METHOD AND DEVICE FOR FUNCTIONAL IMAGING OF THE BRAIN

Description

FIELD OF INVENTION

[0001] The present invention relates to methods and devices for functional brain imaging. BACKGROUND OF THE INVENTION

[0002] Functional brain imaging involves imaging transient changes in a brain region linked to neuronal activity, in animals or humans. The brain regions corresponding to different specific functions (e.g., movement of certain limbs, use of certain sensory receptors, emotions, memory, etc.) are listed in known functional anatomical atlases, corresponding respectively to various animal species or to humans (e.g., the "Paxinos" atlases), and functional brain imaging makes it possible to identify and measure neuronal activity in some of these functional areas, in correspondence with a specific activity of the human or animal on which the functional imaging is performed.

[0003] Some functional brain imaging techniques are based on measuring the electrical currents (EEG) or magnetic fields (MEG) created by neurons. These techniques have very limited spatial resolution due to the complexity of the inverse problem that must be solved to determine the position of the sources of electrical activity.

[0004] Other techniques for imaging these functional changes are based on neurovascular coupling: when neurons are highly active in a brain region, a supply of glucose is required in that region. To achieve this, blood flow increases in that specific area. By imaging blood flow in the brain, it is possible to deduce which functional areas are activated in the functional atlas of the human or animal whose brain is being imaged.

[0005] Several techniques can be used to image these changes in blood flow: Functional MRI (also called "fMRI") which images the change in oxygenation linked to a change in blood flow, nuclear imaging (PET) which images the uptake of glucose supplied by the blood flow, and functional ultrasound imaging.

[0006] In fMRI, identifying the activated functional area is straightforward, since MRI images are of very high quality and allow for relatively easy identification of the anatomy and therefore the activated functional area of ​​the brain. Functional imaging is then performed and superimposed onto an anatomical image acquired by the fMRI machine before the functional image is acquired. However, this technology has several drawbacks: fMRI machines are very expensive and bulky, and offer good spatial resolution only at the cost of a significant decrease in temporal resolution, which makes it impossible to image transient phenomena in the brain (such as epileptic seizures).

[0007] PET has very poor spatial resolution. It provides interesting functional information, but once the activated area is identified, it is impossible to determine which anatomical or functional region of the brain it corresponds to. Therefore, it is necessary to combine a PET imaging machine with an MRI or CT scanner to obtain a usable result. Such an imaging suite is, however, extremely expensive.

[0008] Functional ultrasound imaging is based on ultrasensitive imaging of blood flow (Macé et al, "Functional ultrasound imaging of the brain: theory and basic principles", IEEE Trans Ultrason Ferroelectr Freq Control. 2013 Mar;60(3):492-506), the variation of which over time provides access to activated functional areas (Macé et al, "Functional ultrasound imaging of the brain", Nature Methods, 8, 662-664, 2011).

[0009] Functional ultrasound imaging allows for obtaining, at relatively low cost, an extremely precise vascular image of the brain and the localization of activated areas within this vascular image. However, a challenge lies in correlating the vascular image with an anatomical image to identify functional areas. Indeed, the anatomical image acquired by the functional ultrasound imaging device is of insufficient quality to allow for the identification of imaged functional areas.

[0010] Therefore, only an expert operator is able to identify the functional areas imaged in functional ultrasound imaging: he must identify by eye in the vascular image, global structures of the brain which allow him to position the image in the brain, then try to find where the functional area of ​​interest is located using an atlas.

[0011] He is also known for the imaging techniques of the following documents: an article by Emilie MACE et al. entitled "Functional ultrasound imaging of the brain" , an article by Martijn E. VAN RAAIJ et al. entitled “Functional micro-ultrasound imaging of cerebral rodent hemodynamics” , an article by X. PENNEC et al. entitled "Tracking brain deformations in time sequences of 3D US images" , a document US 2013 / 131495 A1, and a document US 2008 / 275340. SUMMARY OF THE INVENTION

[0012] The present invention aims in particular to overcome these drawbacks.

[0013] To this end, the invention proposes a method for functional brain imaging by ultrasound according to claim 1.

[0014] The aforementioned atlases (a functional brain atlas corresponding to a cerebral vascular atlas) allow for the automatic identification, through pattern recognition, of the brain's functional areas visible on the vascular image being studied. The identification of visualized brain functional areas can thus be performed easily and quickly, even by a non-expert user. Functional and vascular atlases corresponding to different subject classes can be created in advance and made available in a computer-based atlas library. A subject class can, for example, correspond to one or more criteria chosen from among: species, sex, age, and weight.

[0015] According to particular embodiments, the process may also include one or more of the features of claims 2 to 14 when these features are technically compatible.

[0016] In various embodiments of the process according to the invention, one may optionally also use one and / or the other of the following provisions: the vascular image to be studied is representative of blood flow in at least one part of the brain; the vascular image to be studied includes at least one hemodynamic information chosen from: cerebral vascular volume (cf (Macé et al, "Functional ultrasound imaging of the brain: theory and basic principles", IEEE Trans Ultrason Ferroelectr Freq Control. 2013 Mar; 60 (3):492-506), power Doppler, blood flow velocity, color Doppler, a value representative of circulatory resistance (Pourcelot index or other); The reference functional vascular map is a three-dimensional map; for example, it may include a plurality of brain slices in successive planes; each initial vascular image is three-dimensional; each initial vascular image is two-dimensional, the reference functional vascular map being obtained from reference vascular images acquired in a plurality of successive planes; a stimulus is applied to the subject and activation of at least one functional area (1c) is detected on the cerebral vascular image (IVO) following this stimulus; a targeted action towards at least one identified functional area is visualized on the cerebral vascular image, and this action is guided towards said identified functional area;said action is chosen from electrode implantation, fluid injection, remote stimulation by a wave (electromagnetic, sound, ultrasound, light); the imaging step is performed with at least one ultrasound probe movable by motorized means, and said at least one probe is moved until the cerebral vascular image corresponds to at least one desired functional area; the imaging step (a) is performed with at least one ultrasound probe adapted to produce a three-dimensional image, and a two-dimensional image corresponding to at least one desired functional area is extracted from this three-dimensional image.

[0017] Furthermore, the invention also relates to a device according to claim 15. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Other features and advantages of the invention will become apparent during the following description of one of its embodiments, given by way of non-limiting example, with regard to the accompanying drawings.

[0019] Regarding the drawings: there figure 1 is a schematic view of a device for implementing a process according to an embodiment of the invention, the figure 2 is a block diagram of part of the device of the figure 1 , there figure 3 is a perspective view of a rat brain showing an imaging plane in the case of two-dimensional imaging and a reference coronal plane, the figure 4 is an example of a vascular image to be studied that can be obtained by ultrasound imaging, for example, from the brain of the figure 3 , in a coronal plane; the figure 5 is an example of a functional brain map that could correspond, for example, to the brain of the figure 3 , in a coronal plane; the figure 6 shows a set of initial vascular images taken in successive coronal planes on a rat brain in order to contribute to the creation of a cerebral vascular atlas of a certain class of rats; the figure 7 schematically illustrates the creation of the aforementioned cerebral vascular atlas; the figure 8 schematically illustrates the acquisition of a coronal cerebral vascular image incorporating a functional map according to the invention, using only ultrasound imaging, and the figure 9 shows an example of successive functional vascular images obtained by the process of the invention, allowing real-time monitoring of the activation of certain functional areas of the brain. DETAILED DESCRIPTION

[0020] In the different figures, the same references designate identical or similar elements.

[0021] There figure 1This shows an example of a device adapted for performing vascular imaging of the brain 1 of a human or animal subject under study, by emission and reception of compression ultrasound waves, for example with frequencies between 1 and 40 MHz. The emission and reception of ultrasound waves can be carried out by means of an array 2 of ultrasound transducers either by passing through all or part of the skull 1a of the subject, or directly in contact with the brain 1, in particular at the level of one or more trepanations of the skull 1a.

[0022] The imaging system includes, for example, as illustrated in the figures 1 and 2 : an array 2 of n ultrasonic transducers 2a (T 1 -T n ), comprising for example a few hundred transducers 2a, this array 2 being for example a strip of transducers (1D array) adapted to produce a two-dimensional (2D) image or a two-dimensional array adapted to produce a three-dimensional (3D) image; an electronic circuit 3 controlling the array of transducers 2 and adapted to make the array of transducers emit ultrasonic waves and acquire the signals captured by this array of transducers; a computer 4 or similar to control the electronic circuit 3 and visualize the ultrasonic images obtained from said captured signals.

[0023] As depicted on the figure 2 For example, electronic circuit 3 may include: n analog / digital converters 11 (A / D i ) individually connected to the n transducers T 1 -TN of the transducer network 2, n buffer memories 12 (B i ) respectively connected to the n analog / digital converters 11, a central processing unit 13 (CPU) communicating with the buffer memories 12 and the computer 4 and adapted for processing signals sent to and received from the transducer network 2, a memory 14 (MEM) connected to the central processing unit 13.

[0024] The imaging device shown on the figures 1 and 2is adapted to perform synthetic ultrasound imaging of the brain allowing fine imaging of the cerebral vascular network and to provide at least hemodynamic information (i.e. relating to blood flow), as described in particular by Macé et al. (aforementioned publications "Functional ultrasound imaging of the brain: theory and basic principles", IEEE Trans Ultrason Ferroelectr Freq Control. 2013 Mar; 60 (3):492 - 506 ) and "Functional ultrasound imaging of the brain", Nature Methods, 8, 662-664, 2011) and in document EP2101191.

[0025] The hemodynamic information given by the cerebral vascular image thus obtained can be chosen in particular from: cerebral vascular volume, power Doppler, blood flow velocity, color Doppler, a value representative of circulatory resistance (Pourcelot index or other).

[0026] As depicted on the figure 3In the case of two-dimensional imaging, the cerebral vascular image can be acquired in a plane P, for example, with a direction close to a coronal plane P0, but in practice most often different from the perfect coronal plane P0. The resulting cerebral vascular image IV, visible on the figure 4 This allows visualization of the brain's blood vessels (1b) with excellent definition, including microvessels. The aforementioned hemodynamic information can be conveyed by the light intensity of each pixel and / or by a color scale.

[0027] This hemodynamic information reflects local brain activity and therefore allows for functional imaging, provided that each pixel of the image can be located in a functional anatomical atlas of the brain (Paxinos or other), of which a coronal IF section is shown as an example on the figure 5Such a functional anatomical atlas contains a typical anatomical atlas of the brain (shape of brain tissues) and a typical functional atlas consisting of functional areas 1c identified on this typical anatomical atlas. Each of the functional areas 1c of the typical functional atlas corresponds to a specific motor, sensory, or cognitive function and bears a specific reference corresponding to that function. Each functional anatomical atlas is valid for a class of subjects studied, corresponding, for example, to a species, a sex (male / female), a weight range, and an age range of the subject studied (or only some of these criteria, plus others possibly).

[0028] However, it is very difficult for an operator to locate the cerebral vascular image IV in the functional anatomical atlas, since on the one hand, the cerebral vascular image IV does not correspond to the anatomical images that can be extracted from the standard anatomical atlas and on the other hand, the P plane of the cerebral vascular image is generally not exactly a coronal P0 plane as explained above.

[0029] To remedy this drawback, the invention provides for the advance, for at least certain classes of subjects to be studied, of a cerebral vascular atlas type AV corresponding to said class of subjects, in correspondence with a cerebral functional atlas type AF corresponding to said cerebral vascular atlas AV and whose cerebral functional areas 1c are identified in this cerebral vascular atlas AV.

[0030] These two atlases can be established in advance, notably during a preliminary step (p) of functional reference vascular mapping, comprising the following sub-steps: (p1) a reference imaging substep during which both the following are performed: said cerebral vascular atlas (obtained by vascular ultrasound of the brain, or by CT angiography, MRI, CT scan) of at least one subject of the class of subjects considered, and a reference anatomical atlas obtained by imaging the brain of said at least one subject of the class of subjects considered, by a second type of imaging giving a more precise anatomical image of the brain than ultrasound imaging, for example by MRI, (p2) a reference functional mapping substep during which said cerebral functional atlas is determined from a functional anatomical atlas (Paxinos or other) which includes at least one standard anatomical atlas of the brain and functional areas identified on this standard anatomical atlas,This sub-step of functional mapping of reference is carried out by matching said reference anatomical atlas with the standard anatomical atlas, in order to identify the functional zones of the functional anatomical atlas on said cerebral vascular atlas.

[0031] During said substep (p1) of reference imaging, several initial vascular images IV ( figure 6) by vascular ultrasound imaging of the brain and several baseline anatomical images by said second type of imaging. Said cerebral vascular atlas and said reference anatomical atlas are determined by statistical calculation respectively from said baseline vascular images and baseline anatomical images (for example, respectively by averaging said baseline vascular images and baseline anatomical images), as schematically illustrated in the figure 7 for the cerebral vascular atlas in rats.

[0032] During substep (p2) of functional reference mapping, the reference anatomical atlas can be matched with the standard anatomical atlas by pattern recognition. For example, a geometric deformation is determined to transform the reference anatomical atlas into the standard anatomical atlas, and this geometric deformation is then applied to the cerebral vascular atlas, thus matching the cerebral vascular atlas with the functional areas of the functional anatomical atlas.

[0033] Once the AV-type cerebral vascular atlas corresponding to a class of subjects has been established, in correspondence with the AF-type cerebral functional atlas for the same class of subjects, functional areas can be easily and automatically identified on an IVO cerebral vascular image obtained solely by ultrasound imaging, using the process illustrated in the figure 8 : (a) Imaging:

[0034] We image a brain 1 of the subject by ultrasound imaging, to obtain the vascular image to be studied IV0. (b0) Possible preliminary identification:

[0035] Optionally, a preliminary rough localization is first performed during which at least one characteristic area normally present in all subjects under study and in the cerebral vascular atlas AV is automatically detected on the vascular image to be studied IV0, and the vascular image to be studied IVO is thus roughly located in the cerebral vascular atlas AV. This characteristic area may be a point or an invariant structure of the brain, for example the circle of Willis, the sylvian veins, the anterior cerebral artery, the largest arteries of the brain. (b) Identification:

[0036] Using pattern recognition, the vascular image to be studied (IV0) is automatically compared by correlation with the corresponding cerebral vascular atlas (AV) for the subject class to which the subject belongs. This allows the vascular image to be studied (IVO) to be located within the cerebral vascular atlas (AV). In practice, this step of locating the vascular image to be studied (IVO) within the AV vascular atlas is performed by determining, through pattern recognition, a vascular image (IVR) from the AV vascular atlas that best corresponds to the vascular image to be studied (IV0).The AV vascular atlas may be composed of a series of two-dimensional IV cerebral vascular images taken in successive adjacent planes (e.g., coronal planes, or others), and the aforementioned localization may consist, in a simplified version, of choosing the IVR cerebral vascular image closest to the IVO vascular image to be studied (which in this case is a two-dimensional image). In a more elaborate version, and when the IVO vascular image to be studied is two-dimensional, this localization also consists of determining, by shape recognition, the orientation and position of the plane of the IV0 image, and reconstructing an IVR image from the AV vascular atlas in this plane (whether the atlas is three-dimensional or composed of a series of two-dimensional IV cerebral vascular images taken in successive adjacent planes). (c) Identification:

[0037] Using the AF functional cerebral atlas corresponding to the AV cerebral vascular atlas, at least one 1c cerebral functional area is identified on the IV0 vascular image to be studied. More precisely, an IFR image is determined from the AF functional atlas, corresponding to the aforementioned IVR image, and this IFR image is correlated with the IV0 vascular image to be studied, by geometric deformation, to determine a functional IFO map (formed by the limits of the 1c functional areas), which is superimposed on the IVO vascular image to be studied to obtain a functional IVF0 vascular image of the brain.

[0038] More specifically, during the localization step (b), a geometric deformation can be determined allowing the transition from the IVR image to the vascular image to be studied IV0, and during the identification step (c), the functional IFO map of the vascular image to be studied IVO can be determined by applying this geometric deformation to the IFR image from the AF brain functional atlas.

[0039] Once the IFO functional map is established, it remains unchanged as long as the ultrasound imaging probe 2 remains stationary (this probe can, for example, be rigidly fixed to the subject's skull 1a, or via a motorized actuator allowing controlled movement of this probe). IVO cerebral vascular images can then be acquired over time (see figure 9), for example, in synchrony with external or internal stimuli applied to the subject under study (including conscious or unconscious brain activity). These cerebral vascular images are superimposed on the IF0 functional map, in order to visualize and possibly record the activation of functional areas 1c over time.

[0040] It is also possible to visualize and guide the implantation of electrodes or the injection of specific therapeutic agents towards targeted functional areas, or even remote stimulation by a wave (electromagnetic, ultrasonic, sound or light).

[0041] It should also be noted that the method described above makes it easy to move the ultrasound probe 2 to a desired position to image certain defined functional areas, either when said probe 2 is rigidly fixed to the subject's skull 1a, or via a motorized actuator allowing controlled movement of this probe. Indeed, since the cerebral vascular image IVO is located in the functional atlas AF, it is easy to determine the movement necessary to visualize a desired functional area 1c.

[0042] Similarly, when ultrasound imaging is performed with at least one ultrasound probe 2 adapted to produce a three-dimensional image, a two-dimensional image corresponding to at least one desired functional area 1c can be extracted from this three-dimensional image.

Claims

1. A method for functional imaging of the brain by ultrasonic waves, comprising the following steps: (a) a vascular imaging step during which a brain (1) of a human or animal subject is imaged by ultrasonic imaging, in order to obtain a vascular image to be studied (VI0), the vascular image to be studied being representative of the blood flows in at least one portion of the brain, said subject belonging to a certain class of subjects, (b) a localization step, during which, by shape recognition, at least one portion of the vascular image to be studied (VI0) is automatically compared with a brain vascular atlas (VA) corresponding to said class of subjects, and the vascular image to be studied (VI0) is thereby located in the brain vascular atlas (VA), (c) an identification step during which a brain functional atlas (FA) corresponding to said brain vascular atlas (VA) and comprising brain functional areas (1c) located in this brain vascular atlas (VA) is used, so that at least one brain functional area (1c) is identified on the vascular image to be studied (VI0).

2. The method according to claim 1, wherein at least during the localization step (b), a portion (VIR) of the brain vascular atlas (VA) is matched with the vascular image to be studied (VI0), and during the identification step, a map of functional areas (FI0) corresponding to said portion of the brain vascular atlas (VA) is determined, and the vascular image to be studied (VI0) is superposed with said map of functional areas (FI0).

3. The method according to claim 2, wherein, during the localization step, a geometrical deformation is determined giving the possibility of passing from said at least one portion (VIR) of the brain vascular atlas (FA) to the vascular image to be studied, and during the identification step, the functional map (FI0) of the vascular image to be studied is determined by applying this geometrical deformation to a portion (FIR) of the brain functional atlas (FA) corresponding to said portion (VIR) of the brain vascular atlas (VA) which corresponds to the vascular image to be studied (VI0).

4. The method according to any one of the claims 1 to 3, wherein in the localization step (b), said shape recognition is carried out by correlation between the vascular image to be studied (VI0) and the brain vascular atlas (VA).

5. The method according to any one of the claims 1 to 4, wherein the vascular image to be studied (VI0) includes at least one piece of hemodynamic information selected from among: the brain vascular volume, the Doppler power, the flow rate of the blood, the Doppler color, a value representative of the circulation resistance.

6. The method according to any one of the claims 1 to 5, wherein the brain vascular atlas contains data stemming from an imaging selected from among: ultrasonic imaging, angioscanner, MRI, CT scanner.

7. The method according to any one of the claims 1 to 6, further comprising, between the imaging step (a) and the localization step (b), a preliminary localization step (b0) during which on the vascular image to be studied (VI0), at least one characteristic area normally present in any subject to be studied and in the brain vascular atlas (VA) is automatically detected, and the vascular image to be studied (VI0) is thus roughly located in the brain vascular atlas (VA).

8. The method according to claim 7, wherein said characteristic area is selected from among: the Willis polygon, the Sylvian veins, the anterior brain artery, the largest arteries of the brain.

9. The method according to any one of the claims 1 to 8, including at least one preliminary step (p) for reference functional vascular mapping, comprising the following sub-steps: (p1) a reference imaging sub-step during which are produced both: - said brain vascular atlas (VA) of at least one subject from the class of considered subjects, - and a reference anatomic atlas produced by imaging the brain of said at least one subject of the class of considered subjects, by a second type of imaging giving a more accurate anatomic image of the brain than ultrasonic imaging, (p2) a reference functional mapping sub-step during which, said brain functional atlas (FA) is determined from a functional anatomic atlas which comprises at least one typical anatomic atlas of the brain and of the functional areas located on this typical anatomic atlas, this reference functional mapping sub-step being carried out by matching said reference anatomic atlas with the typical anatomic atlas, in order to locate the functional areas of the functional anatomic atlas on said brain vascular atlas.

10. The method according to claim 9, wherein at least during the reference imaging sub-step (p1), said brain vascular atlas (VA) is produced by vascular ultrasonic imaging of the brain.

11. The method according to claim 9 or claim 10, wherein during the reference functional mapping sub-step (p2), the reference anatomic atlas is matched with the typical anatomic atlas by shape recognition.

12. The method according to claim 11, wherein during the reference mapping sub-step (p2), a geometrical deformation is determined which gives the possibility of passing from said reference anatomic atlas to the typical anatomic atlas, this geometrical deformation is applied to the brain vascular atlas (VA) and the brain vascular atlas (VA) is thus matched with the functional areas of the functional anatomic atlas.

13. The method according to any one of the claims 9 to 12, wherein during said reference imaging sub-step (p1), respectively, several initial vascular images (VI) are successively carried out by vascular ultrasonic imaging of the brain and several initial anatomic images by said second type of imaging, are successively carried out on several subjects of the class of considered subjects, and said brain vascular atlas (VA) and said reference anatomic atlas are determined by a statistical calculation respectively from said initial vascular images and from initial anatomic images.

14. The method according to any one of the claims 1 to 13, wherein each class of subjects corresponds to at least one criterion selected from among: species, gender, age, weight.

15. A device for applying a method according to any one of the claims 1 to 14, comprising: - vascular imaging organ (2-4) adapted for imaging a brain of a human or animal subject by ultrasonic imaging, in order to obtain a vascular image to be studied (VI0), said subject belonging to a certain class of subjects, - localizing organ (4) by shape recognition, adapted for automatically comparing the vascular image to be studied (VI0), with a brain vascular atlas (VA) corresponding to said class of subjects, and for thereby locating the vascular image to be studied (VI0) in the brain vascular atlas (VA), - identification organ (4) using a brain functional atlas (FA) corresponding to said brain vascular atlas (VA) and comprising brain functional areas (1c) located in this brain vascular atlas (VA), said identification organ being adapted for identifying at least one brain functional area (1c) on the vascular image to be studied (VI0).