method be with add by the capable accordingly apposition that forms of first matrix column and second matrix and.

In order to introduce the application of PDP in the doppler filtering problem, we at first the doppler filtering equation is become real-valued matrix operation and.The method of finishing doppler filtering is the discrete Fourier transformation (DFT) to the pulse echo of the depth calculation process weighted of each perception interest.If we represent the degree of depth-Doppler's increment with g (kj), wherein k is Doppler's index, 0≤k≤N-1, and j is the depth indexing, so $g(k,j)=\underset{n=0}{\overset{n-1}{\mathrm{\&Sigma;}}}v\left(n\right)f(n,j)\exp (-\mathrm{<figure-callout\; id="2"\; label="j"\; filenames="A9718098400441.png,A9718098400591.png"\; state="\{\{state\}\}">j</figure-callout>}2\mathrm{\&pi;kn}/N)$

Weighting function can merge with the nuclear of DFT, so that obtain element by the given doppler filtering transform coefficient matrix of following formula:

W(k，n)＝W _k，n＝ν(n)exp(-j2πkn/N)。The real component of signal and imaginary component can be write as now behind doppler filtering: $g_{r,\mathrm{kj}}=\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}(W_{r,\mathrm{kn}}{f}_{r,\mathrm{nj}}-W_{i,\mathrm{kn}}{f}_{i,\mathrm{nj}})$ $g_{r,\mathrm{kj}}=\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}(W_{r,\mathrm{kn}}{f}_{r,\mathrm{nj}}+W_{i,\mathrm{kn}}{f}_{i,\mathrm{nj}})$

In following formula, add the index that double-legged target variable can be regarded as matrix entirely.So in matrix expression, doppler filtering can be expressed as the product calculation of matrix.People can see that the PDP device can be used to finish each matrix multiplication operation, realize the doppler filtering operation whereby.

The block scheme of the PDP device of Jie Shaoing is shown in Figure 22 in the present invention.This device comprises the Output Shift Register of J level CCD tapped delay line, a J CCD multiplication D/A converter (MDACs), J * K totalizer, J * K Doppler's increment impact damper and parallel input string line output (PISO).8 bit digital input equipments of MDACs sharing of common also transmit on this input equipment from the element of matrix of coefficients.Tapped delay line is finished sampling and hold function, and continuous in time analog input signal is transformed into simulating signal through over-sampling.

The two PDP parts 840 that are used for the mapping of look stream in the ultrasonography system are shown in Figure 23.In these parts, between a pulse return period, the PDP element at top calculates W _kf _rAnd W _if _kThe item of form (shown in the top), and bottom PDP element calculating-W _iFi and W _kf _iThe item of form.Then the output of each element is added and, so that alternately obtain g _rAnd g _i

Doppler and colored flow pattern are handled and are related to quite heavy calculated amount.This processing can utilize general purpose microprocessor to finish with software.There is the instruction (such as Inter MMX group of functional units) of optimization matrix-matrix operation can improve combination property greatly.With the FFT computational algorithm is that the software flow pattern of the colored flow pattern of basic calculation is illustrated with Figure 24.After initialization 900, obtain frequency reducing data 902 and make pointer P be positioned at the starting point 904 of sweep trace, calculate the mean value of these data and store 906, apply weighting function 908, calculate FFT910, at each frequency computation part numerical value z (k) 912, calculate then first and second key elements 914 and with color show they 916.Pointer adds 1 and handle every sweep trace as requested.

The software flow pattern that is calculated as the colored flow pattern of basic calculation with crosscorrelation is illustrated with Figure 25.

After initialization 940, obtain scan-line data 942, obtain then apart from interval censored data 944.Crosscorrelation calculates 946 and ask average 948, and computing velocity distributes 950 then.Obtain first and second key elements 952 and show they 954.Increase repeats this process then apart from interval censored data 956.

Although showed the present invention particularly and introduce with reference to the preferred embodiments of the invention, the people who is familiar with this technology is to be understood that not breaking away from the defined spirit and scope of claims of the present invention can do various variations in form and details.

Claims (20)

1. one kind is transformed into the method for display format with ultrasound scan data, and this method comprises the steps:

A) provide one by the input array of forming from the input data of ultrasonic device; And

B) the input data are carried out spatial dithering and handle, become the output array of forming by output data so that will import data conversion.

2. according to the process of claim 1 wherein that the spatial dithering step comprises the rounding position data, become the round values of band remainder values from the numerical value rounding of arbitrary accuracy.

3. according to the method for claim 2, wherein the spatial dithering step further comprises by output data propagation remainder values.

4. according to the process of claim 1 wherein that the spatial dithering step comprises:

By the primary importance in first numerical evaluation output array in the input array; And

Utilize primary importance to calculate the second place of exporting in the array.

5. according to the method for claim 4, wherein the pixel location of demonstration is adjoined in the representative of first and second positions.

6. according to the process of claim 1 wherein that the spatial dithering step finishes on multi-purpose computer.

7. according to the method for claim 6, wherein multi-purpose computer is a kind of portable computer.

8. provide the step of input array further to comprise according to the process of claim 1 wherein by the input data of polar coordinates reception from ultrasonic device.

9. according to the method for claim 1, this method further comprises output Array Transformation one-tenth input array.

10. according to claim 9 method, wherein said shift step comprises that will shine upon array again according to variation is initialised to display format.

11. according to claim 1 method, this method further comprises provides scanning conversion circuit, this circuit carries out spatial dithering to the input data to be handled.

12. one kind becomes the method for display format with the ultrasonic scanning data conversion, this method comprises:

The input array of being made up of the input ultrasound data is provided;

Personal computer is provided, and this computing machine is finished the conversion of input ultrasound data through programming; And will become to have the output array of display format by the input Array Transformation that the input ultrasound data is formed.

13. according to the method for claim 12, this method further comprises with the speed of per second 20 frames at least carries out conversion to a large amount of input ultrasound data frames.

14. according to the method for claim 13, this method further comprises with about per second 30 frames or higher speed carries out conversion.

15. a ultrasonography system, this system comprises:

A transducer array that transmits and receives ultrasonic signal;

A beam shaping circuit that is electrically connected with transducer array, this circuit forms the electronics presentation for the treatment of the imaging target;

A programmable data disposal system, this system receive described electronics presentation, finish the program of electronics presentation scan conversion to computer compilation, so that described electronics presentation is transformed to display format.

16. according to the system of claim 15, wherein the programmable data disposal system is a personal computer.

17. according to the system of claim 15, wherein the programmable data disposal system is the portable computer with keyboard and LCD.

18. according to the system of claim 15, wherein the programmable data disposal system is a palmtop computer.

19. according to the system of claim 15, this system further comprises interface circuit.

20. according to the system of claim 15, this system further comprises hand-held casing, this hand-held casing comprises transducer array and beam shaping circuit.

Images