Ultrasound imaging method and device with prediction of artefacts induced between reconstruction modes

Abstract

The invention relates to an ultrasound imaging method for imaging a part (1), characterized by the implementation of the following steps: selecting a first sub-region ({tilde over (Z)}) of the part from a first image (I.sup.A(Z)) of a region (Z) of the part (1), determining, for each point of the first selected sub-region ({tilde over (Z)}), the times of flight (T.sub.ij.sup.A({tilde over (Z)})) corresponding to the paths according to a first reconstruction mode (A) going through the point from a transmitter i to a receiver j for a set of M*N transmitter-receiver couples of an ultrasound signal; determining a second sub-region of the part, a point (P) of the region belonging to the second sub-region when a time of flight (T.sub.ij.sup.B(P)) of the path according to a second reconstruction mode (B) going through the point (P) from a transmitter i to a receiver j of said set of M*N transmitter-receiver couples coincides with a time of flight (T.sup.A({tilde over (Z)})) of a path according to the first reconstruction mode from a transmitter to a receiver from the transmitter i to the receiver j going through one of the points of the first selected sub-region.

Claims

1. A method of inspecting a structure to detect a presence of potential defects therein, including steps implemented by a processor coupled to a multi element transducer having M transmitters of an ultrasound signal and N receivers of the ultrasound signal, the method comprising: transmitting successively via the M transmitters the ultrasound signal in the structure, and recording in the N receivers echoes combining from propagation of the ultrasound signal in the structure, to acquire a set of M*N ultrasound signals S.sub.ij(t), where i denotes one of the transmitters, j denotes one of the receivers, and t denotes time; determining, for each point of a first sub-region of a region of the structure and for each transmitter-receiver pair i-j, a time of flight of a path in accordance with a first ultrasound wave propagation model that goes through the point from a corresponding transmitter i to a corresponding receiver j; determining a second sub-region of the region of the structure, wherein for each transmitter-receiver pair i-j a point of the region belongs to the second sub-region when a time of flight of a path in accordance with a second ultrasound wave propagation model that goes through the point from the corresponding transmitter i to the corresponding receiver j coincides with a time of flight of a path in accordance with the first ultrasound wave propagation model that goes through one of the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j; forming an image of the region of the structure by synthetic focusing, wherein for each point of the region, and for each transmitter-receiver pair i-j, an ultrasound signal considered at a time of flight of the path in accordance with the second ultrasound wave propagation model that goes through the point of the region from the corresponding transmitter i to the corresponding receiver j is exploited uniquely if said time of flight coincides with one, or differs with all, of the times of flight of the paths in accordance with the first ultrasound wave propagation model that goes through one of the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j.

2. The method of claim 1, wherein forming the image comprises, for each point of the region, determination of a sum of amplitudes of acoustic signals limited to the acoustic signals received at the corresponding receiver j from the corresponding transmitter i when the time of flight of the path in accordance with the second ultrasound wave propagation model that goes through the point from the corresponding transmitter i to the corresponding receiver j coincides with a time of flight of a path in according with the first ultrasound wave propagation model that goes through one of the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j.

3. The method of claim 1, wherein forming the image comprises, for each point of the region, determination of a sum of amplitudes of acoustic signals limited to acoustic signals received at the corresponding receiver j from the corresponding transmitter i when the time of flight of the path in accordance with the second ultrasound wave propagation model that goes through the point from the corresponding transmitter i to the corresponding receiver j does not coincide with one of the times of flight of the paths in accordance with the first ultrasound wave propagation model that goes through the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j.

4. The method of claim 1, wherein each one of the first and second ultrasound wave propagation models includes a type of route comprising several portions and a mode of propagation of the ultrasound signal on each portion of the route.

5. The method of claim 4, wherein the first ultrasound wave propagation model is a model according to which each ultrasound signal follows a direct route with two portions, a mode of propagation of the ultrasound signal being lateral on each of the portions.

6. The method of claim 5, wherein the second ultrasound wave propagation model is a model according to which each ultrasound signal follows a direct route with two portions and a mode of propagation of the signal is different on each of the portions, or a model according to which each ultrasound signal follows an indirect route with three portions at least.

7. A non-transitory computer-readable medium storing computer-readable instructions thereon that, when executed by a computer, cause the computer to perform the method according to claim 1.

8. An inspecting device for detecting potential defects in a structure, comprising: M transmitters of an ultrasound signal in the structure to be image; N receivers of an ultrasound signal S.sub.ij(t) coming from the structure to be imaged, the index i designating one of the transmitters, the index j designating one of the receivers, and t representing time; and a processor coupled to the M transmitters and the N receivers, wherein the processor is configured to: determine, for each point of a first sub-region of a region of the structure, and for each transmitter-receiver pair i-j, a time of flight corresponding to a path in accordance with a first ultrasound wave propagation model that goes through the point from a corresponding transmitter i to a corresponding receiver j; determine a second sub-region of the region of the structure, wherein for each transmitter-receiver pair i-j a point of the region belongs to the second sub-region when a time of flight of a path in accordance with a second ultrasound wave propagation model that goes through the point from the corresponding transmitter i to the corresponding receiver j coincides with a time of flight of a path in accordance with the first ultrasound wave propagation model that goes through one of the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j; forming an image of the region of the structure by synthetic focusing, wherein for each point of the region, and for each transmitter-receiver pair i-j, an ultrasound signal considered at the time of flight of the path in accordance with the second ultrasound wave propagation model that goes through the point of the region from the corresponding transmitter i to the corresponding receiver j is exploited uniquely if said time of flight coincides with one, or differs with all, of the times of flight of the paths in accordance with the first ultrasound wave propagation model that goes through one of the points of the first sub-region from the corresponding transmitter i to the corresponding receiver j.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Other aspects, aims, advantages and characteristics of the invention will become clearer from reading the following detailed description of preferred embodiments thereof, given as non-limiting examples, and made with reference to the appended drawings in which:

(2) FIG. 1a is an illustration of a direct time of flight route;

(3) FIG. 1b is an illustration of an indirect echo-type time of flight route;

(4) FIG. 1c is another illustration of an indirect echo-type time of flight route;

(5) FIG. 2a is an illustration of reconstruction results by TFM from experimental data;

(6) FIG. 2b is another illustration of reconstruction results by TFM from experimental data;

(7) FIG. 3a is an illustration of an artefact from a geometry echo;

(8) FIG. 3b is another illustration of an artefact from a geometry echo;

(9) FIG. 4a illustrates and a first reconstruction mode of direct LL type;

(10) FIG. 4b is an illustration of a second reconstruction mode of direct LT type;

(11) FIG. 5 illustrates different indications present in the image of a part reconstructed by a first reconstruction mode of direct LL type;

(12) FIG. 6a illustrates the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction mode of direct LT type;

(13) FIG. 6b is another illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction mode of direct LT type;

(14) FIG. 6c is a further illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction mode of direct LT type;

(15) FIG. 7a illustrates the filtering of artefacts in a reconstruction according to a mode of direct LT type;

(16) FIG. 7b is another illustration of the filtering of artefacts in a reconstruction according to a mode of direct LT type;

(17) FIG. 7c is a further illustration of the filtering of artefacts in a reconstruction according to a mode of direct LT type;

(18) FIG. 8a is a first illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(19) FIG. 8b is a second illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(20) FIG. 8c is a third illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(21) FIG. 9a is a fourth illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(22) FIG. 9b is a fifth illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(23) FIG. 9c is a sixth illustration of the application of a method according to the invention with a first reconstruction mode of direct LL type and a second reconstruction of corner echo LLL mode type, with selection of two different regions in an image reconstructed while considering a first reconstruction mode of direct LL type;

(24) FIG. 10a is a first illustration of the filtering of artefacts in a reconstruction according to a mode of corner echo LLL type;

(25) FIG. 10b is a second illustration of the filtering of artefacts in a reconstruction according to a mode of corner echo LLL type;

(26) FIG. 10c is a third illustration of the filtering of artefacts in a reconstruction according to a mode of corner echo LLL type; and

(27) FIG. 10d is a fourth illustration of the filtering of artefacts in a reconstruction according to a mode of corner echo LLL type.

DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS

(28) The invention relates to an ultrasound imaging method for imaging a part, implementing a reconstruction by synthetic focusing, for example total focusing, by means of a processor coupled to a multiple element transducer 1.

(29) With reference to FIG. 4a, the method exploits the multiple element transducer 1 arranged on the surface of a part 2 to be imaged, comprising M transmitters of an ultrasound signal, indexed i, and N receivers of an ultrasound signal, indexed j, supplying signals S.sub.ij(t) representative of the echo of the ultrasound wave received for a transmission by a transmitter i and a reception by a receiver j.

(30) Still with reference to FIG. 4a, the method comprises a first step of selecting a first sub-region {tilde over (Z)} of the part 2 from a first image I.sup.A(Z) of a region Z ({tilde over (Z)}⊂Z) of the part 2.

(31) This selection may be performed manually by an operator who visualises and selects a sub-region that contains an echo, the origin of which is identifiable (echo of the bottom of the part, or defect echo for example). In a variant, the selection of the sub-region may be automated, for example using automatic image analysis tools able to plot the contours of sub-regions having the largest amplitudes (typically by thresholding of the image and conservation of the points for which the amplitudes are above the threshold).

(32) The method then comprises a step of determining, for each point P of the first selected sub-region {tilde over (Z)}, times of flight of the paths according to a first reconstruction mode going through the point P from a transmitter i to a receiver j for the set of M*N transmitter-receiver couples. In other words, the times of flight are extracted according to the first reconstruction mode associated with each point of the selected sub-region for a given reconstruction mode. One then forms, for each of the M*N transmitter-receiver couples, a sub-set of times of flight T.sub.ij.sup.A({tilde over (Z)}) comprising the set of times of flight according to the first reconstruction mode (A mode) from the transmitter i to a receiver j going through one of the points of the selected sub-region {tilde over (Z)}.

(33) In a possible embodiment of the invention, the method comprises a step of reconstructing said first image I.sup.A(Z) of the region Z by a synthetic focusing according to the first reconstruction mode of ultrasound waves (A mode) exploiting, for each point P of the region, the M*N ultrasound signals S.sub.ij(t) considered, for each transmitter-receiver i-j couple, at a time of flight T.sub.ij.sup.A(P) corresponding to the path according to said first reconstruction mode from the transmitter i to the receiver j while going through the point P of the region. The set of these times of flight, calculated for all the transmitter-receiver i-j couples and for all the points P of the reconstructed region Z, is denoted T.sup.A(Z)={T.sub.ij.sup.A(Z)}.sub.i=1 . . . M, j=1, . . . N. Within the scope of this embodiment, the different sub-sets of times of flight T.sub.ij.sup.A({tilde over (Z)}) are then known from the reconstruction of the first image according to the first A mode.

(34) The method further comprises a step of determining a second sub-region of the part, a point P of the region Z belonging to the second sub-region when a time of flight T.sub.ij.sup.B(P) of the path according to a second reconstruction mode (B mode) going through the point P from a transmitter i to a receiver j of said set of M*N transmitter-receiver couples coincides with a time of flight of a path according to the first reconstruction mode going through one of the points of the first selected sub-region from the transmitter i to the receiver j (T.sub.ij.sup.B(P)∈T.sub.ij.sup.A({tilde over (Z)})).

(35) As an example, the echoes present in the selected sub-region {tilde over (Z)} of FIG. 4a induce artefacts in an image reconstructed according to the B mode in the area surrounded by dotted lines in FIG. 4b which comprises the points for which the time of flight in B mode corresponds to a time of flight in A mode going through a point of the selected sub-region {tilde over (Z)}. It will thus be understood that the invention performs a transformation in B mode of the selected sub-region {tilde over (Z)}.

(36) In a possible embodiment of the invention, the method further comprises a step of reconstruction of a second image I.sup.B(Z) of the region Z of the part by a synthetic focusing according to the second reconstruction mode (B mode). But unlike a conventional reconstruction exploiting the set of times of flight for ail the transmitter-receiver couples and all the points of the region, the reconstruction of the second image according to the invention implements an isolation, or a contrario a filtering, of reconstruction artefacts.

(37) The method may thus comprise a reconstruction of a second image I.sup.B(Z) of the region Z of the part according to the second reconstruction mode (B mode) exploiting, for each point P of the region Z, and for each transmitter-receiver i-j couple, an ultrasound signal S.sub.ij(t) considered at a time of flight T.sub.ij.sup.B(P) of the path according to the second reconstruction mode from the transmitter i to the receiver j while going through the point P of the region, uniquely if said time of flight T.sub.ij.sup.B(P) coincides with (isolation of artefacts), respectively differs from (filtering of artefacts), a time of flight of a path according to the first reconstruction mode going through one of the points of the selected sub-region from the transmitter i to the receiver j.

(38) In other words, the reconstruction of the second image according to the B mode exploits, during an isolation of artefacts, and for a given transmitter-receiver i-j couple, uniquely the times of flight extracted beforehand forming the sub-set T.sub.ij.sup.A({tilde over (Z)}). A contrario, the reconstruction of the second image according to the B mode exploits, during a filtering of artefacts, and for a given transmitter-receiver i-j couple, uniquely the times of flight not belonging to the sub-set extracted beforehand T.sub.ij.sup.A({tilde over (Z)}).

(39) In the case of an isolation, the indications present in the selected sub-region {tilde over (Z)} from the first image I.sup.A(Z) have the same origin as those present in the reconstruction of the second image. In the example of FIGS. 6a and 6b, it is then possible to deduce that the indications linked to the echo from the bottom of the part and to the detected defect echo (zone {tilde over (Z)}) according to the first A mode induces the artefacts surrounded by dotted lines in the second image I.sup.B(Z).

(40) To carry out the isolation of reconstruction artefacts in the second image, one determines, for each point P of the region Z, a sum of amplitudes of the acoustic signals only taking account of the acoustic signals received at a receiver j from a transmitter i when the time of flight T.sub.ij.sup.B(P) of the path according to the second reconstruction mode from the transmitter i to the receiver j while going through the point P coincides with a time of flight of a path according to the first reconstruction mode from the transmitter i to the receiver j while going through one of the points of the selected sub-region (T.sub.ij.sup.B(P)∈T.sub.ij.sup.A({tilde over (Z)})).

(41) Thus, during the reconstruction according to the B mode, the contributions S.sub.ij(T.sub.ij.sup.B(P)) of the points for which the times of flights coincide in B mode with those of the selected sub-region in A mode are only considered.

(42) For the isolation, the sum of amplitudes of the acoustic signals determined at each point P may thus be expressed according to

(43) $I^B\left(P\right)=\underset{i=1}{\overset{M}{.Math.}}\underset{j=1}{\overset{N}{.Math.}}{A}_{ij}^B\left(P\right),$
where the amplitude A.sub.ij.sup.B(P) is given by

(44) $A_{ij}^B\left(P\right)=\{\begin{array}{cc}{S}_{ij}\left(T_{ij}^B\left(P\right)\right)& \mathrm{if}{T}_{ij}^B\left(P\right)\in T_{ij}^A\left(\tilde{Z}\right)\\ 0& \mathrm{otherwise}\end{array}$

(45) To carry out the filtering of reconstruction artefacts in the second image, one determines, for each point P of the region Z, a sum of amplitudes of the acoustic signals only taking account of the acoustic signals received at a receiver j from a transmitter i when the time of flight T.sub.ij.sup.B(P) of the path according to the second reconstruction mode from the transmitter i to the receiver j while going through the point P does not coincide with the times of flight of the paths according to the first reconstruction mode from the transmitter i to the receiver j while going through the points of the selected sub-region (T.sub.ij.sup.B(P).Math.T.sub.ij.sup.A({tilde over (Z)})).

(46) Thus, during the reconstruction according to the B mode, the contributions S.sub.ij(T.sub.ij.sup.B(P)) of the points for which the times of flights coincide in B mode with those of the selected sub-region in A mode are not considered.

(47) For the filtering, the sum of amplitudes of the acoustic signals determined at each point P may thus be expressed according to

(48) $I^B\left(P\right)=\underset{i=1}{\overset{M}{.Math.}}\underset{j=1}{\overset{N}{.Math.}}{A}_{ij}^B\left(P\right),$
where the amplitude A.sub.ij.sup.B(P) is given by

(49) $A_{ij}^B\left(P\right)=\{\begin{array}{cc}0& \mathrm{if}{T}_{ij}^B\left(P\right)\in T_{ij}^A\left(\tilde{Z}\right)\\ S_{ij}\left(T_{ij}^B\left(P\right)\right)& \mathrm{otherwise}\end{array}$

(50) Given the celerity difference existing between the waves L and T, the analysis of the images is preferentially carried out by considering firstly the fastest modes of propagation (direct LL for example) then the slowest (direct LT and corner echo LLL for example). In fact, in the case of the direct LL mode, the risks of presence of reconstruction artefacts stemming from a slower mode of propagation are reduced. In particular, for the sub-region considered, this mode will not be able to contain any indication linked to a geometry echo detected according to direct LT or corner echo LLL modes, which makes the nature of the observable indications more easily interpretable. FIG. 5 gives in this respect an example of image reconstructed according to the direct LL mode in which may be observed the presence of an echo from the bottom of the part EF right in front of the transducer, of a geometry echo EB linked to the welding bead, of indications linked to the echo EE of emission and an echo ED linked to the emerging cut.

(51) The previously described method may be applied to these different indications in order to predict the artefacts that will be associated therewith in other reconstruction modes.

(52) FIGS. 6a-6c give an example of implementation of the method during an isolation of artefacts in a reconstruction according to the direct LT mode. FIG. 6a is an image reconstructed according to the LL mode (A mode) in which echoes may be observed in the selected sub-region {tilde over (Z)}. These echoes induce artefacts in an image reconstructed according to the LT mode (B mode) in the area surrounded by dotted lines in FIG. 6b which comprises the points for which the time of flight of the path in LT mode correspond to a time of flight of the path in LL mode going through a point of the selected sub-region {tilde over (Z)} (transformation in B mode of the sub-region {tilde over (Z)}). FIG. 6c illustrates a reconstruction in standard LT mode. Comparison of FIGS. 6b and 6c then makes it possible to interpret the nature of a large number of indications, in particular the presence of artefacts A.

(53) FIGS. 7a-7c illustrate for their part an example of implementation of the method during a filtering of artefacts in a reconstruction according to the LT mode. FIG. 7a corresponds to FIG. 6c and illustrates a reconstruction in standard LT mode. FIG. 7b corresponds to FIG. 6b and illustrates a transformation in B mode of the selected sub-region {tilde over (Z)}. FIG. 7c illustrates for its part the result of the filtering according to the invention, where the second image of the region is cleared of the artefacts A of FIG. 6c induced by the echoes observed in the sub-region {tilde over (Z)} of the first image.

(54) FIGS. 8a-8c, respectively 9a-9c, give an example of implementation of the method during an isolation of artefacts in a reconstruction according to the corner echo LLL mode. FIG. 8a, respectively 9a, is an image reconstructed according to the direct LL mode (A mode) in which echoes are observable in a selected sub-region {tilde over (Z)} 1, respectively {tilde over (Z)} 2. These echoes induce artefacts in an image reconstructed according to the LLL mode (B mode) in the area surrounded by dotted lines in FIG. 8b, respectively 9b, which corresponds to the transformation in B mode of the sub-region {tilde over (Z)}. FIG. 8c, respectively 9c, illustrates a reconstruction in standard LLL mode. Comparison of FIGS. 8b and 8c, respectively 9b and 9c, then makes it possible to interpret the nature of a large number of indications, in particular the presence of artefacts.

(55) FIGS. 10a-10d illustrate for their part an example of implementation of the method during a filtering of artefacts in a reconstruction according to the LLL mode. FIG. 10a corresponds to FIGS. 8c and 9c and illustrates a reconstruction in standard LLL mode. FIG. 10b corresponds to FIG. 8b and illustrates transformation in B mode of the sub-region {tilde over (Z)} 2. FIG. 10c corresponds to FIG. 9b and illustrates the transformation in B mode of the sub-region {tilde over (Z)} 1. FIG. 10d illustrates for its part the result of the filtering according to the invention, where the second image of the region is cleared of the artefacts induced by the echoes observed in the sub-regions {tilde over (Z)} 1 and {tilde over (Z)} 2 of the first image.

(56) The invention is not limited to the previously described method, but also extends to an ultrasound imaging device for imaging a part 1 capable to implement this method, particularly a device comprising M transmitters of an ultrasound signal in the part to be imaged 1 and N receivers of an ultrasound signal S.sub.ij(t) coming from the part to be imaged, the index i designating one of the transmitters, the index j designating one of the receivers, and t representing the time, and a processor coupled to the M transmitters and the N receivers, the processor being configured to: select a first sub-region of the part from a first image of a region of the part 1, determine, for each point P of the first selected sub-region, times of flight of the paths according to a first reconstruction mode going through the point P from a transmitter i to a receiver j for a set of M*N transmitter-receiver couples of an ultrasound signal; determine a second sub-region of the part, a point P* of the region belonging to the second sub-region when a time of flight of the path according to a second reconstruction mode going through the point P* from a transmitter i to a receiver j of said set of M*N transmitter-receiver couples coincides with a time of flight of a path according to the first reconstruction mode from the transmitter i to the receiver j going through one of the points of the first selected sub-region.

(57) And the invention also extends to a computer program product comprising code instructions for the execution of the steps of the previously described method, when said program is run on a computer.

(58) It will have been understood from the present description that the invention makes it possible to analyse the origin of indications present in reconstructions by multimodal synthetic focusing and thus to distinguish echoes due to defects from those due to the geometry of the part. The invention has moreover the advantage of applying to all multimodal reconstructions. It moreover uses the same parameters as standard synthetic focusing, and thus does not require additional calculations or measurements. It can in addition be used not only for an identification of artefacts, but also for filtering of the latter.