FR2839876A1 - Ultrasound analysis technique for bone uses transmitter and multiple receivers producing matrix for analysis and deduction of bone characteristics - Google Patents

Abstract

The ultrasound probe consists of an ultrasound transmitter (3) and a series of receivers (4) distributed closely along the propagation direction. The excitation level is read on each receiver (4) at each sampling time, thereby providing a spatio-temporal image of the excitations of the receivers. This is analyzed using a Fourier transform, leading to a curve whose shape allows deduction of bone properties. The ultrasound process uses a probe consisting of at least one ultrasound transmitter (3) and a series of receivers (4) distributed relatively closely along the propagation direction. Ultrasound excitations are applied, and the excitation level is read on each receiver (4) at each sampling time, thereby providing a spatio-temporal image of the excitations of the receivers. The image is a matrix within which each column may, for example, correspond to one receiver, and each line may correspond to one sampling time. The matrix is processed with a digital bi-directional Fourier transform to obtain a spatial frequency-temporal frequency diagram. On this diagram the various vibratory modes generated by excitation of the observed medium form curves, and some geometrical features of these curves enable geometrical or mechanical characteristics of the observed medium to be deduced.

Description

<Desc / Clms Page number 1>

Method, probe and apparatus for ultrasonic characterization of a study medium, in particular the cortical layer of a bone.

Field of the invention
The present invention relates to a method of characterizing a study medium by applying ultrasonic excitations on an accessible surface.

 The accessible surface may belong to the study medium itself, or to an intermediate medium that the ultrasound excitations will have to cross to reach the study environment.

 Indeed, the present invention is particularly interested in the characterization of the cortical layer of a bone, non-invasively and in vivo, the accessible surface then being the surface of the skin while a spacer medium constituted Soft tissue is interposed between the accessible surface and the cortical layer of the bone.

 The present invention also relates to a probe and an apparatus for the implementation of the method and in particular for the non-invasive and in-vivo characterization of the cortical layer of a bone.

Background, State of the art
It is known in the art that the propagation of ultrasound in a material depends on the mechanical properties and geometry of that material. Thus, many devices use the propagation of ultrasound in the bones, in particular the measurement of propagation velocity, in order to detect the changes undergone by the bone during osteoporosis or other bone pathologies known to affect the bone. resistance or bone integrity. The majority of

<Desc / Clms Page number 2>

 Commercial devices use a so-called transverse propagation technique in which two ultrasonic transducers are arranged on either side of the bone to be studied, to study the propagation of ultrasound through the bone [Laugier / Berger patent WO 9526160]. The particular arrangement of the sensors inherent in such a technique limits it to the study of peripheral bone sites such as the heel, the phalanges or the distal radius, which are accessible to the measurement on their two opposite faces. Techniques allowing the specific study of the cortical bone, bone which presents only one accessible face, applicable on long bones such as the tibia or the body of the radius have been proposed since many years, putting into play the Ultrasound propagation along the surface of the bone [G. Lowet and G. Van der Perre, "Ultrasound velocity measurement in long bones: Measurement method and simulation of ultrasound wave propagation," J Biomech 29 (10), 1255-1262 (1996), Siegel I., GT Anast and T. Melds. , "The determination of fracture healing", "Surgery, Gynecol, Obstetrics, 327-332 (1958).]. The present invention particularly relates to these propagation phenomena along the surface of the bone.

 For the evaluation of bone in vivo, it is known that the speed of propagation, which depends on the mechanical and geometrical properties of the bone, may allow a diagnosis of the state of the bone [C. Njeh, D. Hans, T. Fuerst, C. Glüer C. and H. K. Genant (1999). Quantitative ultrasound assessment of osteoporosis and bone status. London, Martin Dunitz.]. It has thus been shown that the measurement of the velocity of an ultrasonic wave (frequency of the order of 1 MHz) propagating along the surface of the cortical bone at the radius makes it possible to discriminate a healthy population from a osteoporotic population who has had a hip fracture [M. Weiss, A. Ben-Shlomo, P. Hagag and S. Ish-Shalom, "Discrimination of proximal hip fracture by quantitative ultrasound measurement at the radius", Osteoporosis Int. 11, 411-416 (2000)].

<Desc / Clms Page number 3>

 Known devices, by making it possible to determine the speed of propagation of the first signal reaching one or more receivers following the application of an ultrasound excitation, allow only a relatively limited characterization of the study medium and only allow to open, depending on the result obtained, a number of diagnostic hypotheses, between which the practitioner may have difficulty making a choice.

 The object of the present invention is thus to propose a method of characterization which makes it possible to considerably refine the results provided by the application of ultrasound excitations to a study medium such as the cortical layer of a bone.

 Another object of the present invention is to provide a probe and an apparatus which are suitable for the implementation of the process.

 According to a first aspect of the invention, the method of characterizing a study medium by applying ultrasound excitations on an accessible surface, in particular to characterize the cortical layer of a bone separated from a surface accessible by a intermediate medium, in particular soft tissue, in which: - by at least one emitter placed on the accessible surface, excitation signals are sent to the study medium, - by pre-positioned receptors with respect to other in a probe which is placed on the accessible surface, signals induced along the study medium are received by the excitation signals, is characterized in that the excitation states, which are successive in the field, are detected. time, multiple receivers distributed over the path of the signals, and is established with the result of these detections a spatio-temporal image of vibratory signals transmitted along the study medium.

 The spatio-temporal image can be in the form of a matrix in which each receiver is represented, for example, by a column while each

<Desc / Clms Page number 4>

 moment, for example each sampling instant, corresponds to a row of the matrix. Thus, using the above example, each column indicates the variation in time of the excitation level of a receiver associated with this column, while each line indicates the distribution of the excitation levels along the study medium. as they were perceived by the receivers at a given instant, associated with this line.

 This records all the information that is accessible with a given number of detectors and a given sampling frequency. The sampling frequency is chosen much larger than the center frequency of the excitation signal, so that the excitation levels recorded unambiguously describe the excitation alternations related to the vibratory phenomena in the study environment.

 To eliminate measurement errors, each measurement can be carried out by means of n tests, for example ten tests, and it is possible to take for example, for each instant and each receiver, the average of the n excitation levels recorded during the tests. n tests.

 When an intercalated medium is interposed between the study medium and the receivers, this intercalary medium delays the reception of each signal feature with respect to the moment at which the same signal feature left the study environment to begin cross the intermediate medium. Such a delay does not affect the quality of the results obtained. But it is also possible that the probe and more particularly the direction of alignment of the receivers of the probe have an angle with respect to the direction of propagation in the study medium, in particular because of a variation of the thickness intermediate medium along the direction of propagation. The delay introduced by the intermediate medium then differs from one receiver to another and the spatio-temporal image perceived by the receivers presents a deformation in time compared to that which would be observed along the study medium. . It is possible to

<Desc / Clms Page number 5>

 correcting this deformation either by a known correction technique, or preferably by using transmitters placed on either side of the probe receiving zone as will be described later.

 It is possible to make several spatiotemporal images for the same study environment, by varying the central frequency of the ultrasound excitation. Indeed, the vibratory modes induced for different central frequencies are also different. Preferably, the center frequency is varied between about 100 kHz and 5 MHz.

 The image is preferably recorded by spatio-temporal sampling comprising: sending the signal in the form of shots; for each shot, selectively activating among the receivers a small number of receiver (s) whose state of being is noted; excitation at predetermined time intervals significantly smaller than the signal periods, different receivers being activated during different shots, - in the spatio-temporal image, bring to the same origin times the times corresponding to the shots from the same issuer.

 Spatiotemporal sampling typically involves activating only one receiver for each shot. Shooting is a transmission of a brief train of ultrasonic alternations as an excitation signal. The spatio-temporal sampling then consists in recording at each instant of sampling the excitation level of the activated receiver. If necessary, as explained above, for each receiver, it is possible to carry out n tests and take as a measure, for each sampling instant, the average of the n excitation levels recorded.

 This process is repeated for all the receivers successively, ensuring that the shot has exactly the same waveform for all the tests.

 Then, all the measurements taken to constitute the matrix corresponding to the desired spatio-temporal image are brought back to the same origin of time.

<Desc / Clms Page number 6>

 By this double sampling process, the means necessary to record and record the spatio-temporal image are considerably saved.

 However, it is also possible to have an excitation level measurement channel and a memory compartment for each receiver, these measurement channels and memory compartments being capable of operating independently and simultaneously with a common sampling timing.

 The spatio-temporal images provided by the method according to the invention are rich in information because they include the excitatory consequences of the different modes caused by the excitation signal in the study environment.

 The spatio-temporal image can be processed, in particular by digital two-dimensional Fourier transformation, to extract a spectrum of temporal frequencies and spatial frequencies, and in particular to establish a correspondence between the values of the spatial frequencies and the values of the temporal frequencies.

 This spectrum, which can also be visualized, shows, for example in graphical form, parameters from which one can deduce directly or indirectly, as will be described below, physical quantities relating to the study environment, in particular the velocity of propagation of the transverse wave (shear wave), the thickness of the propagation medium, in particular of the cortical layer of the bone, etc.

 According to a second aspect of the invention, the ultrasound probe for the implementation of a method according to the first aspect, in particular for bone auscultation, comprising at least one ultrasound emitter in at least one d-zone. emission and ultrasonic receivers in a reception area, distributed along an adapter blade, is characterized in that there are at least about ten receivers placed in the receiving area. Because of its high number of receivers, the probe offers a sufficiently precise mesh for

<Desc / Clms Page number 7>

 perform the spatial part of the spatiotemporal sampling according to the method.

 According to a third aspect of the invention, the apparatus for characterizing a study medium, in particular for characterizing a bone in vivo and in a non-invasive manner, is characterized in that it comprises: a probe according to second aspect of the invention; emission control means for controlling the emission of successive ultrasonic fires by the at least one transmitter; selective activation means for activating at least one receiver for each firing; and deactivating other firing means. , respectively, receivers, means for recording and recording, at successive sampling times, the state of excitation of the at least one activated receiver.

 Preferably, the apparatus comprises display means, such as a video screen, for displaying a diagram (spatial frequency) - (time frequency) extracted from the space-time image by appropriate mathematical processing. This visualization allows the practitioner to carry out a graphical identification of the information he is looking for, and also to guide himself to modify the experimental conditions, for example the central excitation frequency, according to the type of modal response he is looking for.

 Other features and advantages of the invention will result from the description below, relating to a non-limiting example.

 In the accompanying drawings: FIG. 1 is a schematic view of the probe in section in the plane of the transducers; - Figure 2 is a partial schematic view of the probe of Figure 1, on an enlarged scale, in section in a longitudinal plane perpendicular to the plane of the transducers; FIG. 3 is a block diagram of the apparatus according to the invention:

<Desc / Clms Page number 8>

 Figure 4 is a block diagram of a measurement cycle; FIG. 5 is a diagram of explanatory propagation paths of a first mode of evaluation of the speed according to the invention; Fig. 6 is an explanatory diagram of the alternative evaluation method; Fig. 7 is a diagram illustrating the calibration step; FIG. 8 is a space-time diagram relating to the evaluation of the propagation speed after calibration; Figures 9 to 12 are chart views (spatial frequency) - (time frequency) for four different values of the FE parameter.

 As shown in FIGS. 1 and 2, the probe is of linear array type consisting of planar, non-focussed piezoelectric ultrasound transducers. The transducers are divided into three zones, namely a central receiving zone 2 located between two emission zones 1. The reception zone 2 comprises equidistant transducers 4 operating as receivers and each transmission zone 1 comprises equidistant transducers 3 operating in transmitters.

Intrinsically, any transducer can be either transmitter or receiver, and its effective function in the probe depends only on the electronic transmitter / receiver device to which the probe is connected, and which will be described in detail later as part of the evaluation apparatus according to the invention.

 The probe is said to be unidimensional, in the sense that all the transducers 3,4 are aligned in one and the same rectilinear row, and are arranged in a common plane corresponding to the plane of FIG. 1. In a manner which is conventional for the commercially available echographic strips, the transducers 3,4, are arranged between a rear center 6 (FIG. 2), said backing, and an adapter blade

<Desc / Clms Page number 9>

 or front face 7 for contact with the body (the skin) of the patient, with the interposition of a gel.

 In each application, the dimensions of the surface of each transducer element are determined by the directivity and sensitivity of the elements. The transducers preferably have a width equal to # e / 2, where #e is the wavelength in the ultrasound water at the center frequency of the probe. This width makes it possible to obtain a probe with a large angular aperture. The width of the transducers is oriented parallel to the alignment direction Ox of the transducers 3 and 4 along the probe. In each zone, the transducers 3 or 4 are adjacent to each other. The length of each transducer 3 and 4, measured in the plane of the transducers (plane of Figure 1) but in the direction Oy perpendicular to their alignment direction Ox, is chosen sufficiently large to obtain a sufficiently sensitive element. In the non-limiting example of Figure 1, the length of the transducers is four times #e.

 The central frequency used may be for example between 100 kHz and 5 MHz, without limitation, for a probe used on the human or animal body. If the center frequency is 1 MHz, we have #e = 1.5 mm. The central frequency of the probe is the frequency calculated from the passage of the signal by two consecutive zeros in the same direction, that is to say for example by going from a positive value to a negative value of the intensity.

 Each zone 1 or 2 is framed by two elements 8 connected to the mass. Thus, when they are activated, the transducers 3 and 4 which are in limit of region 1 or 2 operate under the same conditions as a transducer 3 or 4 situated between two other transducers of the same region. It is further provided between the receiving region 2 and each of the two emission regions 1 an absorbent barrier 11 which in the example shown is made of cork. The barriers 11, called front-facing barriers, are intended to strongly attenuate the waves likely to propagate in the face.

<Desc / Clms Page number 10>

 before the probe. They each form an interruption in the adapter blade 7 (Figure 2), extend throughout the thickness zone (in the direction Oz) of the transducers 3 and 4, and protrude in the rear medium 6 without, however, crossing completely . Each barrier 11 has an anterior face 12 intended to be separated from the contact with the skin of the patient only by a sealing film which also covers the blade 7. The face 12 is flat and coplanar with the front face of the adapter blade 7. Each barrier 11 has in the direction Oy a dimension at least equal to the transducers, that is to say in the example four times #e. In the Ox direction, the barrier 11 extends over for example 10 mm.

 With the probe in the air, for a 160 Volt electrical signal emitted on a transducer on one side of a barrier 11, the electrical signal received by a transducer on the other side of the barrier is less than 160 microvolts , so that the amplitude attenuation is at least equal to a factor of 106, which equals 120 dB.

 In the example shown in Figure 1, the receiving zone comprises 32 transducers 4 adjacent to each other, and each emission zone comprises three transducers 3 adjacent to each other. There are therefore a total of 38 transducers, six elements 8 connected to the ground and two barriers 11, and thus the total length of the probe is of the order of 50 mm with the sizing exposed higher.

In practice, the reception zone can typically, but not exclusively, have 10 to 64 receivers depending on the use of the probe and the desired probe size. The choice depends on a compromise between the chosen central frequency and the site of the skeleton that is being explored.

 It is theoretically possible to provide a single transmitter 3 in each transmission zone 1. However, it is preferred to provide more than one and for example three as in the example shown, for two different reasons. The first reason is that if a transmitter 3 is defective, the probe is still usable with the other two transmitters. Another benefit of having multiple transmitters is to allow

<Desc / Clms Page number 11>

 simultaneously operating at least two transmitters 3 of the same transmission zone 1 by temporally stalling them so that they are phase-matched in phase. In other words, during a transmission, the signal emitted by the transmitter 3 closest to the reception zone is superimposed in phase matching with the signal coming from further along, along the interface to be examined, by from the other transmitter in operation, further from the reception zone 2.

 The probe further comprises a plastic protective casing 13 which is only shown schematically in FIG. 2. The sealing film covering the adapter blade 7 and the anterior face 12 of the absorbent barriers 11 is visible on an entire front face of the casing . On the opposite side to the probe itself, the housing 13 encloses a connector 14 through which the transducers 3 and 4 are connected to a single cable 16 in which the individual electrical connections with each transducer are spatially collected. In the schematic representation of FIG. 2, the mass 17 is represented inside the housing 13, but in practice it is an additional wire in the cable 16, via the connector 14.

 As shown in FIG. 3, the apparatus according to the invention for evaluating in vivo and in a non-invasive manner the mechanical properties of a bone, comprises: a probe 21 which can be as described with reference to FIGS. 2 and which is only very schematically represented; a transmitter / receiver electronic module 22 which is connected on the one hand to the probe 21 by the cable 16 and, on the other hand, to a PC 23 in which a digital acquisition card of the signal 24 is installed, as well as a software part 26 for controlling the digital signal acquisition card 24 and the transmitter / receiver electronic module 22. The software 26 also stores the data in time and space, and the processing of these data to provide the expected evaluations . The data

<Desc / Clms Page number 12>

 in time and space above include in particular the location of the emitters and receivers activated as well as the corresponding transmission and reception times.

 The electronic module 22 contains a transmitting part (left part in FIG. 3) and a receiving part (right part in FIG. 3), as well as a power supply 27. The electronic module comprises in particular an interconnection matrix 28 comprising -Even an emission part 29 provided with terminals such as 31 to which are respectively connected the cable links 16 connected to the emitters 3 of the probe 21, and a receiving part 32 having terminals 33 to which the links of the cable 16 are connected which are connected to the receivers 4 of the probe 21.

 The transmission part of the module thus comprises q independent transmitting channels 34, for example sixteen channels in a typical embodiment, by which electrical excitations of amplitude 160 V are sent to the emitters 3 of the probe 21.

The sixteen channels can transmit either one at a time, or several at a time with adjustable relative delays so as to achieve a beam synthesis, ie the spatial phase concordance described above. The reception part comprises r channels 36, for example 64 in a typical embodiment, which connect each receiver 4 of the probe 21, via the reception part 32 of the interconnection matrix 28, with a multiplexer 37 which allows the successive reading of the signals. channels 36by the acquisition card 24.

 There is still in the transmitter / receiver electronic module a sequencer and gain control stage 38 which is connected to a control input 39 of a beam synthesis stage 41 from which the emitter channels 34 leave.

By the control input 39, the sequencer 38controls the stage 41 so as to define on the one hand the times of the shots (signal transmissions by the emitters 3 of the probe 21), and on the other hand the transmitters concerned by the shot to be made at each moment of shooting. The apparatus according to the invention comprises in

<Desc / Clms Page number 13>

 in addition to a control (not shown) for adjusting the central excitation frequency between about 100 kHz and about 5 MHz.

 The sequencer 38 is also connected to a control input 42 of the multiplexer 37 to define, at each instant, the number of the reception channel 36 whose excitation state must be transmitted to an output amplifier 43.

 The sequencer and gain control stage 38 is further connected to a control input 44 for controlling the gain G of the output amplifier 43.

 The links between the PC 23 and the electronic module 22 comprise: a link 46, typically but not limited to a serial link, with the sequencer 38, allowing the software 26 to program the firing sequences to be performed by the transmitters 3 , as well as the position of the receiver 4 whose signal must be transmitted to the input of the amplifier 43 for each shot, respectively; a synchronization link 47 by which the acquisition card 24 is synchronized with the shots set by the sequencer 38; and a data link 48 between the output of the amplifier 43 of the module 22 and the acquisition card 24, to enable reading and recording of the data in the PC computer 23.

 The method according to the invention gives access to the knowledge of various propagation speeds corresponding to various vibratory modes. Some speeds are accessible only by mathematical processing to establish a diagram (spatial frequency) - (time frequency) according to process steps which will be described later. Other speeds are also available on the time data relating to certain peculiarities of the signals propagating in the study environment.

 The use of these space-time data requires the determination of the respective travel times of the signal to the various receivers 4 of the probe. As the module

<Desc / Clms Page number 14>

 22 monitors only a single receiver 4 at a time, it detects the aforementioned travel time difference not by detecting the arrival of the signal consecutive to the same firing successively on the two detectors, but by a method d spatio-temporal sampling consisting in making two different shots by activating for each shot a respective one of the two receivers, after which the difference in signal travel time between the two receivers can be calculated by calculating the difference between the times of respective paths from the transmitter to each of the receivers.

 In addition, to know the space-time diagram of a signal feature along the bone, that is to say the diagram giving the distance traveled by this feature along the interface as a function of time, is performed respective shots for each of the receivers 4 so as to obtain, in the example of a 32-receiver probe, 32 points of the space-time diagram.

 Moreover, in order to minimize the effects of measurement errors, each measurement of the travel time between one or more transmitters and a given receiver 4 is carried out n times and the travel time measured on the average of the shots is taken as the travel time. .

 This process is visualized in FIG. 4. A main loop 49 is traversed for each of the r reception channels 36. For each receiver 4 having the position j (j varying from 1 to r), two sets of n shots are produced. The first series of shots corresponds to the emission by a group of transmitters located on one side of the reception zone, the second series corresponding to the emission by the other group of transmitters 3 located on the other side from the reception area 1.

For each receiver 4, and for a zone of emission, the system emits n shots of which one is averaged, and one then records the average shot resulting from the average of the shots. This average allows a reduction of the random noise. In total, a measurement cycle corresponds to 2. r.n shots that lead to the recording of 2r averaged time signals. In addition, it is possible to carry out as many successive cycles as desired, in

<Desc / Clms Page number 15>

 specifying a number p of measuring cycles in the control software 26 of the apparatus of FIG. 3. This gives a final data file containing 2. r.p averaged time signals. At the end of each series of r. The data is transferred from the card 24 to the random access memory of the computer 23. The data set is left in the RAM until all the cycles are completed. At the end of p cycles, the data is saved to a file. At the end of each series of r. n shots (2. p series in total), we therefore have data in RAM for possible real-time processing.

 The control software 26 installed in the computer 23 to control the electronic module 22 makes it possible to specify all the parameters necessary for the definition of a complete acquisition: the name of the recording file and the path in the computer ; the choice of transmitters 3 and receivers 4 involved; the possible relative delays between the firing of a group of transmitters 4 in order to achieve the beam synthesis effect described above: the relative delays are adjusted according to the effect observed on the receivers (4), where the search is carried out a superposition of the signals corresponding to the emitters simultaneously activated; the gain associated with each receiver (4), because it is for example necessary to provide a stronger gain for the receivers (4) furthest removed from the emitter group (3) which is activated, or because a gain more strong is necessary for some (4) less sensitive receptors, which can be determined by prior tests, or depending on the individuals; the number of shots (n) to average for each receiver; the number of cycles (p) of a complete acquisition; the possible delay between these cycles; the time between successive shots;

<Desc / Clms Page number 16>

 the sampling frequency, that is to say the rate at which are noted the successive excitation states of the receiver 4 which is activated. This frequency is at least equal to 10 times, preferably 50 times the central frequency of the ultrasound signal emitted; the duration of the acquired signals (thus the number of acquisition points per time signal), which is a significant parameter when it is desired to record information in addition to that of the arrival time of the first signal.

 The device has been optimized so that a large number of measurements can be made as quickly as possible. The minimum time between two successive shots is for example 125 s in a typical embodiment. This minimum time can be dictated by a reloading time of a capacitor, or by a delay of disappearance of the previous ultrasonic signal. By way of example, the time taken to carry out a complete measurement cycle with n = 10 shots per receiver, and if the probe comprises 50 receivers (4), is typically 125 s × 2 × 10 × 50 = 125 ms. Therefore, even if the probe is held manually by an operator on the body of a patient, it is almost immobile during a measurement cycle.

 For each shot, the excitation state of the receiver 4 that is activated is read and recorded at each sampling instant. A base of values is thus obtained giving the state of excitation of the receiver as a function of time during the duration of the measurement relative to this shot, for example 125 s.

As instant of arrival of the first signal, it is possible to take the first sampling instant when the excitation state exceeds a certain predetermined level, the first passage by a maximum, the first change of sign, etc. Once the shots are fired, all the readings are taken to the same origin of the times and the average of the shots is averaged by calculating the average level of excitation of each receiver at each instant of sampling. An average reading is also obtained on which

<Desc / Clms Page number 17>

 for example, the signal characteristic serving as a criterion for the arrival of the first signal is sought.

 The lifting method which has just been described also makes it possible to produce a spatio-temporal image of the excitation modes which propagate along the interface as a consequence of a shot. The spatio-temporal image in the sense of the present invention is a two-dimensional matrix of the excitation states as a function of the receiver, each receiver position corresponding for example to a column of the matrix, and each sampling instant (identified by its delay with respect to the moment of firing) corresponding for example to a line of the matrix.

 For each receiver, and therefore for each column in the example, the succession of values used may be, according to the applications: - The values recorded during any one, for example the first of the n shots in a given direction for this receiver ; - The values recorded for this receiver in a direction during that of the shots that best corresponds to the average value calculated for the arrival time of the first signal; - For each instant the average of the excitation values recorded for this moment during the n shots in a direction of propagation.

 It is also possible, not only for the arrival of the first signal but also for each point of the signal, to calculate by the method according to the invention the time taken by this point of the signal to propagate from one receiver to the other and to obtain a spatio-temporal image freed from the deformation introduced by the intermediate medium.

 The image thus obtained by one or the other method reveals the direct transmission mode of the signal along the bone, and so-called longitudinal longitudinal modes, corresponding to compression-expansion vibrations with forces parallel to the direction of propagation, and so-called transverse induced modes, corresponding to constraints of

<Desc / Clms Page number 18>

 shear between planes perpendicular to the direction of propagation. The spatio-temporal image is shifted and possibly distorted by the soft tissue effect, but its configuration remains identifiable and analysable for diagnostic purposes.

 The type of acquisition (time / space) thus makes it possible to carry out an analysis (time frequency) - (spatial frequency) in the sense that the lines (in the aforementioned example) of the matrix give an image of the excitation states of the row of receivers at a given instant, and the columns (still in the aforementioned example) of the matrix give an image of the successive excitation states of a given receiver in the time that follows a shot.

 The spatio-temporal image is obtained with very simple means since only one receiver is activated at every moment. Spatial-temporal sampling is the method used, which consists in noting the state of excitation of only a small number of detectors (a single detector in the preferred example) for each shot (spatial sampling) and to be recorded for this detector the state of excitation at successive instants following this firing (temporal sampling), then to obtain the spatio-temporal image, that is to say the two-dimensional matrix by bringing all the readings to the same origin of time, for example the instant of fire.

 When performing p successive measurement cycles, the system offers the possibility of obtaining a real time value of the speed of the first signal, calculated from two successive series of r. n shots, namely a series in each direction, the data associated with a cycle being present in the random access memory of the computer 23. The time between two successive shots is also adjustable between the aforementioned technical minimum of for example 125s, at a higher value, for example for recording signals beyond 125 s or real-time visualization of speed versus cycle.

This can, among other things, make it possible to visualize in real time the speed of the first signal measured as a function of the position of

<Desc / Clms Page number 19>

 the probe on the bone, to identify for example a zone of extreme speed.

 The data processing software uses the data read by the acquisition card 24, either in post-processing from the data files recorded on the hard disk of the computer, or directly after each cycle from the random access memory of the computer for real-time processing. The use of ultrasound shots in two opposite directions to the same receivers 4 makes it possible to calculate the speed of a signal along the underlying refraction interface, that is to say in the application considered on the surface of the the bone, regardless of the effect of the soft tissues that are interposed between the interface and the probe, and regardless of the effect of a possible inclination between the studied direction of the interface and the plane of the receptors.

 This method of eliminating the soft tissue effect and inclination will be explained hereinafter with reference to the propagation of the first signal, but the method is applicable to the other types of signals that are produced in the bone for a shot composed by a train of individual alternations.

 We will first describe how from a pair of receivers 4 having between them a known distance and which receive the wave radiated by the interface in one direction of propagation then in the other, can be obtained by a calculation very simple wave velocity in the bone, that is to say mainly in the superficial layer of the bone adjacent to the soft tissue interface, regardless of soft tissue thickness and even d a possible inclination [alpha] between the plane of the two receptors composing the pair, and the direction of propagation studied along the surface of the bone.

 FIG. 5 shows a portion of the probe 21 placed on the surface of the skin 51 of a patient whose bone to be examined is designated 52 and the soft tissues interposed between the probe and the bone are designated by 53. Thicker lines show the geometric paths followed by the wave arriving in two receivers located in B and E, separated by a

<Desc / Clms Page number 20>

dans laquelle VTm et Vos sont respectivement la vitesse du signal dans les tissus mous 53 et dans l'os 52. distance #r. The wave propagating along the bone 52 is at the same time refracted back through the soft tissues towards the probe at an angle (3 which, in a known manner, is given by the relation

in which VTm and Vos are respectively the signal velocity in the soft tissues 53 and in the bone 52.

 The critical angle P thus determined also determines the points A and C from which, along the bone 52, the refracted signals which will be received by the receivers situated at B and E respectively, leave.

Moreover, the difference in arrival time at the two receivers is written:

Point D is the point on the straight line of the CE segment such that CD = AB.

S1 l'onde arrive dans l'autre sens le long de la surtace de l'os, il suffit de changer le signe de a dans l'expression précédente. Finalement, l'écart de temps d'arrivée s'écrit, pour les deux sens de propagation: (2)

The different relations which have just been exposed make it possible to deduce, after a few trigonometric manipulations, the following expression:

S1 the wave arrives in the other direction along the surtace of the bone, it is enough to change the sign of a in the preceding expression. Finally, the arrival time difference is written for both directions of propagation: (2)

<Desc / Clms Page number 21>

 the index corresponding to the two possible directions of propagation along the surface.

 The interest of this two-way shooting process towards the same two receivers is to provide very simply the velocity in the bone starting from # t + and # t-: by summing # t + and # t-, we simply obtain .

L'angle a est inconnu a priori lors d'une expérience in vivo, mais on peut néanmoins raisonnablement le supposer petit, et écrire cos([alpha])#1, à l'ordre 2 en a. On a alors très simplement comme première approximation Vosao de la vitesse de l'onde dans l'os la relation (4)

Par exemple, pour a<4 , l'erreur relative commise est inférieure à l-cos(a)=0.2 %. (3)

The angle a is unknown a priori during an in vivo experiment, but we can nevertheless reasonably suppose it small, and write cos ([alpha]) # 1, at order 2 in a. We then very simply as the first approximation Vosao of the speed of the wave in the bone the relation (4)

For example, for a <4, the relative error committed is less than l-cos (a) = 0.2%.

This formula is simply to take for the signal travel time along the study environment:

The actual travel time for an identifiable characteristic of the signals propagating in the study medium between two points corresponding to two receivers 4 is thus known.

 An interest of the correction method which has just been described is to provide a very accurate estimation of speeds and travel times along the surface of the bone, knowing only very approximately the speed in the soft tissues, speed that it is enough to take equal to 1500 m / s (approximate velocity in the soft tissues given in the literature). The speed in the bone is obtained very simply

<Desc / Clms Page number 22>

 calculating the sum At + + At- arrival time differences in both directions, with an order 2 approximation is clearly sufficient for in vivo measurements. (If higher precision is required, it can be obtained by estimating a as previously described). This correction of the angle a is possible only because the reception zone is common to shots in both directions, a condition necessary for the angle a between the receivers used and the surface of the bone to be the same for both shots in opposite directions.

qui ont des valeurs absolues différentes si a est non nul, ce qui est le cas illustré à la figure 6. La vitesse corrigée (à l'ordre 2 en a) est simplement obtenue en procédant à des substitutions dans l'expression (4) qui devient

valeur qui peut être précisée par le procédé itératif déjà décrit. Ce procédé de mesure par la pente fournit une mesure plus robuste qu'une simple mesure de #t+ et #t- obtenue à #r #r partir de deux récepteurs, puisqu'on intègre un plus grand The probe according to the invention has a group of receivers 4, rather than a single pair of receivers, in order to perform spatio-temporal sampling. The simplest way to calculate the corrected speed to eliminate the effect of the angle α between the receiver alignment and the probe surface, is to plot the graph of the arrival times at the receivers as a function of the position receivers, this for shots in both directions, as shown in Figure 6, where the receivers 4 are individually designated by Rl ...... Rr. We obtain two series of points, which are aligned on two lines of slopes

which have different absolute values if a is non-zero, which is the case illustrated in figure 6. The corrected speed (at order 2 in a) is simply obtained by making substitutions in expression (4) who becomes

value that can be specified by the iterative method already described. This slope measurement method provides a more robust measure than a simple measure of # t + and # t- obtained at #r # from two receivers, since a larger

<Desc / Clms Page number 23>

 number of receivers, which compensates for measurement uncertainties if they are of a random nature. This method gives very accurate results if the sensors are perfectly aligned on the same line, and their relative positions on this line are known precisely (even if they are not evenly distributed). The relative accuracy required on the receiver positions is at least 1%, if you want to get the Vos speed with less than 1% error. However, it is technologically difficult to position the receivers with such precision in the manufacture. To remedy this, it is preferred, according to the invention, to calibrate the probe, preferably according to a method of calibrating the probe, using the angle correction previously described.

Method of calibrating the probe

 The problem of the relative position of the receivers 4 in the probe 21 is of two kinds: distances between adjacent receivers are unknown on the one hand, and on the other hand the angle made by a pair of adjacent receivers with the surface of the os depends on the pair considered. This is illustrated in Figure 7, with six receivers. The basic process described so far automatically corrects the angulation of each pair, but does not correct the uncertainties on the distance between successive receivers. The calibration principle that will now be described consists in determining the distance between the two receivers of each pair. In practice, the probe is placed opposite a calibrated material in which the propagation velocity is known precisely, with the average alignment of the receivers approximately parallel to the surface of the material (so that all the angles considered are small ). We then measure the differences # t + and # t- of arrival time for each pair of adjacent receivers. For each pair of adjacent receivers, the difference in arrival time that would be obtained if the angle of the pair with the surface was zero is #t = # t + + # t- (at order 2 in a).

 2

<Desc / Clms Page number 24>

 Knowing the speed Vcal, which is equal to Vcal = #r, we deduce #t for each pair the distance #r = Vcal # #t which separates the 2 receivers.

 When measuring on a bone surface for which we want to determine the propagation speed, we start with # t + + # tdetermine the corrected travel times #t = # t + + # t- / 2. . Since the distances #r are also known by calibration, the measuring points can be arranged on a graph (time-distance), so that each pair of adjacent receivers is separated from #t along the time axis and #r along the distance axis. The velocity is then obtained by simple measurement of the slope of the regression line passing through all the points (see FIG. 8). In this version of the method, ie with prior calibration, the angle α which does not need to be known, is different for each pair of detectors considered because of the uncertainty on the exact positioning of each detector.

 A diagram (time frequency) - (spatial frequency) obtained from the data sampled in time and space by the present invention makes it possible, for example, to measure the speed of the transverse waves.

The shape of the diagrams (time frequency) - (spatial frequency) obtained depends on the central frequency of emission and the cortical thickness at the measurement site. We call FE the product (central excitation frequency) x (cortical thickness of the bone). The diagrams obtained for weak FE products (see FIGS. 9 and 10) clearly distinguish the dispersion curves of the propagation modes S0 and A1. The slope at the origin of the dispersion curve of the symmetrical mode So (propagating in a layer of cortical bone) has as value the speed in low frequency of the mode

<Desc / Clms Page number 25>

 where V1 and Vt are respectively the longitudinal and transverse velocity velocities of the ultrasounds in the bone. This formula and those that follow are valid only in the case where the study medium such as the cortical layer of bone is similar to a plaque. After having determined V1 by a velocity measurement of the first signal Vos = Vi, we can deduce a measure of the transverse velocity in the bone. In addition the value of the cutoff frequency of the mode A1 (time frequency of the mode when the spatial frequency tends to 0) is given by the expression fA1 = 1/2 # Vt / e, where e is the thickness of the layer cortical. From the measurement of the cutoff frequency fA1, we can therefore deduce the thickness e knowing Vt, where to deduce Vt knowing the thickness e.

0.7l8-[-) 1 onde de Rayleigh, TT Rayleigh = V . t 2 on peut déduire Vt 075-( J 1 connaissant V1 et réciproquement. De plus, lorsque la densité de mode est importante (produit FE grand), la vitesse V1 apparaît sur le diagramme (fréquence temporelle)-(fréquence spatiale) comme la pente d'une droite limitrophe de l'ensemble des courbes de dispersion (cf. figures 11 et 12). On a donc également accès à V1 à partir du diagramme (fréquence temporelle)-(fréquence spatiale). For larger FE products (see Figures 11 and 12), the mode density is too important to measure the parameters of the Ao and So modes in low spatial frequency, but we can then deduce the speed of the Rayleigh wave, obtained by measuring the slope of the dispersion curves of modes So and Ao which merge at high frequency with the dispersion curve of the Rayleigh wave. From the following expression of

0.7l8 - [-) 1 Rayleigh wave, TT Rayleigh = V. t 2 we can deduce Vt 075- (J 1 knowing V1 and vice versa Moreover, when the mode density is important (product FE large), the velocity V1 appears on the diagram (time frequency) - (spatial frequency) as the slope of a straight line bordering the set of dispersion curves (see Figures 11 and 12), so we also have access to V1 from the diagram (time frequency) - (spatial frequency).

 From the formula (4) of the angulation compensation method described above, it can generally be deduced that the corrected Vcor speed of the angulation of

<Desc / Clms Page number 26>

the probe is deduced from the speed measured in each direction of propagation by the formula:

Thus, all the speeds determined according to the geometrical characteristics of the frequency-frequency diagram can be corrected by producing for each direction of propagation a frequency-frequency diagram on the basis of non-corrected measurements, then by determining by means of the formula above the speed corrected according to the two speeds determined graphically on the two diagrams.

 The choice of the frequency used imposes the choice of the probe used. Thus, a longitudinal velocity measurement V1 can be made with a 2MHz probe on a bone of 5 mm cortical thickness, and then a low frequency So mode measurement can be made with a 200 kHz probe. The set of results of the two measurements gives the longitudinal and transverse velocities V1 and Vt. Low frequency probes have more spaced receivers than high frequency probes.

 The examples of measured parameters given above are illustrative, non-limiting. The invention relates to the recording of a spatio-temporal image by a multi-receiver probe arranged in a tight and immutable mesh, to sampling in space and time, and to obtaining a diagram (time frequency) - (spatial frequency).

Applications Application to the study of bone According to the WHO (World Health Organization), osteoporosis is a systemic disease of the skeleton (affecting the entire skeleton) characterized by low bone mass and deterioration of the bone. micro-architecture of the bone, resulting in a

<Desc / Clms Page number 27>

 increased bone fragility and risk of fracture. On the cortical bone, this results in an increase in the porosity of the bone, as well as a decrease in the cortical thickness. The propagation of ultrasonic waves in the cortical layer, sensitive to the local mechanical properties of the bone and its geometry, is modified by osteoporotic involvement.

 Just as the velocity of the longitudinal waves of volume is a discriminating parameter of the state of the bone, Vso, Vt, and the cortical thickness e are parameters modified by the osteoporotic attack, measurable thanks to the process and the device. of the present invention.

 The use of the invention on bone is not restricted to the detection of osteoporotic changes, but can be extended to the study of any bone disease affecting the mechanical properties of the bone and its geometry.

It can also be extended to pediatrics, for monitoring bone development in children.

Application to Other Materials The use of the present invention is not limited to the study of bone. It is also possible to study all the materials having a surface on which the probe and the coupling gel can be affixed, and which have at least one propagation speed greater than that of the intermediate medium. It is thus possible to apply the proposed measurement method to wood, metal, plastic polymer and glass materials.

For example, by measuring the velocity of the first signal, it is possible to measure the velocity of the longitudinal waves in reed pieces used in the manufacture of reeds for musical instruments in order to discriminate faulty reeds.

 Of course, the invention is not limited to the examples described and shown. In particular, in the probe example, there are two emission zones, but this is only necessary to compensate for the effects of the possible angulation of the probe with respect to the study medium. If we give up this

<Desc / Clms Page number 28>

 possibility of correction, the probe may have only one emission zone.

Claims (35)

A method of characterizing a study medium by applying ultrasound excitations on an accessible surface, in particular for characterizing the cortical layer of a bone separated from a surface accessible by an intermediate medium such as soft tissues, in which: - by at least one transmitter (3) placed on the accessible surface, excitation signals are sent to the study medium, - by receivers (4) pre-positioned with respect to each other in a probe (21) which is placed on the accessible surface, signals induced along the study medium are received by the excitation signals, characterized in that the successive excitation states are detected over time. , multiple receivers distributed over the path of the signals, and is established with the result of these detections a spatio-temporal image of vibratory signals transmitted along the study medium.  2. Method according to claim 1, characterized in that in the spatio-temporal image is noted at least one induced propagation mode along the study medium by the excitation signal.  3. Method according to claim 1 or 2, characterized in that the image is recorded by a spatio-temporal sampling comprising - sending the signal in the form of shots, - for each shot, selectively activate among the receivers (4) a small number of receiver (s) whose excitation state is noted at predetermined time intervals significantly smaller than the signal periods, different receivers being activated during different shots, - bringing to the same point of scale times the moments corresponding to shots from the same issuer.  4. Method according to one of claims 1 to 3, characterized in that a correspondence between the spatial frequency values and the temporal frequency values is established. <Desc / Clms Page number 30>  5. Method according to claim 4, characterized in that the correspondence is established by applying to the spatial-temporal image a two-dimensional Fourier transformation.  6. Method according to one of claims 1 to 5, characterized in that visualizes a correspondence diagram between the spatial frequency values and the time-frequency values.  7. Method according to one of claims 4 to 6, characterized in that by choice of the central frequency of the transmitted signal, an approximate value of the parameter FE calculated as follows:

 Where: Fc denotes the central frequency of the emitted signal emp denotes the thickness of the propagation medium.  8. Method according to one of claims 4 to 7, characterized in that the central frequency (Fc) of the transmitted signal is selected to selectively excite the at least one mode analyzed in the diagram (spatial frequency) - (time frequency ).  9. Method according to one of claims 4 to 8, characterized in that by choice of the central frequency (Fc) of the emitted signal shows the dispersion curve of the So mode and in particular its slope at the origin.  10. Method according to claim 9, characterized in that after determining the velocity (V1) of propagation of the longitudinal wave the velocity (Vt) of propagation of the transverse wave is determined by the relation:

 where Vso is the slope of the dispersion curve of the So mode at the origin.  11. Method according to one of claims 4 to 10, characterized in that by choice of the central frequency (Fc) <Desc / Clms Page number 31>  of the emitted signal, the Al mode is excited to reveal the dispersion curve of the mode Ai and in particular the value FA1 of its temporal frequency at the origin of the spatial frequencies.  12. The method as claimed in claim 11, characterized in that after determining the propagation velocity (Vt) of the transverse wave, the thickness (e) of the propagation medium is determined according to the relationship.

13. Method according to one of claims 4 to 12, characterized in that by selection of the central frequency (Fc) of the transmitted signal, a straight line is shown adjacent to the set of dispersion curves and the speed is taken ( Vi) propagation of the longitudinal mode as equal to the slope of this line.  14. Method according to one of claims 4 to 12, characterized in that the propagation speed (V1) of the longitudinal mode is determined from the departure time difference (#t) of the first signal between at least two receivers (4).  15. Method according to one of claims 4 to 12, characterized in that by selection of the central frequency (Fc) of the emitted signal, the Rayleigh wave is shown, and the slope (Vr) of the Rayleigh wave one of the longitudinal velocity (V1) and transverse velocity (Vt) as a function of the other longitudinal and transverse velocities by the relation:

16. Method according to one of claims 4 to 15, characterized in that by using a probe comprising <Desc / Clms Page number 32>

 two transmission zones (1) on either side of the reception zone (2), a diagram (spatial frequency) - (time frequency) is established for each direction of propagation according to which a speed of corresponding uncorrected propagation (V +, V-), then at least one corrected propagation velocity Vcor is determined from the two uncorrected velocities (V +, V ~) corresponding to the formula 17. Ultrasonic probe for bone auscultation, for the implementation of a method according to one of claims 1 to 16, comprising at least one transmitter (3) of ultrasound in at least one zone of emission (1) and receivers (4) of ultrasound in a reception zone (2), distributed along an adapter blade, characterized in that there are at least about ten receivers (4) placed in the reception area (1).  18. The probe as claimed in claim 17, characterized by an ultrasonic absorbing barrier (11) between the reception zone (2) and the at least one emission zone (3).  19. Probe according to claim 18, characterized in that the barrier (11) is made of cork.  20. The probe of claim 18 or 19, characterized in that the barrier (11) extends to a face (12) for contact with the body of the patient.  21. Probe according to one of claims 17 to 20, characterized in that the barrier (11) extends rearwardly, projecting into a rear medium (6) forming a common support to the at least one transmitter (3) and to the receivers (4).  22. Probe according to one of claims 17 to 21, characterized in that there is on at least one side of the receiving zone (2) a plurality of emitters (3) aligned in the propagation direction (Ox).  23. Probe according to one of claims 17 to 22, characterized in that with emitters (3) and piezoelectric receivers (4) each zone is framed by two elements (8) connected to ground. <Desc / Clms Page number 33>  24. Probe according to one of claims 17 to 23, characterized in that each transmitter (3) and / or each receiver (4) is an elongate element transversely to the direction of propagation (Ox) studied.  25. Probe according to claim 24, characterized in that each transmitter element or receiver (3,4) has parallel to the direction of propagation (Ox) a width between about 100 m and about 1 cm.  26. The probe as claimed in claim 24, characterized in that each emitter or receiver element (3,4) has, parallel to the propagation direction (Ox), a width approximately equal to half the wavelength in the water. for the ultrasonic frequency used.  27. Probe according to one of claims 17 to 26, characterized in that the receivers (4) are in succession side by side.  28. Apparatus for characterizing a study medium, in particular for characterizing a bone in vivo and non-invasively, characterized in that it comprises: a probe (21) according to one of claims 17 to 27, emission control means (26,46,34,41) for controlling the emission of successive ultrasonic fires by the at least one transmitter (3), - selective activation means (38,42) for activating at least one of the receivers (4) for each shot, and disable others, respectively, receivers, - means for recording and recording at successive sampling times the excitation state of the at least one an activated receiver.  Apparatus according to claim 28, characterized in that there are at least two transmitters (3) on the same side of the receiving area (2), and the transmit control means (26,46, 38, 41) are adapted to selectively activate both transmitters so that their signals meet in phase spatial match along the interface.  30. Apparatus according to one of claims 28 or 29, characterized in that it comprises a sequencer (38) which pilot <Desc / Clms Page number 34>  in a coordinated manner, on the one hand the emission of the shots, and on the other hand the state of activation and respectively deactivation of the receivers (4) to be activated and respectively deactivated as a result of each transmission.  31. Apparatus according to one of claims 28 to 30, characterized in that the transmission control means are used to adjust the central frequency (Fc) ultrasonic excitation between about 100 kHz and about 5 MHz.  32. Apparatus according to one of claims 28 to 31, characterized in that it comprises data processing means which establish from the digital space-time image a correspondence between the spatial frequencies and the temporal frequencies of the signals. induced in the study environment by an excitation signal.  33. Apparatus according to claim 32, characterized in that the processing means operate by digital two-dimensional Fourier transform.  34. Apparatus according to claim 32 or 33, characterized in that the processing means are capable of performing at least one determination by mathematical processing according to at least one of claims 10, 12, 13, 14, 15, 16.  35. Apparatus according to one of claims 28 to 34, characterized in that it comprises means for displaying a correspondence diagram between the spatial frequencies and the temporal frequencies.