Imaging method, apparatus and product, associated computer program

Abstract

------ Imaging Method, Apparatus, and Associated Computer Program Product The present invention relates to an imaging method, an imaging apparatus, and an associated computer program product for producing a three-dimensional image of a patient's brain using microwave imaging. This method utilizes target detection within the brain to reduce the imaged area in subsequent scans, thereby enabling real-time monitoring of the target's evolution within the patient's brain. Figure to be published with the abstract: Figure 1

Description

Title of the invention: Imaging method, apparatus and associated computer program product technical field

[0001] The present invention relates to the technical field of medical imaging, and more particularly to an imaging method, an apparatus and an associated computer program product.

[0002] Brain diseases are the subject of intense research, both from the point of view of their treatment and their diagnosis.

[0003] Once diagnosed, certain brain conditions require monitoring to prevent more serious conditions in patients. Thus, clinical results from studies suggest that in patients with a primary intracerebral hemorrhage, the risk of rebleeding is not negligible: 24% of patients experienced one or more episodes of rebleeding during a mean follow-up period of 84.1 months. This risk appears to be highest during the first year following the initial hemorrhage. The frequency of a second hemorrhage was higher in those whose blood pressure was not controlled, according to these same studies. In fact, the higher the patient's blood pressure, the greater their risk of recurrence.

[0004] It is therefore important to be able to find methods for monitoring the patient's condition, either on a periodic basis or on a one-time basis for patients for whom an imminent risk of stroke has been determined. These methods can be anatomical, biological, or physiological, but essentially rely on brain imaging, with the problem that brain imaging, while potentially effective, requires expensive equipment that is difficult to transport and involves calculations that are hardly compatible with rapid diagnosis.

[0005] In approaches aimed at monitoring a condition, for example the progression of a stroke, Alzheimer's disease, epilepsy, Parkinson's disease (PD), and Tourette syndrome, the signals must be based on anatomical and physiological information extracted during tissue scanning and be comparatively selected to improve the speed of image extraction in terms of focusing on changes in the target's position and structure. Current imaging methods, such as magnetic resonance imaging (MRI) and computed tomography (CT), are therefore difficult to apply. Furthermore, functional MRI (fMRI) and electroencephalography (EEG), which are ac Currently the most used techniques because they can provide excellent detailed anatomical and physiological information related to the monitoring of brain disorders, have disadvantages: the temporal resolution of conventional fMRI is rather low and takes a few seconds to reach its peak, while its spatial resolution is excellent, on the order of a square millimeter, whereas EEG has a temporal resolution on the order of a millisecond, but its spatial resolution is rather low and covers a few square centimeters.

[0006] In support of these imaging methods, computational brain imaging has been one of the largest and most challenging research areas in recent years. Due to the complex nature of brain diseases, various approaches have been explored over the past few decades, each with its own advantages and disadvantages. Microwave brain imaging based on the electromagnetic approach has attracted considerable attention in recent years because of its advantages, such as its non-invasiveness, lack of contact, low cost, and portability. Extracting information about the brain from these systems still presents many challenges. Various processing methods have been proposed to increase the capabilities of these devices, as well as the usefulness of the extracted images for detecting and diagnosing brain diseases, including strokes.These conventional approaches require a large number of antennas to collect the scattered inverse field and thus solve the inverse scattering problem, resulting in a high processing time.

[0007] The invention aims to enable continuous monitoring of strokes, applied to the monitoring of bleeding and blood clot proliferation in stroke patients, using an emerging brain imaging system based on an electromagnetic approach, and allows for an ultra-fast and super-resolution monitoring methodology. In this respect, unlike other conventional static microwave imaging systems that can be considered as performing a blind scan, the dynamic microwave imaging system based on a multi-static array according to the invention has great potential to provide useful tools for stroke monitoring.The main objective is to develop ultra-fast destructive resolution methods to pave the way for cerebrovascular monitoring by performing closed-loop cognitive scans based on a common algorithm / design.

[0008] The hypothesis is that, unlike differential imaging where time-varying activities are detected by subtracting static information, closed-loop functional cognitive imaging is modulated in real time by the Feedback from backscatter recordings of intelligent irradiation is faster and can extract more information. To achieve this, depending on the dynamic changes in brain activity, the transmitted waveforms, and therefore the activated structures of subnetworks and focusing networks, will be cognitively modified and continuously adapted to the changing targets. The proposed radical vision, based on cognitive scanning (CS), uses an information-based approach for the tested scenario. The phenomenon underlying this idea is that, for functional monitoring and imaging, it is not necessary to detect the entire volume of the brain, but only to save / exploit the data that contains relevant information about a predefined set of brain activities.To transform the problem into an informational one, the basic parameters of knowledge-based information synthesis must be obtained, one of which is the existence of prior knowledge or a library to create a knowledge development system and online decision autonomy. The main argument for the superiority of cognitive scanning for functional brain imaging is that it is more flexible for unknown patterns of brain activity than multi-static imaging and can therefore be more widely offered as a monitoring option with fewer contraindications. By identifying an activated region as a target of interest, the microwave imaging scenario is transformed into a microwave detection scenario.

[0009] The present invention therefore relates to a three-dimensional brain imaging method using microwave imaging, characterized in that it comprises the following steps:

[0010] a - to produce on a patient on whose head is arranged a network of microwave antennas controlled by a switching network connected to a controllable signal transmitter-receiver device, a global three-dimensional image of the brain by microwave imaging using all the antennas of the microwave antenna network;

[0011] b - detect at least one target in the global brain image obtained in step a;

[0012] c - calculate a position of at least one target in the global brain image obtained in step a;

[0013] d - to produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas;

[0014] e - merge the global image and the local processing image into an inter- image global three-dimensional median;

[0015] f - detect at least one target in the overall intermediate image;

[0016] g - calculate a position of at least one target in the overall intermediate image;

[0017] h - to produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas;

[0018] i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image;

[0019] j - repeat steps f to i a predetermined number of times.

[0020] Advantageously, the antenna array is formed on a helmet, the antennas being uniformly distributed hemispherically on the helmet. Advantageously, the array comprises 24 antennas, although the invention is not limited in this respect.

[0021] The antennas are transmitting and receiving antennas, allowing the transmission and reception of microwave signals.

[0022] Thus, thanks to the invention, due to the reduced number of antennas and / or samples required for local imaging due to target localization, fewer calculations are necessary, and therefore the image is obtained more quickly, enabling more real-time monitoring. Since the overall image is not recalculated at each iteration, the focus is solely on tracking and imaging the target, thereby reducing image acquisition and calculation time. However, at each iteration, a target is searched for within the overall image, ensuring that no target is missed, particularly in the context of functional imaging where the aim is to understand the evolution of the target's size, location, and nature as a function of the patient's condition, even if the entire image is not recalculated at each iteration.The fact that the system includes all the necessary antennas, but that these are only activated based on the target's location, allows for great flexibility, reduced computational complexity, and high target detection efficiency. Furthermore, less powerful computing equipment is required compared to the state of the art. Finally, the image can be captured by a simple headset equipped with microwave antennas connected to a computing device, enabling a level of portability not possible with current methods and systems.

[0023] Advantageously, for better efficiency, both a subgroup of the antenna array and a reduction in the number of samples are used, which reduces the computational load accordingly.

[0024] The invention also relates to a microwave imaging device of the brain, characterized by the fact that it comprises a headset equipped with a microwave antenna array, a switching network controlling the microwave antenna array, a vector network analyzer connected to the switching network, a processing unit connected to the switching network and the vector network analyzer and controlling the switching network and the vector network analyzer, a human-machine interface connected to the processing unit, the processing unit comprising computing means and memory to perform the following steps:

[0025] a - to produce a global three-dimensional image of the brain by microwave imaging using all the antennas of the microwave antenna array;

[0026] b - detect at least one target in the global brain image obtained in step a;

[0027] c - calculate a position of at least one target in the global brain image carried out in step a;

[0028] d - to produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas;

[0029] e - merge the global image and the local processing image into a global three-dimensional intermediate image;

[0030] f - detect at least one target in the overall intermediate image;

[0031] g - calculate a position of at least one target in the overall intermediate image;

[0032] h - to produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas;

[0033] i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image;

[0034] j - repeat steps f to i a predetermined number of times.

[0035] The antennas are preferably arranged on the helmet in a hemispherical manner, distributed uniformly.

[0036] The processing unit allows the transmission of signals to the headset and the reception of signals from the headset.

[0037] According to one embodiment, the antenna array comprises 24 antennas, preferably butterfly antenna type antennas.

[0038] Advantageously, the switching network is a 2*24 switching matrix in which the 24 ports are connected to the antennas and the two ports are connected to the vector network analyzer.

[0039] The invention also relates to a computer program product comprising instructions which, when executed by a computer, perform the following steps:

[0040] a - to produce a global three-dimensional image of the brain by microwave imaging using all the antennas of the microwave antenna array;

[0041] b - detect at least one target in the global brain image obtained in step a;

[0042] c - calculate a position of at least one target in the global brain image obtained in step a;

[0043] d - to produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas;

[0044] e - merge the global image and the local processing image into a global three-dimensional intermediate image;

[0045] f - detect at least one target in the overall intermediate image;

[0046] g - calculate a position of at least one target in the overall intermediate image;

[0047] h - to produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas;

[0048] i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image;

[0049] j - repeat steps f to i a predetermined number of times.

[0050] According to one embodiment, each step of three-dimensional image realization by microwave imaging includes the collection of diffusion parameter data representing microwaves scattered by the patient's brain in a diffusion tensor, the generation of differential diffusion parameter data by an adaptive connectome calibration process in order to suppress spurious signals, and the processing of the differential diffusion parameter data by a confocal image reconstruction process to obtain a three-dimensional image.

[0051] The adaptive connectome calibration method consists of calculating the geometric connectome between the hemispheric antenna array and the signals passing through the patient's brain, the term connectome being defined in this application as the set of geometric connections / links between the antennas and the Signals. Based on the symmetry of the medium through which the microwave signals travel, it is possible to categorize the signals into differential pairs that follow symmetrical paths. By subtracting these symmetrical signals, all information not related to the target is removed. In practice, the signals are grouped using a cross-correlation matrix, in which signals from channels that follow the same path through the patient's brain and have a high correlation value are placed in identical groups. The differential scattering parameter data are obtained by subtracting the signals in each group from each other.In subsequent scans, where signals whose differential pair has been removed due to the dropping of a number of antennas in the antenna layout optimization process, an adaptive mode is used to determine the differential pair: the correlation of each signal is calculated with the rest of the signals, and the signal that gives the highest correlation value is selected as the differential pair.

[0052] The confocal image reconstruction method uses delayed-multiply sum (DMAS) beamforming by processing differential scattering parameter data. The maximum value of the coherently focused energy in the reconstructed image refers to the brain region of functional interest, for example, an attack region. The location of these areas is considered the position of a possible target. Such a technique is described, for example, in the following publications:

[0053] - Ahadi, M., Isa, M., Saripan, MI and Hasan, WZW (2015), Three local dimensions lization of tumors in confocal microwave imaging for breast cancer detection (Localisation tridimensionale des tumors en imagerie confocale par micro-ondes pour la detection du cancer du sein), Microw. Opt. Technol. Lett., 57: 2917-2929;

[0054] - Salvador, Sara M., and Giuseppe Vecchi, Experimental tests of microwave breast cancer detection on phantoms (Experimental tests of breast cancer detection by microwaves on phantoms), IEEE Transactions on Antennas and Propagation 57, no. 6 (2009): 1705-1712;

[0055] - Babarinde, OJ, Jamlos, MF, Soh, PJ, Schreurs, DP and Beyer, A., Microwave imaging technique for lung tumour detection (Technique d'imagerie par micro-ondes pour la detection des tumors du pneumo), 2016 German Micro wave Conference (GeMiC), 2016, pp. 100-103.

[0056] According to one embodiment, at least one target is detected by calculating the signal-to-clutter ratio (SCR) and the signal-to-mean ratio (SMR) in two-dimensional planes of the three-dimensional image, the at least one target being detected on the surfaces of each two-dimensional plane on which the values ​​of the The signal-to-noise ratio and the signal-to-mean ratio are simultaneously maximized. Such a technique is described, for example, in the following publications:

[0057] - Reimer, T., Solis-Nepote, M. and Pistorius, S., 2020, The application of an iterative structure to the delay-and-sum and the delay-multiply-and-sum beamformers in breast microwave imaging (Application of an iterative structure to delay-and-sum and delay-multiply-and-sum beamformers in breast microwave imaging), Diagnostics, 17 / 06 / 2022, 10(6), p.411;

[0058] - Babarinde, OJ, Jamlos, MF, Soh, PJ, Schreurs, DP and Beyer, A., Microwave imaging technique for lung tumour detection (Technique d'imagerie par micro-ondes pour la detection des tumors du pneumo), 2016 German Micro wave Conference (GeMiC), 2016, pp. 100-103.

[0059] According to one embodiment, the position of at least one target in the three-dimensional image is calculated as the position of the surfaces of each two-dimensional plane on which the values ​​of the signal-to-noise ratio and the signal-to-mean ratio are simultaneously maximum. The probability that the target will be detected is thus maximized.

[0060] Thus, the location of the target region is approximated by superimposing quantitative metric maps such as the maximum value for the signal-to-noise ratio (SCR) and the signal-to-mean ratio (SMR) in a two-dimensional plane, and the spatial dimensions of the target region correspond to the selected SCR and SMR windows.

[0061] According to one embodiment, the subgroup of the microwave antenna array associated with the position of at least one target consists of the antennas having the shortest distance to the position of at least one target.

[0062] According to one embodiment, compressive sampling is applied to the microwave signals emitted by the antennas of the antenna array. The computing power required is thus reduced. The compressive sampling is advantageously implemented by a convex optimization method of the L1 space (which in this application designates the space of functions taking values ​​in R whose absolute value (or the space of functions taking values ​​in C whose modulus) is Lebesgue integrable).

[0063] According to one embodiment, each step of merging two three-dimensional images consists of replacing, in the global image, the part corresponding to the local image, with the local image to obtain the merged image.

[0064] Since the position of the target is known, we can know the region of the global image corresponding to the target, and replace this region in the global image with the local image of the target.

[0065] Only the additional information obtained during processing is added to the overall 3D image.

[0066] The predetermined number of times steps f to i are repeated depends on the choice of the operator using the device according to the invention. The predetermined number of times can in particular be correlated to a duration, in particular a patient monitoring duration, or be correlated to a lack of evolution of a target over time (the imaging by the device is stopped if the target no longer evolves after a certain duration) or to an excessively rapid evolution of the target (significant spatial development of the target over a predetermined duration).

[0067] After step j, the image obtained is an image of the patient's brain on which the target is positioned in space. An evolution of the target over time is also obtained, from step a to step j, by storing and comparing the images.

[0068] The invention therefore uses an intelligent cognitive scanning program. The imaging device of the invention uses a dynamic intelligent structure to create high-resolution images in the shortest possible time. The cognitive scanning paradigm proposed in this invention is the result of a combination of information theory and electromagnetic theory. In dynamic multi-scan mode, feedback is used to augment the input data for subsequent scans. The selection of the antenna array arrangement that determines the desired signals in the processing stage depends on the information from the previous scan and is selected so as to extract new information in the next scan. The brain is irradiated by an antenna array that propagates electromagnetic waves into the head. These waves are then converted into measured reflected signals.After passing through the receiver, they are converted into data according to the type of scan. From this process, information is extracted from the captured data.

[0069] Unlike conventional multi-static microwave imaging systems, the proposed cognitive scanning also provides an additional information extraction / integration process. The main objective presented here is to create a closed-loop cognitive scanning method as a faster alternative to differential imaging that allows for the extraction of more information. Based on cognitive compression measurement and sampling techniques, the information stream and outputs of the proposed device create highly innovative decision-making schemes for ultrafast dynamic microwave imaging systems. Building on this capability, the simultaneous design of a multi-scan algorithm architecture based on cognitive scanning, which includes antenna selection and sub-Nyquist sampling, is implemented. Compression-based sampling enables ultrafast sampling and complete reconstruction of certain classes of unlimited signals. In information-based scanning, temporal and spatial changes occur within the monitoring scenario. These temporal and spatial changes must be intelligently detected and tracked. Furthermore, only variant samples should be extracted by scanning. The extracted data also adds new information to the Knowledge Growing System (KGS) library to estimate the future state of the Target of Interest (TOI). To this end, the invention combines algorithms with architectures for cognitive scanning, data acquisition, and information integration required for very fast tracking to optimize performance.

[0070] The main argument in favor of the superiority of cognitive scanning for static brain imaging is that it is faster and more accurate for multi-static brain imaging and can therefore be more widely offered as a monitoring option with fewer contraindications. By identifying the target area as the target of interest, the microwave imaging scenario is transformed into a detection scenario.

[0071] To better illustrate the object of the present invention, a preferred embodiment thereof will be described below, in conjunction with the attached drawings.

[0072] On these drawings:

[0073] [Fig.1] is a block diagram of the microwave imaging device according to the present invention;

[0074] [Fig.2] is an example representation of a helmet used by the device in [Fig.1];

[0075] [Fig.3] is a flowchart of the imaging process implemented by the microwave imaging device according to the present invention;

[0076] [Fig.4] represents a two-dimensional imaging domain of the human head;

[0077] [Fig.5A] represents a first step in image reconstruction by a confocal image reconstruction process;

[0078] [Fig.5B] represents a second image reconstruction step by a confocal image reconstruction process;

[0079] [Fig.5C] represents a third image reconstruction step by a confocal image reconstruction process;

[0080] [Fig.5D] represents a fourth image reconstruction step by a confocal image reconstruction process;

[0081] [Fig. 6] represents a 2D image extracted according to the invention; and

[0082] [Fig.7] represents the 2D image obtained from the image of [Fig.6] based on the SCR metric.

[0083] Such a method is described for example in the publication O'Loughlin, D., Elahi, MA, Lavoie, BR, Fear, EC and O'Halloran, M., 2021, Assessing Patient-Specific Microwave Breast Imaging in Clinical Case Studies, Sensors, 21(23), p.8048.

[0084] Referring to [Fig. 1], one can see that a microwave imaging device 1 according to the present invention has been schematically represented.

[0085] The device 1 includes a processing unit 2, comprising means for microprocessor-type computing, microcontroller-type computing, digital signal processor (DSP), processor-type processor, field-programmable gate array (FPGA) or application-specific integrated circuit (ASIC), associated with memory (of the type ROM, EEPROM, RAM, flash memory), as well as input / output ports and / or means for wired or wireless communication with the other elements of the device 1, to implement the steps described below for microwave brain imaging and to transmit signals to the other elements of the device 1 and receive signals from the other elements of the device 1.

[0086] The device 1 further comprises a switching network 3 connected to a headset 4 carrying a plurality of antennas 5, the antennas 5 being controlled by the switching network 3 which also collects the signals from the antennas 5, a vector network analyzer 6, connected to the switching network 3 and processing the signals from the switching network 3 to send them to the processing unit 2, and a human-machine interface 7, comprising a screen and keyboard and / or mouse input means.

[0087] The processing unit 2 and the human-machine interface 7 can, for example, be implemented using a conventional desktop computer or a tablet or even a smartphone and is configured to implement the steps described below.

[0088] Referring to [Fig.2], we can see that a helmet 4 equipped with its antennas 5 is shown there according to a preferred embodiment.

[0089] The helmet 4 is hemispherical in shape, with 24 antennas 5 in this embodiment, by way of example and not limitation, distributed uniformly over the surface of the helmet 4. The numbers on the antennas are used to reference the antennas 5 for ordering purposes. The number assigned to each antenna 5 in [Fig. 2] is, of course, illustrative and not limiting. The invention is not limited to the number of 24 antennas indicated in this example of an embodiment of the invention, which is also illustrative and not limiting.

[0090] The antennas 5 are preferably butterfly antenna type antennas, and advantageously have a bandwidth of about 3 GHz.

[0091] The processing unit 2 is configured in the microwave imaging device 1 to implement the following steps, schematically represented in [Fig.3], when helmet 4 is placed on the head of a patient:

[0092] a - to produce a global three-dimensional image of the brain by microwave imaging using all 5 antennas of the microwave antenna array 4;

[0093] b - detect at least one target in the global brain image obtained in step a;

[0094] c - calculate a position of at least one target in the global brain image produced in step a;

[0095] d - to produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array 4 controlled by the switching network 3 associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas 5;

[0096] e - merge the global image and the local processing image into a global three-dimensional intermediate image;

[0097] f - detect at least one target in the overall intermediate image;

[0098] g - calculate a position of at least one target in the overall intermediate image;

[0099] h - to produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array 4 controlled by the switching network 3 associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas 5;

[0100] i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image;

[0101] j - repeat steps f to i a predetermined number of times.

[0102] After step j, the image obtained is an image of the patient's brain on which the The target is positioned in space. An evolution of the target over time is also obtained, from step a to step j.

[0103] The SCR and SMR metrics described below are used to detect the target in steps b and f.

[0104] The objective is to increase the speed and accuracy of image reconstruction by reducing the complexity of the hardware to a minimum, provided that most of the information can be extracted in the software of the processing unit 2. In the present invention, a complete scan of the brain is performed and at each scan, only the necessary samples are selected in order to reduce the processing speed in the software of the processing unit 2.

[0105] A complete scan of the patient's brain is therefore carried out first.

[0106] A full-scan diffusion tensor is created in the processing unit 2, the tensor being a matrix representing the signals emitted and received by each antenna 5 of the headset 4.

[0107] By applying the adaptive connectome calibration process detailed below to the diffusion tensor, the antennas 5 required for scanning are determined according to information on the target area.

[0108] A reduction in the number of samples (compressive sampling) is applied to the result of the adaptive connectome calibration process applied to the diffusion tensor to obtain a sparse matrix from the diffusion tensor. An improved sparse matrix is ​​then created from the full scan.

[0109] Compressive sampling is advantageously implemented on the sparse matrix improved by a convex optimization method of the Ll space.

[0110] In order to generate sparse signals, suppose that the length of a signal x is Mxl. If the signal is sparse with the factor K (K ≤ M), it can be represented as x^s, where rp is a complete orthogonal dictionary matrix. The dictionary matrix is ​​used as a coefficient in the compressive detection method. Compressive detection, which is considered an inverse reconstruction of the original signal from sparse signals, uses the linear operation of the measure matrix (¢) and the input signal as follows: y = 0x. This theory is based on the assumption that the measure matrix <e>is inconsistent with the dictionary matrix ip. By this condition, the vector-matrix, s, can be reconstructed from G=O(K*log Nt). To solve this equation, we must solve the following convex optimization problem. [YES] [Math.l] min || 5 || n provided that y = ^y)S

[0112] In the present invention, a convex optimization based on the Ll norm is proposed to solve this equation. Furthermore, the matrix operations in compressed detection can be written as: y = 0ips = As, where 0 is a random Gaussian measure matrix and rp is the discrete cosine transform (DCT) matrix: A = 0ip. The compressive method is first applied to generate the sparse signals. Then, from these sparse signals, images are obtained by the confocal image reconstruction process to compare the new SCR. If the SCR is still high, this means that the sparse signal is correct. Otherwise, the sparsity factor must be decreased.

[0113] Optimal sparsity is applied such that fewer undamaged elements are used in the range where there is no signal change. In general, from a time perspective, the initial part of the signal is often omitted because it has values high levels due to reflection on the surface. However, useful information can be extracted from these signals for use in estimating certain parameters, such as the effective permittivity, etc.

[0114] After collecting the received signals, all the signals are reduced in size using sparse signals. Since most natural signals in one or more domains (time, frequency, wavelets, etc.) have a sparse display, this means that the information they contain can be used with a small number of coefficients in a special expressed domain. For example, although images appear very dense or spatially rich and with a lot of information, in the frequency domain they have compact and so-called thin information. Thus, most of their frequency coefficients are null or close to zero. The same is true for many other types of signals. Therefore, a very large signal of length N can only be represented by the coefficient K, which is K ≤ M; this signal is called a sparse order of order K.

[0115] From the hollow matrix, an image is created using the confocal image reconstruction process.

[0116] Meanwhile, in order to extract anatomical information, existing images related to the scan of the patient's head, such as the CT and / or MRI, are transmitted to device 1, preferably wirelessly.

[0117] These existing images are then processed and segmented by a machine learning machine to process geometric and target position information in the brain image.

[0118] The existing images are then merged with the image created by the confocal image reconstruction process and a complete image of the brain is created.

[0119] This scenario being intended for brain monitoring, it has a multi-scan mode allowing to reveal evolutions / changes of the target.

[0120] In radar imaging, after receiving the return signals using a multistatic scanning data collection method, a diffusion tensor is created. This step, which involves both hardware and software, converts the electromagnetic waves into complex data stored in a tensor called the diffusion tensor. In this tensor, which is a matrix, the sub-diagonal portion is eliminated first, based on the useful information, along with the diagonal elements, which correspond to the signals reflected by each antenna, due to their much higher values ​​compared to the other signals. Generally, before converting the signals into data and the data into information, calibration must be performed on the hardware to ensure the accuracy of the received signals.Calibration includes, in particular, eliminating coupling between antennas by adjusting their distance, and creating a plane wave instead of a spherical wave by adjusting the distance from the medium. measurement and elimination of unwanted reflections are achieved by adding an absorber or a metallic backplate to suppress scattering behind the antenna. Once all the necessary steps have been taken to transfer the correct data from the hardware to the software, the software is calibrated.

[0121] Due to the reciprocity property of electromagnetic theory, it is not necessary to scan the entire 24 antenna x 24 antenna state. In other words, the information stored from the 3D simulation model includes the diagonal elements and the upper triangular elements of the diffusion tensor. The diagonal elements are the antenna return losses. The other signals are the transmission losses between different antenna pairs.

[0122] The main objective of the software portion of the processing unit 2 is to reconstruct the accurate image of the received signals. First, in order to map the information, a coordinate hemisphere containing the focal points must be created. To create this hemisphere, physical information such as the ambient radius, the ambient material, and the number of sampling points must be entered into the program of the processing unit 2 using the human-machine interface 7. The ambient radius determines the image boundary. The number of points can also be determined based on the signal bandwidth. Another important parameter is the ambient material, which is defined based on its dielectric permittivity and electrical conductivity. The dielectric permittivity is more important due to changes in wave velocity.In the next step of entering parameters into the program, information from device 1 is entered such as the number of antennas 5, their location and the channels linked to the signals, which show the relationship between the signals and the path between two respective antennas 5 in each signal.

[0123] After inputting all the physical parameters, the first preprocessing section comprises the spurious signal suppression algorithm with in situ calibration. In brain radar imaging, since there are several unknown parameters for reconstructing the overall image, it must be extracted from the feedback signals. Thus, methods based on information theory can be very useful, both for increasing processing speed and for the device to detect certain targets. In this invention, we propose an adaptive form of the in situ calibration algorithm based on the dynamic arrangement of the connectome (the set of connections between the antennas and the signals) in the feedback processing path. In this method, the signals are selected in the diffusion tensor based on geometric information of the imaged medium (brain) and the wave propagation channels in the environment.To this end, we perform processing from different paths that lead to . extracting more information. In general, all information is extracted from the diffusion tensor in addition to the physical information of the system to reconstruct the image.

[0124] The in situ adaptive connectome calibration method uses the symmetrical property of the right and left sides of the elliptical structure of the brain.

[0125] All connections between the antennas are identified.

[0126] The connections are then grouped by distance between a given pair of antennas. Thus, a first group consists of connections between adjacent antennas, a second group consists of antennas separated by one antenna, a third group consists of antennas separated by two antennas, etc., the last group being made up of antennas separated by the largest possible number of antennas 5 on the headset 4. These groups represent signals that travel along symmetrical paths.

[0127] For the helmet 4 shown in [Fig.2], the last group would therefore consist of the antennas 5 separated by five antennas (for example the antennas 5 diametrically opposed on the helmet 4).

[0128] For each group of signals, the first signal is subtracted from the next, and the resulting signal is subtracted from the next until the last signal in the group. This yields a differential signal for the group of signals, removing characteristics unrelated to the target to be detected.

[0129] The differential signal obtained for a group is then subtracted from the average value of the signal for the group.

[0130] These two steps make it possible to eliminate extraneous signals, in particular background and skin effects, to retain only the useful information relating to the target.

[0131] Once this processing is completed, the processing unit 2 performs image reconstruction using a confocal image reconstruction process to obtain a three-dimensional image from the calibrated signals. Fundamentally, the confocal image reconstruction process consists of coherently integrating the energy of each signal reflected at each focal point.

[0132] A two-dimensional imaging domain of the human head, as shown in [Fig. 4], is considered. An antenna array is used where the antennas are placed at equal distances from each other around the head. The positions of each antenna, represented in spherical coordinates (in, r), correspond to Cartesian coordinates given by an = [xn, yn], where n is the number of the nth antenna. The imaging area inside the head is represented by I, where the imaging points (focal points) inside are designated by im = [xm, ym], where m is the number of the mth point in the imaging area.

[0133] In order to ensure consistent signal integration, the effects of the delays between the Focal points and antenna positions must be compensated. In this case, it is necessary to find the phase shift between each antenna and the other antennas. This delay time is equal to the direct distance between the transmitter and the receiver divided by the ground velocity in the propagation medium. The next step is to extract the target location from the reflected signals. For this, all values ​​are set to zero before their calculated delay. The propagation delay of the signal from the nth antenna of the array to the mth point of the imaging area, I, is calculated based on the following equation:

[0134] [Math.2]

[0135] where eeff is the effective dielectric constant of the head. For the model studied, the value of Eeff is calculated to be 38, which corresponds to the average of the dielectric constant values ​​measured on artificial brains. In order to reconstruct an image, a focused beamforming algorithm such as the Delay and Sum (DAS) beamforming algorithm is implemented. The first step is to identify the focal points in order to calculate the energy model of the signal reflected at these points. For the multi-static imaging algorithm, this is done by coherent integration of the signals. A Delay and Sum (DAS) beamforming algorithm creates a coherent integration of the signal energy at each focal point by summing the phase-corrected signals relative to each other, according to the following equation:

[0136] [Math.3] 1(im) = | (2(rn(Û))) |4

[0137] Where An is the signal from the antenna at location n (focal point of n).

[0138] A person skilled in the art knows how to convert from Cartesian coordinates to coordinates spherical.

[0139] Figures 5A-5D illustrate the confocal image reconstruction process. [Fig. 5A] shows the focal points inside the hemisphere with the size of the antenna distances.

[0140] The spatial accuracy between these points is approximately 3 mm. Next, by calculating the delay, these points are calculated from each pair of antennas corresponding to each signal, and the energy accumulation at each point is obtained. [Fig. 5B] shows a three-dimensional diagram of a confocal image with DMAS beamforming. [Fig. 5C] shows a 2D image in coronal view, and [Fig. 5D] finally shows a two-dimensional image containing the target.

[0141] A combined method based on machine learning has been proposed for the segmentation and classification of computed tomography (CT scan) / MRI images. The main steps of the proposed method are presented below.

[0142] A noise suppression method can be used to reduce the noise level in the images.

[0143] Furthermore, to reduce image dimensions, dimensionality reduction methods such as wavelet transform and principal component analysis (PCA) can be used. Finally, K-means methods, fuzzy image segmentation, and multiclass SVM (support vector machine) segmentation can be used to classify types of anomalies (strokes).

[0144] After creating the overall three-dimensional image based on microwave imaging using the patient's MRI / CT scan image, these two images are combined in the post-processing program in processing unit 2. After extracting the necessary information, a decision is generated for feedback to the switching network to determine the parameters of the next scan. As this is a feedback loop method, the image must be displayed in several time frames.

[0145] In order to achieve the ultra-fast processing technique, signal reduction and sampling reduction are applied at two levels.

[0146] The main question is which antennas should be selected to have the maximum information capacity. The elements of the diffusion tensor show all possible connections, and the goal is to select the necessary signals based on the connectome from the diffusion tensor. In the context of cognitive scanning, it is not necessary to input all the elements of the diffusion tensor into the image reconstruction algorithm. For example, among the different signals that pass through the same propagation channel, only one is selected, and the others are eliminated to obtain a sparse matrix. This means that the lower triangular elements of the sparse matrix must be removed. Furthermore, due to the quasi-elliptical symmetry of the human head, the return signals from facing antennas can be used for in situ calibration.It is important to note that the return signals from each port are significantly different from the diagonal transmission signals; namely, the bi-static diagonal transmission signals are much weaker than the return signals in each monostatic port. This phenomenon is of considerable importance because it governs key imaging measurements such as the correlation and contiguity of encoded information from the scene when different channels are scanned.

[0147] A quantitative way to analyze the information capacity (and thus the orthogonality of the spatiotemporal resolution) of the selected signals is to analyze the signal-to-noise ratio (SCR) and the signal-to-mean ratio (SMR) of the reconstructed images and the target location in each scan. According to the invention, an SCR map and an SMR map are used as a means of target detection in steps b and f. Then, depending on the maximum value of these criteria (SCR and SMR), the target position is revealed. After creating the image by the confocal image reconstruction process during the first complete scan, information on the target position and other parameters such as the SCR and SMR are extracted from the image of the current scan. Furthermore, the cross-correlation matrix of the diffusion tensor is to be compared to the SCR and SMR maps.All this information is then adapted to determine the target's position and the area it occupies. This information is presented in relation to the connectome as a decision factor for determining the optimal signal selection. Next, by comparing this position with the cross-correlation matrix, the signals that have the main effect on the target's position are selected by examining the connectome's connectivity. Compressive detection, implemented by processing unit 2, is then applied to these selected signals to generate sparse signals. In this case, the subsequent digitized image is acquired in the fastest possible mode with the fewest samples required, while preserving the necessary information about the detected target. Compressive sampling is advantageously implemented using a convex optimization method in L1 space.

[0148] After extracting information from the patient's brain scan images, at this stage, using the image obtained by the microwave imaging method, new useful information is extracted to match the location of the target with the information extracted from the scan images and provide the necessary information in the feedback phase to decide on the second scan.

[0149] Due to the nature of microwave images, which are obtained by accumulating energy at focal points and exhibit target-like points or blurred targets within the image, target position detection uses data-based metrics. Useful metrics derived from this perspective include the SCR and SMR. The region with the highest value of these two metrics is identified as the target region and compared to the target region obtained in the scan image. As long as this comparison yields the same result, selective signals are used for the next scan, instead of the entire diffusion tensor signal set.

[0150] Due to the nature of microwave images, which are obtained by accumulating energy in focal points that have target-like points in the image or For fuzzy targets, detecting their position using artificial intelligence methods exhibits a high error rate. Therefore, to detect the target and extract its position from the image, techniques based on quantitative metrics are proposed. The useful metrics derived from this perspective are SCR and SMR. For this purpose, the image region was divided into several 2D windows as shown in [Fig. 6]. The central window location with the highest value for these two metrics is identified as the target region.

[0151] The first parameter calculated within the scope of this invention is the SCR ratio, which is used to assess the extent to which the area energy is greater than the spurious signal energy in each 2D S window. Therefore, the SCR value quantifies the presence of an artifact at the target location in the brain. The SCR can be identified as follows:

[0152] [Math.4] VneS

[0153] where [F(n)]Stroke is the energy value in the S2Den window in the presence of a target, and [F(n)]Ciutter is the energy value in the same S region when the target is not present. The spurious signals are due to residual artifacts, and the average energy value of the spurious signals is calculated in a background model, which is based on a model of a healthy head without a stroke.

[0154] The second parameter calculated within the scope of this invention is the SMR, which allows for the evaluation of the extent to which the energy of the target area is greater than the average energy of the spurious signals in the head area. The SMR metric is specified as the ratio between the average value of the backscattered energy in the target area and the average value in the entire brain. The SMR can be identified as follows:

[0155] [Math.5] SMR=.............. mean V nE H

[0156] where [F(n)]Stroke is the intensity of all points inside window S and mean[F(n)]ciutter is the intensity of all points inside window H, where H represents the entire head region.

[0157] Fig. 7 shows the results of the generation of SCR metric maps from a 2D microwave image of Fig. 6, where the SCR value in ordinates varies from 0 to 5.

[0158] The ultrafast processing technique for a super-resolution microwave brain imaging system based on maximizing the extracted information capacity is now described. After locating the target, the signals related to the area The target's signals are selected. Depending on the target's position, relevant signals are selected within the target area to apply adaptive calibration. The adaptive connectome calibration method can be used for calibration. The hemispherical area of ​​the headset 4 can be divided into horizontal or vertical planes, each plane corresponding to a subset of the antenna array 5. The horizontal arrangement of signal selection can be applied based on the height of the target area. In other words, if the detected target is at the same level as one of the circular antenna arrangements, only the signals associated with that circular arrangement will be used to create the image in subsequent scans. Another arrangement in the vertical direction is possible, which, due to the limited number of antennas, can only be effective in assisting with improved calibration.

[0159] The device 1 of the invention therefore performs an initial scan of the patient's brain using the headset 4 equipped with antennas 5, by microwave imaging. A diffusion tensor is obtained, which is processed by the processing unit 2. This unit processes the image obtained using an SCR and SMR detection method to detect and position the target detected in the brain. Once the target is detected, a three-dimensional image is constructed using an adaptive connectome calibration process and compressive sampling. This process selects a subgroup of the antenna array 5 that is best suited to produce a new brain image limited to the target region.Once the image limited to the target is constructed, also using the adaptive connectome calibration and compressive sampling process, this limited image is merged with the previous image or even with a more precise image obtained by fMRI or other methods. These steps are then repeated to obtain a sequence of images that allows tracking the target's evolution without having to recalculate the entire image each time. The power of the device according to the invention lies in the fact that the target is detected in each global image by a computationally inexpensive algorithm. Image acquisition is then limited to the detected target region to obtain a global image resulting from the fusion of a more precise image with the image limited to the imaged area.< / e>

Claims

Demands

1. - Three-dimensional brain imaging method using hypersonic imaging frequency, characterized by the fact that it includes the following steps: a - to produce on a patient on whose head is placed a network of microwave antennas controlled by a switching network connected to a controllable signal transmitter-receiver device, a global three-dimensional image of the brain by microwave imaging using all the antennas of the microwave antenna network; b - detect at least one target in the overall brain image obtained in step a; c - calculate a position of at least one target in the global brain image produced in step a; d - produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas; e - merge the global image and the local processing image into a global three-dimensional intermediate image; f - detect at least one target in the overall intermediate image; g - calculate a position of at least one target in the overall intermediate image; h - produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas; i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image; j - repeat steps f to i a predetermined number of times, at least one target being detected by calculating the signal-to-noise ratio and the signal-to-mean ratio in two-dimensional planes of the three-dimensional image, at least one target being detected on the surfaces of each two-dimensional plane on which the values ​​of the signal-to-noise ratio and the signal-to-average ratio are simultaneously maximum.

2. - A method according to claim 1, characterized in that each step of three-dimensional image realization by microwave imaging includes the collection of diffusion parameter data representing microwaves scattered by the patient's brain in a diffusion tensor, the generation of differential diffusion parameter data by an adaptive connectome calibration method to suppress spurious signals, and the processing of the differential diffusion parameter data by a confocal image reconstruction method to obtain a three-dimensional image.

3. - Method according to claim 1, characterized in that the position of at least one target in the three-dimensional image is calculated as the position of the surfaces of each two-dimensional plane on which the values ​​of the signal-to-noise ratio and the signal-to-mean ratio are simultaneously maximum.

4. - A method according to any one of claims 1 to 3, characterized in that the subgroup of the microwave antenna array associated with the position of at least one target consists of the antennas having the shortest distance to the position of at least one target.

5. - A method according to any one of claims 1 to 4, characterized in that compressive sampling is applied to the microwave signals emitted by the antennas of the antenna array.

6. - A method according to any one of claims 1 to 5, characterized in that each step of merging two three-dimensional images consists of replacing, in the global image, the part corresponding to the local image, with the local image to obtain the merged image.

7. - A microwave brain imaging device (1), characterized in that it comprises a headset (4) equipped with a microwave antenna array (5), a switching network (3) controlling the microwave antenna array (5), a vector network analyzer (6) connected to the switching network (3), a processing unit (2) connected to the switching network (3) and the vector network analyzer (6) and controlling the switching network (3) and the vector network analyzer (6), a human-machine interface (7) connected to the processing unit (3), the processing unit (3) comprising

8. computing and memory resources to carry out the following steps: a - to produce a global three-dimensional image of the brain by microwave imaging using all the antennas (5) of the microwave antenna array; b - detect at least one target in the overall brain image obtained in step a; c - calculate a position of at least one target in the global brain image produced in step a; d - produce a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array (5) controlled by the switching network (3) associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas; e - merge the global image and the local processing image into a global three-dimensional intermediate image; f - detect at least one target in the overall intermediate image; g - calculate a position of at least one target in the overall intermediate image; h - produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array (5) controlled by the switching network (3) associated with the position of at least one target detected in step f and a reduction in the number of samples taken by the antennas; i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image; j - repeat steps f to i a predetermined number of times, at least one target being detected by calculating the signal-to-noise ratio and the signal-to-mean ratio in two-dimensional planes of the three-dimensional image, at least one target being detected on the surfaces of each two-dimensional plane on which the values ​​of the signal-to-noise ratio and the signal-to-mean ratio are simultaneously maximum. - Apparatus (1) according to claim 7, characterized in that each step of three-dimensional image acquisition by microwave imaging includes the collection of diffusion parameter data representing microwaves scattered by the patient's brain in a diffusion tensor, the generation of differential diffusion parameter data by an adaptive connectome calibration process to suppress spurious signals, and the processing of the differential diffusion parameter data by a confocal image reconstruction process to obtain a three-dimensional image.

9. - Apparatus (1) according to claim 8, characterized in that the position of at least one target in the three-dimensional image is calculated as the position of the surfaces of each two-dimensional plane on which the values ​​of the signal-to-noise ratio and the signal-to-mean ratio are simultaneously maximum.

10. - Apparatus (1) according to any one of claims 7 to 9, characterized in that the subgroup of the microwave antenna array (5) associated with the position of at least one target consists of the antennas (5) having the shortest distance to the position of at least one target.

11. - Apparatus (1) according to any one of claims 7 to 10, characterized in that compressive sampling is applied to the microwave signals emitted by the antennas (5) of the antenna array.

12. - Apparatus (1) according to any one of claims 7 to 11, characterized in that each step of merging two three-dimensional images consists of replacing, in the global image, the part corresponding to the local image, with the local image to obtain the merged image.

13. - Microwave brain imaging apparatus (1) according to any one of claims 7 to 12, characterized in that the antenna array (5) comprises 24 antennas, preferably butterfly antenna type antennas.

14. - Product computer program comprising instructions which, when executed by a microwave imaging device according to any one of claims 7 to 13, perform the following steps: a - acquire a global three-dimensional image of the brain by microwave imaging using all the antennas of the microwave antenna array; b - detect at least one target in the global brain image acquired in step a; c - calculate a position of at least one target in the global brain image acquired in step a; d - acquire a local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step a and a reduction in the number of samples taken by the antennas; e - merge the global image and the local processing image into a global three-dimensional intermediate image; f - detect at least one target in the overall intermediate image; g - calculate a position of at least one target in the overall intermediate image; h - produce a new local three-dimensional processing image of the target by microwave imaging using at least one of a subgroup of the microwave antenna array controlled by the switching network associated with the position of at least one target detected in step f and a reduction in the number of samples produced by the antennas; i - merge the global intermediate image and the new local processing image into a new global three-dimensional image also called the intermediate image; j - repeat steps f to i a predetermined number of times.

Images