Method for estimating anisotropy of a region of interest and associated methods and devices

Abstract

The present invention relates to a method for estimating a parameter representative of an anisotropy of the structure of a region of interest. This method relies on ultrasound sensing with plane waves and calculation with a calculator to deduce such parameter. Anisotropy is expected to be a biomarker of the diseases inducing an anisotropy effect, such as cancer. This opens the way to many applications of diagnostic, prognostic or follow-up of patients. Anisotropy may be used for other applications, notably for real-time guiding of thermal therapy.

Description

[0001]METHOD FOR ESTIMATING ANISOTROPY OF A REGION OF INTEREST AND ASSOCIATED METHODS AND DEVICES FIELD OF THE INVENTION The present invention concerns a method for estimating a physical parameter relative to an area. The present invention also relates to a method for controlling a thermal therapy of an area. The present invention also concerns devices involved in said method, namely a calculator, a device for estimating and a therapy system. BACKGROUND OF THE INVENTION The guiding of thermal therapies, and more specifically the estimation of tissue temperature, is achieved in most of the case by magnetic resonance imaging (MRI). However, it comes with high costs and is unsuitable for some patients, such as those with pacemakers or those suffering from obesity. It would thus be desirable to be able to guide thermal therapies with another imaging modalities, notably ultrasound imaging. For this, it is known to study the temporal evolution of specific features of radiofrequency signals or B-mode images of the area. However, to date, measuring temperatures above 50°C in biological tissues with ultrasound-based methods is challenging. Methods that uses or relies on changes in the speed of sound are limited by the non- linear relationship between the speed of sound and the temperature. Separation of unheated and heated sample using quantitative ultrasound (QUS) estimates occurred only in the 15-25 MHz frequency range and differences above 50°C seems to be beyond the QUS sensitivity. Consequently, ultrasound-guided focused ultrasound treatments rely on the appearance of hyperechogenicity to estimate whether the targeted tissues have been necrotized. Hyperechogenicity appears with cavitation or boiling at temperatures greater than 90°C. Furthermore, in most cases, there is no need to reach such a high temperature to create the desired thermal ablation. Ideally, temperatures between 56°C and 70°C are sufficient to irreversibly damage biological tissue. In addition, this temperature range enables purely thermal effects to be achieved, allowing excellent control over the extent of the damaged zone. For example, treatments guided with MR-thermometry usually use temperatures between 55°C and 65°C to ablate tissues. Moreover, the movements of the subject, which necessarily occur because of breathing or heart beating, are problematic in so far as they prevent from being able to follow the treated area with time because the same points for the ultrasound probe does not always correspond to the same point of the tissue. This renders the use of known ultrasound imaging techniques incompatible with a real-time guiding of thermal therapy. SUMMARY OF THE INVENTION There is therefore a need for a method for estimating a physical parameter relative to an area by using ultrasound sensing, which is able to measure temperature over a range from 37°C to 90°C and not sensitive to the natural movement of a subject, notably for the aim of achieving real-time thermal therapy guiding. To this end, the specification also describes a method for estimating a physical parameter relative to a region of interest of an area, the method for estimating being carried out estimating by a device for estimating and comprising : - a step of collecting backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times in presence of at least two of ultrasound excitations of the area with plane waves having a different angle with relation to a reference axis, - a step of calculating, for each angle, a value representative of the backscattered energy based on the backscattered signals, to obtain calculated values, and - a step of estimating a parameter representative of the anisotropy of the structure of the region of interest based on the variation of the calculated value with the angle of the plane wave exciting the area. Said method for estimating is a specific imaging technique for determining an original parameter. The imaging technique is here a plane wave ultrasound imaging. This imaging technique is different from the shear wave elastography. Both techniques are advanced techniques used in ultrasound imaging, but they are based on different physical principles and serve distinct purposes. Plane wave imaging involves emitting ultrasound as flat wavefronts, rather than focused beams, allowing rapid acquisition of data over a large area. This method enhances temporal resolution and enables faster image reconstruction, which is particularly useful for dynamic imaging. In contrast, shear wave elastography measures the propagation of shear waves generated in tissues by an ultrasound pulse. It assesses tissue stiffness, which is especially useful for detecting anomalies such as fibrosis or tumors. Thus, while plane wave imaging optimizes temporal resolution and image quality, shear wave elastography provides mechanical information about tissues. The original parameter is a parameter representative of the anisotropy of the structure of the region of interest. Such a parameter is especially useful for quantifying the elongation of a structure in a specific direction, and potentially its evolution over time. The present method therefore provides an efficient and non-invasive method providing a parameter, which can then be used as a marker for biological phenomenon of the imaged subject. According to further aspects, which are advantageous but not compulsory, the method for estimating might incorporate one or several of the following features, taken in any technically admissible combination: - the excitations are obtained by generating several plane waves. - an angle formed by one of the plane waves with the reference axis is comprised between 10° and 20°. - the parameter representative of the anisotropy is a parameter representative of the angular dependence of the calculated values. - the step of estimating further comprises deriving a physical parameter representative of the temperature of the region of interest based on the parameter representative of the anisotropy. - the method further comprises: - calculating the backscattered energy variation of the region of interest during the time measurement interval at a given position to obtain a calculated backscattered energy variation by adding the elementary backscattered energy variations between two consecutive measuring times, the elementary backscattered energy variation depending from a first signal and a second signal, the first signal being a signal obtained from the backscattered signal at the given position of the region of interest at a first measuring time of the two consecutive measuring times, the second signal being a signal obtained from the backscattered signal at said position at a second measuring time of the two consecutive measuring times, and - estimating a physical parameter representative of the temperature of the region of interest based on the calculated backscattered energy variation, - deducing a thermal contribution by subtracting the estimated parameter representative of the temperature based on the calculated backscattered energy variation and the physical parameter representative of the temperature of the region of interest based on the parameter representative of the anisotropy. - the area is a tissue of a subject. - the subject is an animal, notably a rabbit, a pig, a rodent or a human The present description also concerns a method for controlling a thermal therapy of a region of interest of an area, the thermal therapy being carried out by a therapy apparatus according to a configuration of the therapy apparatus, the method comprising the step of: - carrying out a method for determining as previously described, to obtain a physical parameter representative of the region of interest of the area and - controlling the configuration of the therapy apparatus by setting the configuration of the therapy apparatus in function of the physical parameter representative of the region of interest. According to further aspects, which are advantageous but not compulsory, the method for controlling might incorporate one or several of the following features, taken in any technically admissible combination: - the configuration of the therapy apparatus is a set of values for parameters of the thermal therapy applied by the therapy apparatus, each parameter being chosen in the list consisting of the shape of the emitting surface of the therapy apparatus, the number of elements of the emitting surface of the therapy apparatus, the duration of the thermal therapy, emitted power by the therapy apparatus, duty cycle of the therapy apparatus, frequency of the therapy apparatus, beam steering of the therapy apparatus and therapy apparatus location. - the therapy apparatus comprises a heating unit, the heating unit being an ultrasound probe, a radiofrequency ablation probe, a laser or a microwave ablation probe. The present description also deals with a method comprising carrying out a method for estimating as previously described, to obtain a physical parameter relative to a region of interest of an area, the method being chosen among: - a method for nondestructive testing, - a method for predicting that a subject is at risk from suffering from an anisotropy related disease, - a method for diagnosing an anisotropy related disease, - a method for identifying a therapeutic target for preventing and / or treating an anisotropy related disease, - a method for identifying a marker, the biomarker being a diagnostic marker of an anisotropy related disease, a susceptibility biomarker of an anisotropy related disease, a prognostic biomarker of an anisotropy related disease or a predictive biomarker in response to a treatment of an anisotropy related disease, and - a method for screening a compound useful as a probiotic, a prebiotic or a medicine, the compound having an effect on a known therapeutical target, for preventing and / or treating an anisotropy related disease. The present description also concerns a calculator adapted to estimate a physical parameter relative to a region of interest of an area, the calculator being adapted to : - obtain backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times in presence of at least two of ultrasound excitations of the area with plane waves having a different angle with relation to a reference axis, - calculate, for each angle, a value representative of the backscattered energy based on the backscattered signals, to obtain calculated values, - estimate a parameter representative of the anisotropy of the structure of the region of interest based on the variation of the calculated value with the angle of the plane wave exciting the area. The present description also deals with a device for estimating a physical parameter relative to a region of interest of an area, the device for estimating comprising : - an ultrasound probe adapted to collect backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times in presence of at least two of ultrasound excitations of the area) with plane waves having a different angle with relation to a reference axis, and - a calculator as previously described The present description also concerns a therapy system comprising: - a therapy apparatus, the therapy apparatus being adapted to carry out a thermal therapy of a region of interest of an area according to a configuration of the therapy apparatus, and - a device for estimating as previously described, the device for estimating being further adapted to control the configuration of the therapy apparatus by setting the configuration of the therapy apparatus in function of the physical parameter representative of the region of interest. The specification also describes a method for estimating a physical parameter relative to a region of interest of an area, the method for estimating being carried out by a device for estimating and comprising : - a step of collecting backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times, - a step of calculating the backscattered energy variation of the region of interest during the time measurement interval at a given position to obtain a calculated backscattered energy variation by adding the elementary backscattered energy variations between two consecutive measuring times, the elementary backscattered energy variation depending from a first signal and a second signal, the first signal being a signal obtained from the backscattered signal at the given position of the region of interest at a first measuring time of the two consecutive measuring times, and the second signal being a signal obtained from the backscattered signal at said position at a second measuring time of the two consecutive measuring times, and - a step of estimating a physical parameter representative of the temperature of the region of interest based on the calculated backscattered energy variation . According to further aspects, which are advantageous but not compulsory, the method for estimating might incorporate one or several of the following features, taken in any technically admissible combination: - the elementary backscattered energy variation is defined as: where: •^^ is an integer strictly superior or equal to 1,• Δ^^^^^^^^ designates the elementary backscattered energy variation between thefirst measuring ‒ 1and the second measuring time ^^^^, • designates the first signal,• ^^2 designates the second signal,• |^^| designates the absolute value of A, and• 〈^^〉 designates the spatial average of the value A.- the first signal is the backscattered signal at the given position at a first measuring time of the two consecutive measuring times and the second signal is the backscattered signal at the given position at a second measuring time of the two consecutive measuring times. - the method further comprises: - a step of obtaining the attenuation coefficient of the medium situated between an ultrasound probe used at the step of collecting and the area, - a step of post-processing each backscattered signal, to obtain a post-processed signal, the step of post-processing comprising compensating the attenuation of the medium in the backscattered signal by using the obtained attenuation coefficient, the first signal being the post-processed backscattered signal at the given position at a first measuring time of the two consecutive measuring times, the second signal being the post-processed signal obtained from the backscattered signal at the given position at a second measuring time of the two consecutive measuring times. - the step of obtaining comprises calculating the attenuation coefficient by determining the argument of the intensity of the backscattered signals at one measuring time. - the step of post-processing further comprises applying a Hanning window on the signal obtained after carrying out the compensating of the attenuation of the medium in the backscattered signal by using the determined attenuation coefficient, to obtain the post-processed signal. - the physical parameter representative of the temperature of the region of interest is thermal dose in cumulative equivalent minute at a predetermined temperature. - the physical parameter representative of the temperature of the region of interest is a state of the region of interest. - the state of the region of interest is chosen among three states, the first state corresponding to a treated region of interest, the first state being representative to a temperature strictly inferior to a first threshold, the second state corresponding to non-treated region of interest or a potential necrotic region of interest, the second state being representative of a temperature comprised between the first threshold and a second threshold, and a third state corresponding to a necrotic region of interest, the third state being representative of a temperature strictly superior to a second threshold. - the area is a tissue of a subject, the subject being an animal, notably a rabbit, a pig, a rodent or a human. The present description also deals with a method for controlling a thermal therapy of a region of interest of an area, the thermal therapy being carried out by a therapy apparatus according to a configuration of the therapy apparatus, the method comprising the step of: - carrying out a method for estimating as previously described, to obtain a physical parameter representative of the temperature of the region of interest, and - controlling the configuration of the therapy apparatus by setting the configuration of the therapy apparatus in function of the physical parameter representative of the temperature of the region of interest. The present description also concerns a method comprising carrying out a method for estimating as previously described, to obtain a physical parameter representative of the temperature of a region of interest of an area, the method being chosen among: - a method for nondestructive testing, - a method for predicting that a subject is at risk from suffering from a temperature related disease, - a method for diagnosing a temperature related disease, - a method for identifying a therapeutic target for preventing and / or treating a temperature related disease, - a method for identifying a marker, the biomarker being a diagnostic marker of a temperature related disease, a susceptibility biomarker of a temperature related disease, a prognostic biomarker of a temperature related disease or a predictive biomarker in response to a treatment of a temperature related disease, and - a method for screening a compound useful as a probiotic, a prebiotic or a medicine, the compound having an effect on a known therapeutical target, for preventing and / or treating a temperature related disease. The present description also deals with a calculator adapted to estimate a physical parameter relative to a region of interest of an area, the calculator being adapted to : - obtain backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times, - calculate the backscattered energy variation of the region of interest during the time measurement interval at a given position to obtain a calculated backscattered energy variation by adding the elementary backscattered energy variations between two consecutive measuring times, the elementary backscattered energy variation depending from a first signal and a second signal, the first signal being a signal obtained from the backscattered signal at the given position at a first measuring time of the two consecutive measuring times and the second signal being a signal obtained from the backscattered signal at said position at a second measuring time of the two consecutive measuring times, and - estimate a physical parameter representative of the temperature of the region of interest of the area based on the calculated backscattered energy variation. The present description also concerns a device for estimating a physical parameter relative to an area, the device for estimating comprising : - an ultrasound probe adapted to collect backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times, and - a calculator as previously described. The present description also deals with a therapy system comprising: - a therapy apparatus, the therapy apparatus being adapted to carry out a thermal therapy of an area according to a configuration of the therapy apparatus, and - a device for estimating as previously described, the device for estimating being further adapted to control the configuration of the therapy apparatus by setting the configuration of the therapy apparatus in function of the physical parameter representative of the temperature of the region of interest. The term ^adapted to^ used here should be understood as meaning ^able to^ or ^configured to^. BRIEF DESCRIPTION OF THE DRAWINGS The invention will be better understood on the basis of the following description which is given in correspondence with the annexed figures and as an illustrative example, without restricting the object of the invention. In the annexed figures: - figure 1 is a schematic representation of an area to be monitored and a device adapted to estimate a physical parameter relative to said area - figure 2 is a flowchart of an example of carrying out a method for estimating by the device for estimating of figure 1, the estimated physical parameter being a physical parameter representative of the temperature of the area, - figure 3 is a flowchart of another example of carrying out a method for estimating by the device for estimating of figure 1, the estimated physical parameter being a physical parameter representative of the anisotropy of the area, - figure 4 is a representation of a therapy system comprising the device for estimating of figure 1, as an illustration of an application of a possible use of the device for estimating of figure 1, and - figures 5 to 15 are figures showing results obtained with experiments carried out by the Applicant when using the device for estimating of figure 1. More precisely: - figure 5 is a graph showing normalized backscattered intensity as a function of depth for an incident plane wave propagating inside the sample. Exponential fit (dotted line) with slope ^^0 = 0.06 ^^^^ / (^^^^.^^^^^^) , - figure 6 is a set of three graphs showing Δ^^^^^^ and temperature evolution over time during treatment for free-field acoustic power values of 9 W (left), 12 W (center) and 16 W (right), inside (black diamonds) and outside (white circles) the focal zone. - figure 7 represents on the left BSE variation as a function of temperature increaseΔ^^ . The Pearson coefficient R is equal to 0.94. (Right) Evaluation of Δ^^^^^^thermometry accuracy and bias. The Δ^^^^^^ thermometry evaluation was based on a comparison with thermocouple measurements. For the accuracy, the dashed red line represents the acceptable mean accuracy threshold (5°C), and the red circle represents the mean Δ^^^^^^ thermometry accuracy. For the bias, the dashed red line represents the acceptable mean accuracy threshold (± 0.5°C), and the red circle represents the mean bias value. In both cases, the interquartile range denotes the middle 50% of the dataset. The top box shows the 75% of the dataset that falls below the upper quartile, while the bottom line indicates the 25% of the dataset that falls below the lower quartile. The middle line represents the median value, and the lines extending from the box represent the 2.5% and 97.5% limits of the dataset. - Figure 8: Five moments during HIFU treatment with different signal processing methods. Acquisitions were performed with the imaging probe placed in the center of the HIFU transducer. (A-E) Bmode imaging, (F-J) temperature maps based on theΔ^^^^^^estimation, (K-O) thermal dose maps and (P^T) treatment maps based on equations 9a and 9b. The dotted and solid lines were determined based on visual inspections of the treated samples (see Figure 9). The dotted lines delineate the unambiguously coagulated region. The solid lines delineate the region between the untreated tissues and unambiguously coagulated tissues. - Figure 9: Top row: Three examples of treated liver tissues sliced in a plane parallel to the HIFU acoustic axis, showing the visually estimated unambiguously coagulated region (inside the dotted lines) and the region between the untreated tissues and coagulated tissues (between the dotted lines and the solid lines). Bottom row: treatment maps obtained by quantifying BSE variations during the treatment. The free-field acoustic power was set 16W for these three samples. - figure 10 is a graph illustrating final Δ^^^^^^ value as a function of the maximal temperature increase during HIFU treatment. - figure 11 illustrates in parts A to E histological sections of different samples that were untreated (part A), treated with HIFU (parts B to D) or heated in a water bath (part E). The second row shows the same picture in black and white to better highlight the structural change (parts F to J) Corresponding maps of ^^^^(^^) (parts K to O) and ^^^^(^^) (parts P to T) are also represented respectively on third row and fourth row. - figure 12 shows histological slides of liver tissue of heifer not treated (part A) or treated by HIFU at 55°C (part B), 65°C (part C) or 75°C (part D). The arrows on parts C and D show the HIFU axis. - figure 13 illustrates Bmode image of liver tissue heated by HIFU. The treated area appears hyper-echoic on the two images obtained by placing the ultrasound probe parallel (part A) or perpendicular (part B) to the HIFU axis. The echogenicity is more marked in case (part A), which shows the impact of anisotropy on the ultrasound signal. - figure 14 is a map obtained by measuring the difference in energy contained in the RF signals measured by the ultrasound probe when the environment is either insonified by 2 waves planes angulated at +-18° or by a plane wave at normal incidence (0°). The dotted line demarcates the near field zone beyond which the signals can be physically exploited. - figure 15 is a macroscopic observation of the lesion formed inside the liver tissue by the HIFU beam. Two tiles of the support correspond to 1cm. In both images, the scale is almost identical, the areas circled in dotted lines are the same size. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A device for estimating 10 a physical parameter relative to a region of interest 11 of an area 12 is shown on figure 1. The physical parameter may vary according to the operating of the device for estimating 10, as will appear later in the specification. It can be noticed here that the corresponding methods can be carried out in-vivo or ex-vivo. In particular, in reference to figure 2, it will be illustrated a case where temperature is the estimated physical parameter, whereas, in reference to figure 3, it will be illustrated a case where anisotropy is the estimated physical parameter. Preferably, as illustrated in figure 1, the area 12 is a tissue of a subject; the region of interest 11 being thus a tissue of a subject. According to the example, the tissue is the tissue of a liver. However, the device can be used to any other biological tissue and mainly breast, pancreas, muscle, brain and kidney. The subject is, for instance, an animal, notably a rodent, a rabbit, a pig or a human. In the present case, the subject is a heifer. The device for estimating 10 comprises an ultrasound probe 14 and a calculator 16. The ultrasound probe 14 is adapted to send ultrasound waves to the area 12 and to collect the backscattered signal emitted by the area 12 in response to the ultrasound waves. Said ultrasound waves may come from another apparatus, as is the case in the example of a therapy apparatus in figure 4. More precisely, the ultrasound probe 14 is a set of transducers, each transducer being adapted to collect the radiofrequency backscattered signals from the ultrasound excitation. The number of transducers and their spatial arrangement may vary according to the application. As a specific example, the number of transducers is here set to 128 and the transducers are assumed to be arranged so as to form a rectangular array. The ultrasound probe 14 is thus adapted to provide radiofrequency backscattered signals, which may be processed to obtain an ultrasound image of the area 12 if desired. The ultrasound probe 14 is adapted to apply ultrasound waves in a controlled manner to fulfill the requirement of health safety of the area 12. The ultrasound probe 14 is adapted to obtain such backscattered signals at several measuring times during a time measurement interval. The time measurement interval extends from an initial time t0to a final time tN-1. A time measurement is thus indexed with an integer strictly superior or equal to 1 and N designates the number of time measuring. The time measurement interval comprises N elementary time measurement intervals each extending between two consecutive measuring times. Each elementary time measurement intervals extends between a first measuring time^^^^ ‒ 1and a second measuring time^^^^. The calculator 16 is adapted to collect the backscattered signals acquired by the ultrasound probe 14 and to process them to obtain an estimation of the physical parameter relative to an area 12. The calculator 16 is an electronic circuit designed to manipulate and / or transform data represented by electronic or physical quantities in registers of the calculator 16 and / or memories into other similar data corresponding to physical data in the register or memory memories. other types of display devices, transmission devices or storage devices. As specific examples, the calculator 16 is produced in the form of a programmable logic component, such as an FPGA (Field Programmable Gate Array), or even an integrated circuit, such as an ASIC (Specific Integrated Circuit). Alternatively, when the method is carried out in the form of one or more software programs, that is to say in the form of a computer program, also called a computer program product, it is also capable of being recorded on a medium, not shown, readable by computer. The computer-readable medium is, for example, a medium capable of storing electronic instructions and of being coupled to a bus of a computer system. For example, the readable medium is an optical disk, a magneto-optical disk, a ROM memory, a RAM memory, any type of non-volatile memory (for example FLASH or NVRAM) or a magnetic card. A computer program comprising software instructions is then stored on the readable medium. An example of the operating of the device for estimating 10 is now described in reference to figure 2, which is a flowchart showing an example of carrying out a method for estimating. In this case, the method for estimating aims at obtaining an estimation of a physical parameter representative of the temperature of the area. According to this example, the method for estimating comprises a step of collecting S20, a step of obtaining S30, a step of post-processing S40, a step of calculating S50, a step of estimating S60. During the step of collecting S20, the ultrasound probe 14 collects the backscattered signals at several measuring times during the time measurement interval. During the step of obtaining S30, the calculator 16 obtains the attenuation coefficient of the medium situated between the region of interest 11 and the ultrasound probe 14. For this, the calculator 16 calculates the backscattered intensity of the ultrasound probe 14 by summing the contribution from each transducers of the The backscattered intensity I(z) was obtained based on all the M transducers of the ultrasound probe 14 as follows: where^^^^(^^)is the RF signal acquired by the j-th element. The calculator 16 then extracts the argument of the signal according to the followingequation: where: •^^(^^) is the attenuation of the middle, and• ^^ is the depth obtained by time-of-flight analysis.The calculator 16 then deduces the attenuation ^^(^^) by using a logarithmic regression and the previous relation. During the step of post-processing S40, the calculator 16 applies different operations on the backscattered signals to obtain post-processed signals. In the present case, two operations are carried out, one first operation being a compensation of the attenuation and a second operation being applying a Hanning window the signal obtained after carrying out the compensating of the attenuation of the medium. During the step of calculating S50, the calculator 16 calculates the backscattered energy variation during the time measurement interval at a given position to obtain a calculated backscattered energy variation. For this, the calculator 16 adds the elementary backscattered energy variations between two consecutive measuring times for each pair of consecutive measuring times included in the time measurement interval. The elementary backscattered energy variation depends from a first signal S1 and a second signal S2. The first signal S1 is a signal obtained from the backscattered signal at the givenposition at a first measuring time ^^^^ ‒ 1 of the two consecutive measuring times ^^^^ ‒ 1 and ^^^^ , ^^being there an integer strictly superior or equal to 1. Similarly, the second signal S2 is a signal obtained from the backscattered signal at the given position at a second measuring time^^^^of the two consecutive measuring times^^^^ ‒ 1and ^^^^. In the present case, the elementary backscattered energy variation is defined as: where: •Δ^^^^^^^^ designates the elementary backscattered energy variation between thefirst measuring time^^^^and the second measuring time ^^^^ ‒ 1, •|^^| designates the absolute value of A, and• 〈^^〉 designates the spatial average of the value A.According to a first embodiment, the first signal S1 is the backscattered signal at the given position at a first measuring time^^^^ ‒ 1of the two consecutive measuring times^^^^ ‒ 1and ^^^^and the second signal S2 is the backscattered signal at the given position at a second measuring time^^^^of the two consecutive measuring times ^^^^ ‒ 1and ^^^^,. According to a second embodiment, the first signal S1 is the post-processed backscattered signal at the given position at a first measuring time ^^^^ ‒ 1of the two consecutive measuring times ^^^^ ‒ 1and ^^^^and the second signal S2 is the post-processed backscattered signal at the given position at a second measuring time^^^^of the two consecutive measuring times^^^^ ‒ 1and ^^^^. This leads to the following formula: where •^^^^^^^^^^ is the RF signal obtained after post-processing, and• ^^ designates the given position.This expression can also equivalently be expressed in a logarithm form, leading to the following expression More developments on such expression can be found hereinafter in the specification. During the step of estimating S60, the calculator 16 estimates a physical parameter representative of the temperature of the region of interest 11 based on the calculated backscattered energy variation. According to a first example, the physical parameter representative of the temperature of the region of interest 11 is the thermal dose in cumulative equivalent minute at a predetermined temperature. Here, the predetermined temperature is set to 43°C. For this, the calculator 16 implements the following calculation: where: •^^43(^^) is the thermal dose in cumulative equivalent minutes (CEM),• T is the current tissue temperature,• ^^1 is the time at which the temperature has started to evaluate, and• ^^^^ is the time at which the treatment ended.According to a second example, the physical parameter representative of the temperature of the area 12 is a state of the area 12. For instance, the state of the area 12 is chosen among three states ST1, ST2 and ST3. The first state ST1 corresponds to a treated area, the first state ST1 being representative to a temperature strictly inferior to a first threshold. The second state ST2 corresponding to non-treated area or a potential necrotic area, the second state ST2 being representative of a temperature comprised between the first threshold and a second threshold. The third state ST3 corresponds to a necrotic area, the third state ST3 being representative of a temperature strictly superior to a second threshold. Any technique enabling to determine the thresholds can be considered here. Notably, calibration temperature curves obtained with other experiments may be advantageously used here. As apparent from the experiments carried out in the experimental section, such method enables to estimate the temperature of the region of interest 11 by using ultrasound sensing. The method is accurate and can be carried out in real time, so that the method enables to monitor temperature of a region of interest 11. An example of the operating of the device for estimating 10 is now described in reference to figure 3, which is a flowchart showing an example of carrying out a method for estimating. In this case, the method for estimating aims at obtaining an estimation of a physical parameter representative of the anisotropy of the structure of the region of interest 11. A physical parameter representative of the anisotropy is a parameter quantifying the deformation of a structure, like cells or collagenic or connective tissues, in one specific direction compared to its initial state. According to this example, the method for estimating comprises a step of collecting S120, a step of calculating S130, a step of estimating S140. During the step of collecting S120, the ultrasound probe 14 collects the backscattered signals at several measuring times during a time measurement interval in presence of at least two of ultrasound excitations of the area 12 with plane waves having a different angle with relation to a reference axis. In the present case, in addition, the ultrasound probe 14 is arranged to generate each plane wave forming an angle inferior to 90°, preferably comprised between 10° and 20°, with a reference axis. The reference axis is, for instance, an axis perpendicular to the plane defined by the transducers. In addition, the plane wave has a frequency comprised between 1 MHz and 80 MHz. Alternatively, instead of an electrical command enabling the ultrasound probe 14 to generate the excitation, the position of the ultrasound probe 14 is changed manually, mechanically or using a robotized arm to generate each plane wave. During the step of calculating S130, the calculator 16 calculates, for each angle, a value representative of the backscattered energy based on the backscattered signals. Such value is for instance a spatial map giving the value of the backscattered energy for a plane wave sent with the angle. Alternatively, it could be considered a statistical value of the backscattered energy over all the region of interest 11, for instance a mean value. As another example, the value is the backscattered energy at said angle. Therefore, the calculator 16 obtains a set of values representative of the backscattered energy for each angle of the wave used to excite the area 12. During the step of estimating S140, the calculator 16 estimates a parameter representative of the anisotropy based on the variation of the calculated value with the angle of the plane wave exciting the area 12. In the present case, the parameter representative of the anisotropy is a parameter representative of the angular dependence of the backscattered energy. For instance, the parameter is the ratio between the maximum backscattered energy at an angle and the minimum backscattered energy at an angle. The method for estimating does not need a temporal follow up of the same area. Indeed, the anisotropy can be acquired by using multiplane wave imaging. In addition, the anisotropy is a biomarker of the local temperature (as notably illustrated experimentally in the experimental section). Compared to other known biomarkers, anisotropy can be validly used for temperature over 55°C. This confers to anisotropy the property of being a biomarker of area undergoing irreversible effect linked to temperature. Therefore, advantageously, the method for estimating further comprises a step of deriving a physical parameter representative of the temperature of the region of interest based on the parameter representative of the anisotropy. For both methods for estimating (corresponding to figures 2 and 3), it provides the user with the ability of accurately estimating the searched parameter in a non-invasive way with a set-up, easily available clinically. Such property is advantageous for many applications, some of which being now described. One application is the guiding of thermal therapies. To illustrate such application, a therapy apparatus 24 is represented on figure 4. The therapy apparatus 24 is adapted to carry out a thermal therapy of the region of interest 11 according to a configuration of the therapy apparatus 24. By definition a configuration of a therapy apparatus 24 is a set of values for parameters of the thermal therapy applied by the therapy apparatus 24, each parameter being chosen in the list consisting of shape of the emitting surface of the therapy apparatus 24, number of elements of the emitting surface of the therapy apparatus 24, duration of the thermal therapy, emitted power by the therapy apparatus, duty cycle of the therapy apparatus 24, frequency of the therapy apparatus 24, beam steering of the therapy apparatus 24 and therapy apparatus 24 location. In an embodiment, the configuration provides with a value for each parameter of the list. The therapy apparatus 24 is part of a therapy system, which also encompasses the device for estimating 10 of figure 1. As apparent from figure 4, the therapy apparatus 24 comprises a heating unit 18 adapted to heat a zone 20, which is monitored by a thermocouple 22. As an example, the heating unit 18 is another ultrasound probe adapted to apply ultrasound waves, which generates the temperature increase. This other ultrasound probe may also be adapted to apply a therapy to the user. In this example, the therapy system therefore comprises two ultrasound probes, one for imaging and another one for heating. In a specific embodiment, the therapy system only comprises one ultrasound probe, the imaging ultrasound probe 14 being also used to heat the area 12. In another embodiment, the heating unit 18 is a radiofrequency ablation probe, a laser or a microwave ablation probe. The arrows 26 and 27 show schematically the different ultrasound waves applied, the arrows 26 corresponding to imaging ultrasound waves and the arrows 27 to heating ultrasound waves. The device for estimating 10 is further adapted to control the configuration of the therapy apparatus by setting the configuration of the therapy apparatus in function of the physical parameter representative of the temperature of the area. According to other applications, the method may be used in other methods, among which: - a method for nondestructive testing, in such case, for instance, the object is a spatial arrangement of crystals, - a method for predicting that a subject is at risk from suffering from a disease, , - a method for diagnosing a disease, - a method for identifying a therapeutic target for preventing and / or treating a disease, - a method for identifying a marker, the biomarker being a diagnostic marker of a disease, a susceptibility biomarker of a disease, a prognostic biomarker of a disease or a predictive biomarker in response to a treatment of a disease, and - a method for screening a compound useful as a probiotic, a prebiotic or a medicine, the compound having an effect on a known therapeutical target, for preventing and / or treating a disease. In such applications, the results of the estimation is compared with a reference to deduce the property. For instance, a disease can be diagnosed by comparing the estimated value for the subject having the disease and the estimated value for the subject not suffering from the disease. Alternatively, the reference value may be issued from a table. Generally, the disease is an anisotropy related disease or a temperature related disease. The term ^related^ meaning here that the disease induces an anisotropy effect or a temperature effect, which will be detected with the use of the method for estimating. A specific example of anisotropy related disease is breast tumor. Such kind of tumor is expected to present an evolution of the anisotropy of the tissue. Indeed, the growth of a cancerous mass involves the extracellular matrix by stiffening it and modifying the geometry of the fiber bundles. The term of Tumor Associated Collagen Signatures is used to designate that these groups of collagen bundles are characteristic of cancerous tumor development. This is notably shown in an article by Kaushik et al. entitled ^From transformation to metastasis: deconstructing the extracellular matrix in breast cancer^. The extracellular matrix therefore has specific tissue fiber orientations that are different from healthy tissues, in or around the tumor. In addition, the orientation of the collagen fibers around the tumor is often linked to the more or less aggressive nature of the tumor, and to its way of progressing into the tissues, either by "spicules" surrounding the tumor radially, or by an enveloping ^capsule^. To illustrate this observation, one can notably refer to an article of M. Costantini et al. entitled ^Association between sonographic appearances of breast cancers and their histopathologic features and biomarkers^ or to an article of L. Gole entitled ^Quantitative stain-free imaging and digital profiling of collagen structure reveal diverse survival of triple negative breast cancer patients^. Therefore, an estimating method for determining an anisotropy effect is expected to be an efficient method to detect a breast tumor. It should also be stressed that both method may be carried out on the same region of interest to be able to separate the thermal effect from the mechanical effect of a heating. The thermal effect would be obtained by subtracting the temperature obtained by carrying out the method of figure 2 to the temperature obtained by carrying out the method of figure 3. EXPERIMENTAL SECTION The use of high intensity focused ultrasound (HIFU) therapy for tissue ablation has recently gained momentum. Guidance is provided by either magnetic resonance imaging or ultrasound imaging. To date, ultrasound imaging is limited by its inability to provide temperature measurements at thermal therapies temperature ranges (between 55°C and 70°C). Here, variations in ultrasonic backscattered energy (∆BSE) were used to monitor temperature increases in liver tissue up to 100°C during HIFU treatment. Results showed a linear correlation between ∆BSE and temperature (r = 0.94, p < 0.001). Monitoring can be performed at ultrasound scanners frame rate with an accuracy of 5°C. For moderate temperature rises, ∆BSE can be used to monitor cooling. Above 70°C, irreversible changes occur that were investigated through histology, highlighting the role of HIFU beam directivity in creating microstructural anisotropy. The results demonstrate the ability of ∆BSE to estimate temperature within therapeutic range, while providing safety control. This method can be readily implemented clinically and potentially applied to other thermal therapies. 1 Introduction In recent years, many non-invasive high intensity focused ultrasound (HIFU) treatments have been developed. In most cases, guidance is provided by either magnetic resonance imaging (MRI) or conventional BMode ultrasound imaging (sonography). Some applications have been approved for commercial use and are available in medical treatment centers worldwide. Other potential uses of the technology are in the early stages of technical research. These treatments require efficient real-time monitoring of internal body temperature to ensure adequate temperature increases in the targeted area while preventing lesions in healthy tissues. To date, MRI is the main technology used to achieve this real-time monitoring. Typically, the spatial resolution of MRI is on the order of 1 mm, the tissue temperature can be determined within 1◦C, and the temporal resolution is on the order of 1 s. Therefore, MRI offers excellent three-dimensional imaging and temperature monitoring capabilities; however, MRI has high costs and is unsuitable for some patients, such as those with pacemakers or those suffering from obesity. Moreover, as HIFU treatment time often exceeds one hour, the accessibility of MRI scanners is a limitation in many hospitals. Many alternative strategies have been developed to address these limitations, such as computed tomography (CT), microwave radiometry and photoacoustic imaging. Among them, ultrasound imaging is a low-cost approach that enables high temporal (on the order of 1 ms) and spatial resolution (on the order of 100 µm) without ionizing radiation. Moreover, ultrasound imaging is easily accessible clinically. To date, ultrasound cannot be used to determine tissue temperature. Consequently, ultrasound-guided focused ultrasound treatments rely on the appearance of hyperechogenicity to estimate whether the targeted tissues have been necrotized. Hyperechogenicity appears with cavitation or boiling at temperatures greater than 90°C. Furthermore, in most cases, there is no need to reach such a high temperature, as these temperatures may lead to uncontrolled and unwanted damage. Ideally, temperatures between 56°C and 70°C are sufficient to irreversibly damage biological tissue. In addition, this temperature range enables purely thermal effects to be achieved, allowing excellent control over the extent of the damaged zone. For example treatments guided with MR- thermometry usually used temperatures between 55°C and 65°C to ablate tissues. Thus, many studies have focused on the development of other imaging methods to monitor hyperthermic or ablative treatments over time with ultrasound imaging. Moreover, new technologies have been developed by major ultrasound manufacturers to allow large scale access to radiofrequency data for research purposes, leading to technical innovations with state-of-the-art ultrasound scanners. However, regardless of the method used (speed of sound, attenuation, elasticity, concentration of diffusers, etc.), no method has been able to measure reliably tissue temperature during treatment over the ranges corresponding to the target temperatures during ablative thermal therapies (between 55°C and 70°C). In this experimental section, radiofrequency (RF) signals and ultrasonic backscattered energy variations (∆BSE) were used to monitor temperature increases in liver tissues up to 100°C during and after HIFU treatment. The temperature measurements were correlated with actual measurements obtained using thermocouples, and the extent of the treated region was correlated with the region estimated based on the ∆BSE data. Moreover, the increase in the backscattered energy was investigated at the cellular level by considering the cellular network as a continuous scattering phase immersed in a fluid medium. 2 Methods HIFU device The toroidal HIFU device used is known to be suitable for creating large ablation. This device was placed into acoustic contact with the sample using degassed water contained in a polyurethane coating (CIV-Flex Transducer cover, CIVCO, Kalona, IA, USA). This coating attenuated the ultrasound pressure by approximately 2% at 2.5 MHz. To prevent the transducer from becoming too hot during the HIFU procedure, the water was cooled at 6 ◦C and flowed at a continuous rate of 0.6 L.min-1 using a peristatic pump (Masterflex LS Model 7555-05, Cole Parmer Instruments Co., Chicago, IL, USA) in a closed cooling circuit. The amplifier (Image Guided Therapy, Pessac, France) has 32 channels with programmable phase and power. The phase of each channel could be adjusted with a resolution of 1 degree. The electrical power delivered by each channel could be adjusted up to 20 Watts. The time needed to change the phase and / or power delivered was 1 ms. The spatial repartition and intensity of the produced pressure field can be controlled electronically by modulating the amplitude and the phase applied to each of the 32 individual transducers. The phase, emitted power and reflected power of each channel were analyzed and registered during the HIFU sonications with a custom-made user interface. Ultrasound imaging probes A sectorial ultrasound imaging probe working at a frequency of 7.5 MHz (Vermon, Tours, France) was placed at the center of the HIFU transducer and connected to an ultrasound scanner (EB4012, B-K Medical, Herlev, Denmark). The ultrasound imaging plane was aligned with the HIFU acoustic axis. The scanner was modified in order to allow acquisition of radiofrequency (RF) lines by a computer via an analog / digital converter (CompuScope^ CS14100; GaGe, Lockport, IL) at a sampling frequency of 40 MHz. This probe was used to guide the treatment and to acquire RF signals during the off cycles of the HIFU sequence to compute of ∆BSE. Additional ultrasound imaging was performed using a Verasonic L7-4 imaging probe connected to a Verasonics ultrasound scanner (Vantage 256, Verasonics, Kirkland, WA, USA) for attenuation measurements. This second ultrasound imaging probe was placed perpendicularly to the HIFU acoustic axis, at the depth of the focal zone (FZ). The Verasonics system provided the raw radio frequency (RF) signals at a sampling frequency of 20 MHz, which is 4 times larger than the center frequency of the L7-4 probe. The imaging sequence is composed of a single plane wave. Experimental setup Bovine livers were collected from a local slaughterhouse and were sliced into rectangular-shaped samples of approximately 10 cm x 10 cm x 6 cm. The samples were then placed in a crystallizing dish filled with degassed water (dissolved [O2] = 2^3 mg / L). The samples in the crystallizing dish were then degassed with a vacuum pump (0.7 bar for 30 min) prior to the experiment, to remove any bubbles that might have formed during the process. To minimize contact with air, the sample and degassed water were both transferred to an experimental tank containing degassed water. While underwater, the sample was placed in a holder. The water in the tank was then heated to 37°C using a thermostat, and the sample was allowed to reach 37°C (which took approximately 30 min) before the start of the experiment. A custom holder was designed and 3D printed to align the Verasonics L7-4 probe perpendicularly to the HIFU acoustic axis when the imaging plane passed through the focus of the HIFU transducer. Once the HIFU transducer and imaging array were placed in the holder, no further adjustment was needed. To continuously estimate temperature in the focal zone, a needle thermocouple (MT-29 / 5HT, Phymep, Paris, France) was inserted into the liver, 0.5 mm behind the imaging plane; thus, the thermocouple did not interfere with the region undergoing RF analysis. During HIFU sonication, electronic beam steering was used to ensure that the maximum pressure zone was at a depth of 22 mm deep in the liver tissues. With these settings, the -6 dB beamwidth and depth of field were 6.8 mm and 12.9 mm, respectively. The free-field acoustic power was set to 9W, 12W or 16W in the different experiments. HIFU sonications were performed using a duty cycle to allow ultrasound imaging and raw radiofrequency data acquisition during exposures. HIFU sonications were performed with 2410-s on-cycles, separated by 2-s off-cycle (duty cycle of 83%). During the cooling phase, radiofrequency signals were also acquired every five degrees until the sample reached 40 degrees. The experimental setup is schematically presented in figure 4 (left part) and the sequences are shown in figure 4 (right part). Ten additional liver samples were included in this experimental section and heated using a water bath at 70°C for histological comparisons. The size of the samples was smaller (2 cm x 2 cm x 2 cm) to obtain homogeneous heating using the same heating time process as used for the HIFU-treated samples. When the HIFU treatment was completed, the treated zone of the liver was sliced along a plane passing through the middle of the HIFU lesion and perpendicular to the sample surface. For samples heated in the water bath, the slicing plane passed through the middle of the sample. Each sample was then placed on an even surface containing a reference scale, and a photograph was taken to generate a near-isometric view of the sample. The samples were then fixed in a 4% formaldehyde solution. After 48h, the samples were transferred to phosphate-buffered saline, dehydrated using increasing concentrations of alcohol, treated with intermediate medium and embedded in paraffin. The embedded samples were sliced and stained using hematoxylin and eosin. The tissues were examined with a brightfield microscope. Macroscopic and microscopic views of the treated samples were then analyzed, as explained in the following sections. Attenuation measurements The Verasonic ultrasound imaging probe with its axis perpendicular to the HIFU acoustic axis was used to measure the attenuation coefficient of the liver sample using plane wave emissions. This coefficient is essential to compensate for attenuation effects between the region of interest (ROI) and the imaging probe. The backscattered intensity I(z) was obtained based on all N = 128 elements of the probe as follows: ^^(^^) = 1∑^^ |^^ (^^)|2^^ ^^ = 1 ^^ (equation 1)where ^^^^(^^) is the RF signal acquired by the i-th element. Considering that the average backscattered intensity can also be represented as a plane wave it follows that this quantity can be described by the Beer-Lambert law : ^^(^^) = ^^0^^ ‒ 4^^(^^^^)^^ (equation 2)where: •^^^^ is the central frequency of the probe, and• ^^ is the depth obtained by time-of-flight analyzis.In soft tissue, such as the liver, it is relevant to consider that the attenuation varies linearly with frequency: ^^(^^) = ^^0^^ (equation 3)where: •^^0 is the attenuation slope, and• ^^ is the frequency.Even if the attenuation is expected to significantly change with temperature, because of the change in tissue viscosity, the Applicant has chosen to use a constant value to evaluate the temperature variation with ∆BSE without any prior bias on the temperature. BSE computation. To compute the BSE variation over time, RF signals were gated with a Hanningwindow. Then, the energy ^^ was calculated at position x and time step ^^^^ with respect to theprevious acquisition at time step^^^^ ‒ 1as: (equation 4) where •^^^^^^^^^^ is the RF signal compensated for the attenuation effect and• the operator ^.^ denotes spatial averaging.Δ^^^^^^ is then deduced based on ^^(^^^^,^^) as follows:Δ^^^^^^(^^^^,^^) = 10^^^^^^10(^^(^^^^,^^)) (equation 5)This processing method only enables the measurement of energy value higher thenthe one measured at first time step ^^ = 0 ( ^^ must remain positive to compute Δ^^^^^^ ).Nevertheless, a relative cooling measure between time steps ^^ > 0 is possible.∆BSE prediction accuracy The accuracy of theΔ^^^^^^thermometry measurements (see equation 6) was used to determine the consistency between the measured temperature change and the actual temperature change (i.e., temperature measured using the thermocouple). Given the importance of maintaining a proper heating range, the Applicant considers an accuracy of ≤ 5°C to be suitable. The bias was calculated as the mean error between ∆BSE thermometry and thermocouple measurements (see equation 7). This parameter shows if the temperature is over- or underestimated. A bias of ≤ 0.5+C was considered appropriate. Accuracy and bias measurements were described as the mean ± standard deviation and are shown in equations 6 and 7: 6) where •^^∆^^^^^^^^^^^^,^^ is the average Δ^^^^^^ thermometry in the ROI, T• is the average temperature measured with the thermocouple,and •n is the number of measured time points.The ROI was a two-dimensional region with 2 mm width and 2 mm length (corresponding to approximately 10% of the focal zone) centered at the thermocouple location. Estimation of thermal damage The accumulation of energy inside a tissue can be historically quantified with a thermal dose model based on the equivalent time at 43°C,^^43, defined, at time^^^^^^^^of treatment, as : (equation 8)where:• ^^43(^^) is the thermal dose in cumulative equivalent minutes (CEM),• T is the current tissue temperature and• t is the time of observation.The commonly accepted threshold value for irreversible damage is of ^^43 = 250 ^^^^^^in the liver. This quantity provides a conservative predictor of the extent of the thermal lesions, according to hyperthermia and previous HIFU literature on various types of soft tissue. Hence, equation 8 was used to monitor energy accumulation in HIFU procedure. However, temperature reached during such procedure leads to fast increase of ^^43: one second at 65 degrees leads to an increase of 3^^105^^43. Thus, thermal lesions were considered to be immediately damaged above such temperatures. To go beyond thermal dose analyzis, two thresholds were defined based on equation 18 to determine whether an ROI was likely or certainly treated: Δ^^^^^^^^^^^^^^^^^^^^^^^^ = 60 ‒ ^^(^^ = 0)8.7^^^^ (equation 9a)Δ^^^^^^ 70 ‒ ^^(^^ = 0)^^^^^^^^^^^^^^ = 8.7^^^^ (equation 9b)where : •^^ = ^^^^^^10(^^^^^^^^^^) ≈ 2 β is a constant that compensates for the surfaceaverage performed based on the^^^^^^^^^^pixels to compute Δ^^^^^^ , and •T = 60°C and T = 70°C are the temperature thresholds chosen in order toaccount for the uncertainty in the calibration curve (see equation 18 and figure 7). Microstructure analysis The backscattered energy of a collection of discrete particles of the same size can be written as: ^^^^^^(^^) ∝ ^^0^^^^(^^)^^(2^^) (equation 10)Where: •^^^^(^^) is the differential backscattering cross section of a particle and ^^0 isthe scatterer density. •^^(^^) is the structure factor that accounts for spatial correlations between thescatterers. This quantity depends on the scattering vector q = ki− ks, where ki= k is the incident wavevector and ks= −k is the backscattered wavevector. For isotropic dilluted media with no spatial order, S(q) is constant and equal to 1. Thus, in this case, the BSE depends only on the scattering properties of each particle. For dense media, ^^(^^) behaviour becomes more complex and can induce various propagation behaviors. Describing the backscattered energy with equation 10 requires several assumptions. First, the scattering is assumed to be due to the impedance mismatch between the host medium and well-separated particles of the same size. Second, the attenuation is assumed to be neglectable (or are compensated for) in the host medium. Finally, it is assumed that no shear waves propagate in the host medium. Under these assumptions, in the case of fluid particles, the scattering cross-section can be described by Anderson^s theory and is of the form: (equation 11)A reasonable assumption is that the hepatocyte network is responsible for scattering inside the surrounding host medium, even if the weights of nuclear and cell scattering remain unknown. In this experimental section, the Applicant considers the whole ensemble of cells and nuclei as a single phase. This choice is relevant as long as an acoustic impedance is not associated with this phase, which is acceptable as the interest here is in the spatial organization of the surrounding medium. This choice leads to the description of a liver sample as a two-phase fluid medium composed of the hepatocyte network and extracellular network. The development that follows provides a framework to study the microstructure of two-phase media with quantities that are linked to the structure factor^^(^^)and are easily accessible based on the histological sections. This development is valid under the assumption of an infinite and statistically homogeneous medium. Considering a medium composed of N well-separated particles, the microstructure is commonly described as a random variable (^^) : (equation 12)where ^^^^is the position of particle j. Alternatively, a continuous two-phase medium can be described by binarizing histological images, leading to a binary matrix ^^^^^^^^ = 1 ^^^^^^^^^^^^ ^^ℎ^^ ^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^{ 0 ^^^^^^^^^^^^^^ (equation 13)The ^^^^(^^) defined in Eq. (13) is a generalization of ^^(^^) (see equation 12) in thecase of a two phase medium. The binary picture ^^^^(^^) is also a straightforward tool to compute the concentration of the scattering phase inside the medium. The concentration^^is defined as :(equation 14)where ^^^^^^^^is the total number of pixels in the histological image. The results are provided in terms of the surface fraction even if the medium is three dimensional. With the goal of evaluating the spatial fluctuations inside the continuous two-phase medium, the autocovariance function^^^^(^^)is either written straightforwardly as Alternatively, this function can be formulated based on the Wiener-Khinchin theorem as (equation 15)The second expression (equation 15) was used in this experimental section for the sake of numerical simplicity. In the case of a medium composed of discrete particles and described by equation 12, the corresponding function,^^^^(^^), is proportional to the pair correlation function ℎ2(^^) . To further this analysis, the spectral properties of the microstructure were derived based on the spectral density of ^^^^(^^) : (equation 16)Notably,^^^^(^^)has long been used to elucidate heterogeneous media through scattering experiments. In the case of discrete disorder, the structure factor ^^(^^) can be defined with thehelp of ^^^^(^^) (analogous to ^^^^(^^) ) by the relation:^^= ^ (^^)^^(^^) ^^^0 (equation 17)The previous work, among others, by Zachary and Torquato describes the linkbetween ^^^^(^^) and ^^(^^) in the case of random heterogeneous media composed of sphericalparticles. Temperature elevation visualization Figure 5 shows the exponential decrease in the acoustic intensity ^^(^^) measured with an L7-4 ultrasound imaging probe placed perpendicular to the HIFU acoustic axis. This finding demonstrates that the plane wave assumption for the backscattered beam isjustified. Based on the slope an attenuation coefficient of ^^0 = 0.06 ^^^^ / (^^^^.^^^^^^) wasdeduced; this value is consistent with values commonly described in the literature. This method was based on the plane wave assumption and a plane wave imaging probe was required to obtain the initial measurement of the attenuation coefficient. This value was used to compensate for losses occurring between the HIFU probe and heated area that impact the radiofrequency data. HIFU sonications were created in 11 liver samples with different HIFU acoustic powers of 9 W, 12 W or 16 W. The treatment time was fixed at 288 s using 2410-s on- cycles separated by 2-s off-cycles (duty cycle of 83%). Under these conditions, the temperature increase was slow enough to be measured precisely during the ultrasound imaging process. The RF data acquired by the imaging probe in the center of the HIFU transducer (in this case, the imaging plane contains the HIFU acoustic axis) were then usedto estimate Δ^^^^^^ in the focal zone. Δ^^^^^^ was first calculated in a two-dimensional regionof interest that was centered at the thermocouple placed in the focal zone. The size of this region of interest was set to 2 mm, which corresponded to approximately 10% of the focal zone; this size was determined so that the region of interest was not affected by spatial temperature changes. The temporal variation in the temperature and Δ^^^^^^ was determined using three acoustic powers, as presented in Figure 6. There was good agreement between the Δ^^^^^^ and temperature variations over time during heating. Δ^^^^^^ was also consistent with the temperature variations during cooling when the maximal temperature was less than 70°C. Interestingly, when the temperature exceeded 80°C, the value ofΔ^^^^^^remained approximately constant over time as the temperature decreased. This suggests that for temperatures higher than 70°C, tissuechanges linked to Δ^^^^^^ variations become permanent. Variations in Δ^^^^^^ outside the focalzone were less than 10% of the maximal value that was reached. Correlation between temperature increase and BSE variation Δ^^^^^^values measured in all samples were grouped in a global dataset. A high correlation was found between the temperature increase andΔ^^^^^^(R = 0.94, p < 0.001). A linear fit was used to correlate the temperature increase withΔ^^^^^^(Figure 7). Importantly, the linear relation allowed to base the signal processing on two consecutive acquisitions, limiting the consequences of organ motion for in vivo applications. The temperature increaseΔ^^can be written as: Δ^^ = 8.7 ^^ Δ^^^^^^ (equation 18)The robustness and accuracy of the BSE thermometry prediction were evaluated by quantifying the temperature accuracy using a comparison with thermocouple measurements in the focal zone. As presented in Figure 7, the average accuracy and bias were equal to or better than the acceptable threshold, which were fixed at 5°C for the accuracy and ± 0.5°C for the bias, respectively. The median accuracy was 4.4°C, and the median bias was 0.0°C. The linear relationship between the temperature and Δ^^^^^^ was then applied to the ultrasound images, as shown in Figure 8. Although no significant changes were observed in the sonogram data, a clear temperature increase was observed in the focal zone. The temperature maps (F-J) as a function of time were used to calculate the thermal dose ^^43(K-O). Because accurately defining the borders of the treated area in the sliced tissues is difficult, two closed contour lines were drawn. These lines were determined based on visual inspections of the detected lesions, as shown in Figure 9. The inner dotted lines delineate the unambiguously coagulated tissue region. The outer solid lines define a region between the inner coagulated region and the clearly untreated tissues. As shown in Figure 8, most of the energy first accumulates at the center of the lesion (K) and then expands due to thermal diffusion. When the thermal dose concept was applied, the region in the uncertain area was underestimated compared with the region determined by equations 11a and 11b (P^T), which are more accurate in detecting the unambiguous ablated region. Three examples of HIFU lesions are presented in figure 9, with the zones visually estimated as unambiguously coagulated (dotted lines) and likely coagulated (solid lines) regions. Table 1 describes the estimated treated areas using Δ^^^^^^ and the HIFU lesion visual inspection results. In each case, the proportion of the unambiguously treated area is larger than 80%, showing the reliability of the proposed method for monitoring HIFU treatment. In contrast, a relatively large proportion of certain and uncertain treated areas can sometimes be observed outside the outer edges (for sample 2, 22%). Importantly, the uncertainty in the estimation of the coagulated area based on the macroscopic observations is mainly due to two reasons. First, tissue decoloration is not binary, and it is difficult to estimate whether slight decoloration can be considered as tissue necrosis. For this reason, a straightforward link between coloration and ablation is difficult to define, and therefore, outer edges of lesions are difficult to delineate. Second, although the liver samples were sliced carefully, the sliced samples may not have been exactly in the imaging plane, leading to uncertainties in the estimation of the lesion boundaries. The most relevant indicator of the reliability of the method described in this experimental section is the correlation betweenΔ^^^^^^and the temperature measured using thermocouples. Thermocouples were placed in the focal zone with high precision using mechanical parts. With this system, the location of the thermocouple can be precisely identified in the imaging plane. IrreversibleΔ^^^^^^changes were observed after cooling only if the maximal temperature reached during treatment was higher than 70°C (see figure 5). Therefore, under these conditions, the structural changes are independent from the thermal effects. Since ex vivo samples can be considered quasistatic during the procedure,Δ^^^^^^was compared before heating and after cooling to quantify the structural changes. As reported in figure 5, only small variations (plotted as small black markers) in ∆BSE were observed outside the focal zone. The order of magnitude of these variations was less than 10% of the maximal temperature. Therefore, the Δ^^^^^^ variation inside the focal zone was analyzed.The variation in Δ^^^^^^ between the initial state ( ^^ = ^^0 , before heating) and final state ( ^^ = ^^^^ ,after cooling) is which can be straightforwardly defined as Δ^^^^^^^^ = ^^^^^^(^^ = ^^^^) ‒ ^^^^^^(^^ = ^^0) (equation 19)Sample 1 2 3 Average (n = 11)Treated area inside inner edges (%) 96 87 81 87 ± 9.3Treated or Uncertain area inside outer 83 68 64 67 ± 11edges (%) Treated or Uncertain area outside 16 22 6 17 ± 9.5outer edges (%) Treated area outside outer edges (%) 9.0 15 0.0 7.5 ± 5.0Table 1: Estimated treated areas using the BSE variation. The corresponding samples are shown in figure 9. This is reported as a function of the maximal temperature increase in figure 10. For the lowest temperature increase, Δ^^^^^^^^remained close to 1 dB, which was of the same order of magnitude as the variation outside the focal zone. If the maximal temperature was lower than 70°C, no or very weak structural changes were observed, and the impact of the heating process on the microstructure appears to be reversible. When the maximal temperature reached during heating was higher than 70°C, Δ^^^^^^^^increased irreversibly up to 3.4 ± 1 dB, with a region of high variation between 75°C and 80°C, which was likely due to both hot spots (creating boiling regions) and regions with temperatures below the boiling point. These irreversible changes can have numerous structural origins. The following results are based on histological analyses and theoretical predictions. Liver samples were viewed as two-phase fluid media composed of the hepatocyte and extracellular networks. Examples of binary images are shown in figure 11. For unheated samples (see figure 11(parts A and F)), cells were well separated from one another. In this situation, the scattering phase can be described as an ensemble of discrete spheres. For heated tissues (see figure 11 parts B to E and G to I), cells were merged, and the aforementioned discretedisorder description is no longer appropriate. The structure factor ^^(^^) in BSE expression(see equation 10) is commonly defined for media with discrete particles. Therefore, to elucidate the changes in the scattering response of a medium composed of two continuous phases (i.e., where scatterer radii and structure factors were not easy to define), the autocorrelation function ^^^^(^^) of the binary histological images and its Fourier transform^^^^(^^) were considered. Both quantities were linked with the structure factor ^^(^^) .Figures 11 (parts A to D) show four histological sections of samples heated at different temperatures during these experiments. Figure 11(part E) shows a histological view of liver tissues heated in a water bath at temperature T = 70°C for the same time as the HIFU treatments. Figures 11 (parts A to D) and figure 11 (part E) can be compared to analyze the effect of HIFU treatment in biological tissue from a structural point of view. When the maximal temperature was below 60°C (see figures 11(part A) and 11(part B)), the cells were well separated from each other. The autocovariance function^^^^(^^)was almost perfectly isotropic for both cases (see figure 11(part K) and figure 11(part L)), as were the spectral densities (see figure 11(part P) and figure 11(part Q)). No visible macroscopic modifications were observed in the liver samples. When higher temperatures were reached (up to 65°C), anisotropy appeared in the cell spatial organization (see figure 11(part C) and figure 11(part D)). The anisotropic orientation is indicated by the arrow andclearly appears in both ^^^^(^^) (see figure 11(parts M and N) and figure 11(I)) and ^^(^^) (seefigure 11(part R) and figure 11(part S)). Two parameters that may explain these results are the maximal temperature reached inside the sample and the direction that the HIFU beam propagates inside the sample. To determine which one of these factors was involved in the microstructure modifications, samples treated with HIFU were compared with samples heated in a water bath at a precise controlled temperature T = 70°C for the same time. The histologicalsections, ^^^^(^^) and ^^^^(^^) are reported in Figures 8(part E), (part O) and (part T) respectively.The comparison of these results with those obtained after HIFU treatments clearly highlights the strong influence of the directivity of the HIFU beam on the organization of the cellular network. High temperatures cause the cells to merge in both procedures, whereas an anisotropic microstructure appears only when the samples are treated with HIFU. The rate of the temperature increase was not responsible for the anisotropy since the heating time was the same in both conditions (HIFU and water bath). The densities were calculated using equation 14 and are reported in Table 2. The density results indicates that the cellular network surface fraction (i.e., the scattering phase) increases as a function of the temperature that was reached during treatment. Very similar densities were obtained for samples treated with HIFU at T = 75°C and those heated in a water bath at T = 70°C. This result suggests that, unlike the spatial arrangement of the cellular network, the concentration was primarily impacted by the temperature increase and not by the direction of the HIFU beam. Considering that the number of cells was constant over time, the increase in the concentration for temperature T > 55°C was linked to an increase in the cell radius a. This observation was consistent with the observed increase inΔ^^^^^^ and Δ^^^^^^ being proportional to ^^6 (equations 10 and 11). Furthermore, theassumption that the cellular network was responsible for scattering was supported by the experimental results. Heating 37°C 55°C 65°C 75°C 70°Ctemperature Concentration 54% 50% 56% 69% 97%Table 2: Surface fraction of the cellular network estimated based on histological observations of tissues heated at different temperatures with HIFU (white tabular cells column 1 to 4) or a water bath (column 5 corresponding to 70°C). Other experiments regarding anisotropy were also carried out in reference to figures 12 to 15. Figure 12 parts A to D illustrate an histological slide of samples of the area 12, namely a liver of a heifer for different experimental conditions. More specifically, figure 12 part A corresponds to an area 12 not having been treated by HIFU. By contrast, figure 12 parts B to D illustrates cases wherein the area was treated by HIFU but reached a respective temperature. In parts C and D of figure 12, the axis of the HIFU is represented by an arrow. For part B of figure 12, the temperature is 55°C; for part C of figure 12, the temperature is 65°C and for part D of figure 12, the temperature is 75°C. The observation of the cellular structure of each part of this figure 12 shows a change of the arrangement of the cells within the tissue induced by the HIFU. In addition, by analyzing the samples corresponding to parts B to D of figure 12, it appears that the anisotropy increases proportionally to the increase of the temperature reached by the tissue. In other words, a stronger temperature implies a stronger anisotropy. Parts A and B of figure 13 also illustrate the influence of the anisotropy on the ultrasound signal. In these experiments, B-mode imaging of a liver tissue heated by HIFU was achieved. This implies that RF signals were acquired according to different angles for observing both the treated zone and the non-treated zone. Part A of figure 13 corresponds to an acquisition achieved along an axis parallel to the HIFU axis whereas part B of figure 13 corresponds to an acquisition achieved along an axis perpendicular to the HIFU axis. The retrodiffused energy by the treated zone strongly varies as visible on parts A and B of figure 13. More specifically, the treated area appears in a clearer color, meaning that the treated area is more echogeneous. The treated are appears to be even more echogeneous in the case of part A of figure 13, wherein the image is taken along the HIFU axis. This is a proof that tissue anisotropy can be detected in the echography signal. This is also confirmed by the results shown in figures 14 and 15. More specifically, the use of an ultrasound probe having the capacity of emitting plane waves with different angles in the area also enables to detect tissue anisotropy. The emission of plane waves with different angles is known as plane wave imaging. In the experiment, the plane wave were emitted by an ultrasound probe situated along the HIFU axis and were at an angle of 18° and an angle of -18° with relation to the HIFU axis. Such position of the ultrasound probe and value of angles are usually feasible in HIFU apparatus. Figure 14 is a map showing the retrodiffused energy difference between two cases: a first case wherein the two plane waves at +18° and -18° are used and a second case in which a plane wave at 0° is used. The dotted line delimits the short field are beyond which the signals are physically exploitable. Figure 15 illustrates a photo of the tissue showing the injury, which can be observed with the eye. This macroscopic observation of the injury formed within the liver tissue by the HIFU treatment. Two cases of the support correspond to 1 cm. On the two images of figure 15, the scale is quasi-identical, the area, which is delimited by the dotted lines have substantially the same size. On the photo of figure 15, at the center of the delimited area, the temperature is higher than the temperature on the edges of the delimited area. The same observation can be drawn from the observation of the map of figure 14, wherein it appears that the value of the retrodiffused energy difference is higher in the central part of the delimited area with relation to the peripheral part of the delimited area. This is another proof that an imaging method based on anisotropy is achievable. Discussion This experimental section demonstrated for the first time that reliable temperature measurements in tissues can be performed using ultrasound imaging between 37°C to 70°C during heating and cooling processes. Moreover, tissue temperature measurements can be obtained up to 100°C if only the heating phase is considered. The experimentalΔ^^^^^^ measurements show a linear relationship between Δ^^^^^^ (in dB) and temperature,with a high correlation. Monitoring can be performed at the frame rate of ultrasound imaging scanners with an accuracy that is within the acceptable threshold of 5°C. Accuracy and bias measurements were performed in regions that were carefully identified based on Δ^^^^^^ thermometry measurements (time and location) and compared with the gold standard, i.e., temperature measurements using thermocouples. The method described in this experimental section exploits the high spatial (on the order of some µm) and temporal resolution (on the order of some ms) of ultrasound imaging. When using MR thermometry, temperature measurements are typically performed with an accuracy of less than 1°C but with lower spatial (on the order of 1 mm) and temporal (on the order of 1 s) resolutions. Given the temperature increase when treating tissues with HIFU (up to 60°C), an accuracy of 5°C was considered acceptable even if it was not ideal. Our approach allows us to estimate that the focal region is correctly placed in tissues during treatment and to evaluate the extent of the treated zone. Furthermore, ultrasound systems have low costs and are available for all patients in clinical settings. Moreover, although the accuracy can be improved, this experimental section was focused on the describing the new method and providing the initial physical explanations. Many methods have been developed to estimate the temperature increase during thermal treatments using ultrasound imaging. These methods were mainly based on speed of sound, elasticity, quantitative ultrasound, attenuation or signal amplitude estimations. Temperature measurements are possible only in a limited range (up to 55°C) when using the first three methods, whereas the other two methods detect the presence or absence of coagulated regions rather than a temperature increase. For temperature increases lower than 70°C, the cooling phase can also be monitored using Δ^^^^^^ , as long as cavitation or boiling is not induced by the treatment. Sublethal temperature elevations (less than 10°C) can be reliably estimated by the Δ^^^^^^ variation to localize the focal zone and validate the focal zone location before using higher temperatures to damage the targeted tissues. Irreversible changes in ^^^^^^^^at the maximal temperature provide information on the structural changes that occur within biological tissues during HIFU procedures. The representation of the soft medium as a two-phase medium allowed us to elucidate the impact of the directivity of the HIFU beam on the cellular network and the increase in cell size through the increase in the cellular network volume fraction. Therefore, after HIFU exposure, the treated area becomes anisotropic at lengthscales close to 10 µm. Under the assumption that backscattering is due to the cellular network, this anisotropy may therefore be involved in the irreversible changes observed in Δ^^^^^^ . Although the two-phase representation is an important simplification compared to reality, the proposed analysis tools can be adapted to random fields with continuous acoustic index variations. Associating an acoustic impedance with each phase will allow us to build acoustic impedance maps, which are known to be powerful tools for predicting Δ^^^^^^ based on histological sections. In conclusion, this experimental section demonstrates that ultrasound imaging can be used to monitor temperature during HIFU treatment. Computing Δ^^^^^^ is a reliable approach to estimate temperature, with an accuracy of 5°C during heating and cooling if the maximal temperature reached is below 70°C. For temperatures higher than 70°C, only the heating phase can be monitored since irreversible changes in Δ^^^^^^ occur. This method could also be used to monitor thermal ablations created using other physical agents, such as radiofrequency, microwaves or lasers. The potential to use ultrasound imaging for temperature monitoring during thermal ablation treatments is of paramount importance because of the high spatiotemporal resolution and clinical availability of ultrasound systems.

Claims

CLAIMS 1.- Method for estimating a physical parameter relative to a region of interest (11) of an area (12), the method for estimating being carried out estimating by a device for estimating (10) and comprising : - a step of collecting backscattered signals, the backscattered signals being ultrasound waves backscattered by the area (12) at several measuring times in presence of at least two of ultrasound excitations of the area (12) with plane waves having a different angle with relation to a reference axis, - a step of calculating, for each angle, a value representative of the backscattered energy based on the backscattered signals, to obtain calculated values, and - a step of estimating a parameter representative of the anisotropy of the structure of the region of interest (11) based on the variation of the calculated value with the angle of the plane wave exciting the area (12). 2.- Method for estimating according to claim 1, wherein the excitations are obtained by generating several plane waves. 3.- Method for estimating according to claim 2, wherein an angle formed by one of the plane waves with the reference axis is comprised between 10° and 20°. 4.- Method for estimating according to any one of the claims 1 to 3, wherein the parameter representative of the anisotropy is a parameter representative of the angular dependence of the calculated values. 5.- Method for estimating according to any one of the claims 1 to 4, wherein the step of estimating further comprises deriving a physical parameter representative of the temperature of the region of interest based on the parameter representative of the anisotropy. 6.- Method for estimating according to claim 5, wherein the method further comprises: - calculating the backscattered energy variation of the region of interest (11) during the time measurement interval at a given position to obtain a calculated backscattered energy variation by adding the elementary backscattered energy variations between two consecutive measuring times,the elementary backscattered energy variation depending from a first signal and a second signal, the first signal being a signal obtained from the backscattered signal at the given position of the region of interest at a first measuring time of the two consecutive measuring times, the second signal being a signal obtained from the backscattered signal at said position at a second measuring time of the two consecutive measuring times, and - estimating a physical parameter representative of the temperature of the region of interest (11) based on the calculated backscattered energy variation, - deducing a thermal contribution by subtracting the estimated parameter representative of the temperature based on the calculated backscattered energy variation and the physical parameter representative of the temperature of the region of interest (11) based on the parameter representative of the anisotropy. 7.- Method for estimating according to any one of the claims 1 to 6, wherein the area is a tissue of a subject. 8.- Method for estimating according to claim 7, wherein the subject is an animal, notably a rabbit, a pig, a rodent or a human. 9.- Method for controlling a thermal therapy of a region of interest (11) of an area (12), the thermal therapy being carried out by a therapy apparatus (24) according to a configuration of the therapy apparatus (24), the method comprising the step of: - carrying out a method for determining according to any one of the claims 1 to 8, to obtain a physical parameter representative of the region of interest (11) of the area (12), and - controlling the configuration of the therapy apparatus (24) by setting the configuration of the therapy apparatus (24) in function of the physical parameter representative of the region of interest (11). 10.- Method for controlling according to claim 9, wherein the configuration of the therapy apparatus (24) is a set of values for parameters of the thermal therapy applied by the therapy apparatus (24), each parameter being chosen in the list consisting of the shape of the emitting surface of the therapy apparatus (24), the number of elements of the emitting surface of the therapy apparatus (24), the duration of the thermal therapy (24), emitted power by the therapy apparatus (24), duty cycle of the therapy apparatus (24), frequencyof the therapy apparatus (24), beam steering of the therapy apparatus (24) and therapy apparatus location. 11.- Method for controlling according to claim 9 or claim 10, wherein the therapy apparatus (24) comprises a heating unit (18), the heating unit (18) being an ultrasound probe, a radiofrequency ablation probe, a laser or a microwave ablation probe. 12.- Method comprising carrying out a method for estimating according to any one of the claims 1 to 8, to obtain a physical parameter relative to a region of interest (11) of an area (12), the method being chosen among: - a method for nondestructive testing, - a method for predicting that a subject is at risk from suffering from an anisotropy related disease, - a method for diagnosing an anisotropy related disease, - a method for identifying a therapeutic target for preventing and / or treating an anisotropy related disease, - a method for identifying a marker, the biomarker being a diagnostic marker of an anisotropy related disease, a susceptibility biomarker of an anisotropy related disease, a prognostic biomarker of an anisotropy related disease or a predictive biomarker in response to a treatment of an anisotropy related disease, and - a method for screening a compound useful as a probiotic, a prebiotic or a medicine, the compound having an effect on a known therapeutical target, for preventing and / or treating an anisotropy related disease. 13.- Calculator (16) adapted to estimate a physical parameter relative to a region of interest (11) of an area (12), the calculator (16) being adapted to : - obtain backscattered signals, the backscattered signals being ultrasound waves backscattered by the area (12) at several measuring times in presence of at least two of ultrasound excitations of the area (12) with plane waves having a different angle with relation to a reference axis, - calculate, for each angle, a value representative of the backscattered energy based on the backscattered signals, to obtain calculated values, - estimate a parameter representative of the anisotropy of the structure of the region of interest (11) based on the variation of the calculated value with the angle of the plane wave exciting the area (12).14.- Device for estimating (10) a physical parameter relative to a region of interest (11) of an area (12), the device for estimating (10) comprising : - an ultrasound (14) probe adapted to collect backscattered signals, the backscattered signals being ultrasound waves backscattered by the area at several measuring times in presence of at least two of ultrasound excitations of the area (12) with plane waves having a different angle with relation to a reference axis, and - a calculator (16) according to claim 13. 15.- Therapy system comprising: - a therapy apparatus (24), the therapy apparatus (24) being adapted to carry out a thermal therapy of a region of interest (11) of an area (12) according to a configuration of the therapy apparatus (24), and - a device for estimating (10) according to claim 14, the device for estimating (10) being further adapted to control the configuration of the therapy apparatus (24) by setting the configuration of the therapy apparatus (24) in function of the physical parameter representative of the region of interest (11).

Images