METHOD FOR THE DETECTION OF STRUCTURAL ANOMALIES ON A MEDICAL IMAGE AND ASSOCIATED SYSTEM

Abstract

METHOD FOR DETECTING STRUCTURAL ANOMALIES ON A MEDICAL IMAGE AND ASSOCIATED SYSTEM The invention relates to a computer-implemented method (1000) for detecting a structural anomaly of a subject's organ from a functional image, comprising the following steps: receiving (100) an image acquired (11) by a non-irradiating image acquisition device; converting (200) said acquired image (11) into at least one synthetic anatomical image (12) by a conversion module; detecting (300) one or more structural anomalies of a subject's organ on the synthetic anatomical image (12) by a detection module. Figure for the abstract: Fig. 1

Description

Title of the invention: METHOD FOR THE DETECTION OF STRUCTURAL ANOMALIES ON A MEDICAL IMAGE AND ASSOCIATED SYSTEM Scope of the invention

[0001] The invention relates to a method for detecting a structural anomaly on a functional image of an organ of a subject.

[0002] The invention also relates to a system, a computer program product, and a data carrier for implementing said method according to the invention. State of the art

[0003] Chronic airway diseases represent a major public health problem. Among them, chronic obstructive pulmonary disease (COPD), asthma, and cystic fibrosis, one of the most common genetic diseases in Caucasians, cause progressive lung damage through chronic inflammation and infections. Computed tomography (CT scans) has long been the gold standard for monitoring structural changes in the lungs, but it poses problems of cumulative exposure to ionizing radiation, particularly for pediatric patients.

[0004] Ultrashort echo time MRI (UTE-MRI) sequences offer a promising, radiation-free alternative for lung imaging. However, UTE-MRI presents specific challenges, including a lower signal-to-noise ratio and reduced spatial resolution compared to CT.

[0005] To date, there is no automated solution for segmenting and quantifying the volume of structural airway lesions such as bronchiectasis, bronchial wall thickening, bronchial mucus, bronchiolar mucus, or foci of consolidation / atelectasis on UTE-MRI due to these technical limitations. These volumetric segmentations can be used to derive a severity score for structural airway damage.

[0006] The invention therefore aims to improve existing methods of segmenting structural abnormalities of an organ with sufficient accuracy while reducing radiation exposure, in order to quantify a holistic severity score of structural alterations of the airways of patients with chronic airway disease. Summary of the invention

[0007] According to a first aspect, the invention relates to a computer-implemented method for detecting a structural anomaly of an organ of a subject from of a functional image. The method includes a step of receiving an image acquired by a non-irradiating image acquisition device. The method further includes a step of converting said acquired image into at least one synthetic anatomical image by a conversion module. The method further includes a step of detecting one or more structural anomalies of an organ of a subject on at least one synthetic anatomical image by a detection module.

[0008] One benefit of such detection is to allow the calculation of a holistic segmentation score of one or more structural anomalies.

[0009] In one execution mode, the conversion step is implemented by a first trained learning function configured to receive the acquired image as input and generate the synthetic anatomical image as output.

[0010] In one execution mode, the detection step is implemented by a second trained learning function configured to receive as input the synthetic anatomical image and generate as output at least one presence score of at least one predetermined structural anomaly.

[0011] In one execution mode, the training of the first learning function is implemented using data produced by a loss function implemented during the execution of the second trained learning function.

[0012] In one execution mode, the learning function is trained using a loss function that incorporates a learning reinforcement agent over deep supervision

[0013] In one execution mode, the data produced includes segmentation information extracted from the loss function of the second trained learning function.

[0014] In one execution mode, the first learning function is trained using a discriminator which acts on a latent space of said first learning function to evaluate whether the features generated by the first learning function respect predetermined properties extracted from the second trained learning function.

[0015] In one execution mode, the first learning function is trained using a loss function which incorporates an adversarial or corrective term.

[0016] In one execution mode, the adversarial or corrective term guides the latent space of the first learning function to produce representations aligned with target information extracted from the second trained learning function.

[0017] In one embodiment, the method includes the generation of a presence score by the second trained learning function for each anomaly present on the synthetic anatomical image, said score being a function of the integral of the number of voxels and / or pixels classified for each predetermined anomaly.

[0018] In one embodiment, the generated attendance score is a holistic attendance score.

[0019] In one embodiment, the method includes extracting a bronchial volume and / or a vascular volume from the synthetic anatomical image.

[0020] In one embodiment, the method includes a step of normalizing each presence score generated by the extracted bronchial and / or vascular volume to generate, for each predetermined anomaly, a normalized presence score.

[0021] In one execution mode, the conversion module includes a neural network comprising: • an encoder configured to compress the acquired image into a reduced representation in a latent space, • a decoder configured to generate a synthetic anatomical image from the reduced representation in latent space;

[0022] the method further comprising: • training a learning function to classify at least two types of anomalies on a synthetic anatomical image of a patient's organ, • the extraction of specific knowledge from said trained learning function, • the integration of said specific knowledge extracted within the latent space of the neural network of the conversion module to orient the conversion of the functional image into synthetic anatomical images into an image, said orientation allowing the reconstruction of the scan image to be focused on the regions presenting anomalies.

[0023] In one execution mode, the detection module includes a learning function trained to receive as input at least one portion of a synthetic anatomical image and to generate as output a classification of each elementary part of said portion of the synthetic anatomical image.

[0024] In one execution mode, the first learning function is implemented by a neural network such as an nn-Unet network.

[0025] In one execution mode, the second learning function is implemented by a neural network such as an nn-Unet network.

[0026] In one execution mode, the acquired image is an image of a subject's lung. In one execution mode, the structural abnormality predetermined by the detection module includes at least one of the following abnormalities: bronchiectasis, bronchial wall thickening, bronchial mucus accumulation, bronchiolar mucus accumulation, and / or condensation / atelectasis.

[0027] According to another aspect, the invention also relates to a structural anomaly detection system comprising software and / or hardware means for implementing the method according to the invention.

[0028] According to another aspect, the invention relates to a computer program product comprising code instructions which, when implemented by a computer, lead the system according to the invention to carry out the steps of the process according to the invention.

[0029] According to another aspect, the invention relates to a computer-readable data carrier on which the computer program product according to the invention is recorded. Brief description of the figures

[0030] Other features and advantages of the invention will become apparent from the following detailed description, with reference to the accompanying figures, which illustrate:

[0031] [Fig-1]: a flowchart representing the steps of a method according to a mode of realization of the invention.

[0032] [Fig.2]: a logic diagram representing the steps of a method for training the first and second learning function according to an execution mode of the invention.

[0033] [Fig.3]: a schematic representation of a system according to an embodiment of the invention. Description of the invention

[0034] The present invention relates to a computer-implemented method 1000 for the automatic detection of a structural anomaly of an organ from a functional image of said organ of a subject. The invention also relates to an associated device 1 for implementing said method.

[0035] A structural anomaly should be understood as a structural change in an organ compared to a healthy state and caused by anatomical alterations of the organ that may be caused by disease, injury, or the aggravation of an injury. In one embodiment, the structural anomalies affect the morphology and function of the airways in a lasting and / or significant manner.

[0036] In one embodiment, a structural anomaly refers to anatomical alterations of the lungs caused by cystic fibrosis, such as bronchiectasis (abnormal dilations of the bronchi), bronchial wall thickening (increased thickness of the walls of the airways due to inflammation or chronic infections) or bronchial mucus accumulation (obstructions of the airways caused by excessive mucus production).

[0037] These changes are key markers of disease progression and are studied through imaging techniques to assess the condition of the lungs.

[0038] An example of a method according to one embodiment of the invention is now described with reference to [Fig.1]. Image acquisition

[0039] Method 1000 comprises the reception 100 of a medical image 1 of at least one organ of a subject. The received medical image 11 preferably comprises an image acquired by a non-ionizing imaging device.

[0040] In the context of the present invention, the term "image acquired by a non-ionizing imaging device" specifically designates a visual representation, in the form of digital or analog data, obtained from a medical imaging modality that does not involve the use of ionizing radiation to generate said image.

[0041] This term encompasses, but is not limited to, images produced by devices using technologies based on magnetic fields, radio waves, ultrasound, or photons in the visible or infrared spectrum. By way of non-exhaustive examples, this includes images obtained by magnetic resonance imaging (MRI), ultrasound, optical coherence tomography (OCT), or any other equivalent method guaranteeing the absence of ionizing radiation.

[0042] The images 11 thus generated allow anatomical, functional or metabolic visualization of the tissues and organs of the subject without risk associated with cumulative exposure to radiation, which makes them particularly suitable for vulnerable populations or those requiring frequent examinations.

[0043] In one aspect of the invention, a method includes a receiving step 100, by a data processing device, of an image obtained by non-ionizing imaging such as magnetic resonance imaging (MRI) and its storage in memory.

[0044] In one execution mode, a data acquisition module receives an MRI image file as a data stream or as a digital file generated by an MRI scanner (e.g., an anatomical or functional MRI sequence). The data can be received via a wired connection (e.g., Ethernet, USB) or wirelessly (e.g., Wi-Fi, Bluetooth) from the imaging device or an intermediate server. The received image is then stored in a dedicated memory space, comprising volatile memory (such as RAM) for immediate processing and non-volatile memory (hard drive, SSD, or secure cloud) for long-term archiving.

[0045] In one execution mode, the metadata associated with the MRI image (such as acquisition parameters, resolution, or patient identifier) ​​is checked to ensure data compatibility and integrity.

[0046] The image file may include a single two-dimensional MRI image or a three-dimensional MRI image.

[0047] In the context of the present invention, the term "acquired image" refers to a visual representation obtained from a medical imaging modality, such as magnetic resonance imaging (MRI), and may include two-dimensional ("2D image") or three-dimensional ("3D image") images.

[0048] A 2D image corresponds to a planar projection representing a single cross-section or section of an anatomical volume of the subject, each point of the image being associated with an intensity value indicating the physical or chemical properties of the corresponding region.

[0049] A 3D image, on the other hand, corresponds to a volumetric set consisting of a series of successive 2D images or a reconstructed volume, each voxel (volumetric element) representing a specific spatial unit and containing information relating to the characteristics of the tissue or organ in the volume concerned.

[0050] In one embodiment, the acquired images 11 are images obtained using an ultrashort echo magnetic resonance imaging (UTE-MRI) sequence. These images are particularly well-suited for capturing anatomical structures exhibiting low proton signal, such as the lungs, due to their ability to reduce motion artifacts and enhance the contrast of low-density soft tissues.

[0051] In one embodiment, the acquired images 11 are images obtained by respiratory-synchronized magnetic resonance imaging (MRI) or UTE-MRI. This technique allows for the capture of anatomical images at specific phases of the respiratory cycle, thereby reducing motion artifacts associated with respiration. The images thus acquired provide an accurate representation of thoracic structures, such as the lungs, airways, and surrounding tissues, under consistent and reproducible conditions. This embodiment is particularly well-suited for applications where respiratory variability could compromise the analysis, such as the evaluation of pulmonary abnormalities, the study of airway dynamics, or the mapping of soft tissue movements during respiration. Another benefit is increased spatial and contrast resolution and reduced motion artifacts.

[0052] Conversion of the acquired image into a synthetic image

[0053] In one execution mode, the acquired image 11 is converted 200 into a synthetic image 12.

[0054] In the context of the present invention, the generated synthetic image 12 specifically designates a visual representation obtained from medical data, allowing the description of the physical morphology, shape and organization of the tissues, organs, or internal structures of the human or animal body. These images are intended to capture static and structural features, providing a precise map of anatomical elements such as bones, muscles, organs, blood vessels, or respiratory tracts. In one embodiment, the generated synthetic image is a synthetic image of irradiating and / or ionizing imaging.

[0055] One benefit of this step is to convert the acquired image into an image type on which it is easier to automatically detect one or more structural abnormalities of the organ. Preferably, the acquired image is converted into a synthetic 2D or 3D CT scan image, depending on whether the acquired image was 2D or 3D. The remainder of the description will describe the converted synthetic image as a CT scan image, but other types of anatomical imaging can also be considered.

[0056] In one execution mode, this step is carried out by a first learning function 26.

[0057] The first trained learning function 26 can include any type of neural network. The first learning function 26 is configured to receive as input a medical image 11 as described above and generate as output a synthetic anatomical image 12.

[0058] In one embodiment, the first learning function 26 is implemented by an nnU-Net type neural network.

[0059] In one aspect of the invention, a convolutional neural network model, based on an nnU-Net architecture, is configured to convert an image 11 obtained by magnetic resonance imaging into a synthetic anatomical image 12 such as an image corresponding to a computed tomography (CT) scan acquisition. This nnU-Net model is optimized to receive as input an MRI image, including information on the magnetic properties and contrasts of the tissues, and to produce as output a synthetic image simulating the intensities in Hounsfield units characteristic of a CT scan, while preserving anatomical details.

[0060] The nnU-Net model is trained on a database of matched MRI and CT scan images, where each pair of images is aligned to ensure precise spatial correspondence. In one embodiment, the training method includes registration between a computed tomography (CT) image and a magnetic resonance imaging (MRI) image, said images being matched to represent the same anatomical structures. This step aims to spatially align the two imaging modalities to allow for combined analysis or fusion of anatomical and functional information.

[0061] The nnU-Net model architecture includes an encoder, which extracts the relevant features from the MRI images, a bottleneck for modeling the complex relationships between the MRI and synthetic image properties, and a decoder configured to reconstruct a synthetic image from the extracted features.

[0062] In a convolutional neural network architecture, the bottleneck constitutes a layer or set of layers located between the encoder and the decoder. This region plays a key role in the transformation and compression of features extracted by the encoder, making it possible to represent essential information in a reduced space while eliminating redundancies.

[0063] The training 110 of the first learning function 26 is an iterative process aimed at adjusting the weights and connections between neurons to minimize a loss function and improve the model's performance on a given task. Training an artificial neural network allows for the optimization of weighted connections between the network layers as a function of a loss function.

[0064] The training 110 of the first learning function includes, in particular, receiving as input to the network a set of data 21, 22 (for example, the acquired image or data from said acquired image) which are propagated through the layers of the network. At the output of the network, a predictive result is generated.

[0065] The training includes calculating a loss function to measure the difference between the predicted output and the CT scan image matched to the acquired image received as input to the network. The loss function may include, but is not limited to, the mean squared error (MSE), cross-entropy, and / or Dice Loss. A learning agent is then implemented to adjust the POIs by minimizing the loss function while preserving the network's ability to generalize to new data. In one embodiment, the loss function comprises a combination of both Dice Loss and cross-entropy. One benefit of this combination is to advantageously leverage the advantages of both loss functions.

[0066] Within the framework of the present invention, Dice Loss is a loss function used to evaluate and optimize the correspondence between segments predicted by a neural network and a ground truth in image segmentation tasks. Detection of the presence of an anomaly

[0067] In an execution mode, a detection step 300 is generated by a detection module.

[0068] The detection module is configured to receive as input a synthetic anatomical image 12 and to generate as output at least one presence score 14A, 14B, 14C of at least one predetermined structural anomaly of an organ of the subject.

[0069] In one embodiment, the attendance score 14A, 14B, 14C is a holistic attendance score.

[0070] A "holistic" score refers to an approach that comprehensively considers all available information without simplification or excessive hierarchical prioritization. Unlike methods that simplify or summarize data (for example, by focusing solely on a dominant anomaly or specific sections), a holistic approach aims to analyze all voxels or pixels of an image or structure to produce a complete and representative result.

[0071] A holistic presence score implies that each anomaly is evaluated over the entire analyzed volume, taking into account all anomalies present in the image without excluding any.

[0072] In one execution mode, the score is calculated from the integral of all voxels or pixels classified as belonging to a given anomaly over the entire volume studied (for example, the entire lung in the case of bronchi).

[0073] In one execution mode, the holistic score takes into account all anomalies simultaneously, whether major or minor, unlike hierarchical scores, where a dominant anomaly can "mask" the others.

[0074] In one embodiment, the holistic score is based on an exact quantification of the volume occupied by each anomaly, rather than on visual simplifications or approximate averages.

[0075] One benefit of such a holistic score is that it avoids biases associated with common simplifications (for example, considering only the most severe abnormalities or analyzing only a few image slices). It thus offers a robust and comprehensive quantification of the abnormalities present, improving the accuracy and reproducibility of diagnoses.

[0076] In one execution mode, this step of calculating an attendance score is carried out by a second trained learning function 25.

[0077] The second trained learning function 25 is configured to receive as input anatomical images or synthetic anatomical images of an organ of a subject such as CT-scan images or synthetic CT-scan images and to generate as output a score of presence of at least one predetermined structural anomaly on an organ of a subject.

[0078] In a first embodiment, the second learning function 25 comprises a first sub-function configured to generate a segmentation of the anatomical image and / or the synthetic anatomical image and a second sub-function configured to generate, from the generated segmentation, a presence score for each predetermined anatomical anomaly.

[0079] In a second alternative embodiment, the second learning function 25 comprises a plurality of second sub-functions, each configured to generate, from the segmentation generated by the first sub-function, a score for the presence of a single predetermined anatomical anomaly different from the other second sub-functions.

[0080] In one embodiment, the second learning function 25 is configured to generate directly from each voxel of the anatomical image or synthetic anatomical image, a probability of presence of each predetermined anatomical anomaly.

[0081] The presence score 14A, 14B, 14C generated of an anomaly may include the integral of the probabilities of each voxel of the anatomical image and / or synthetic anatomical image belonging to said anomaly.

[0082] In one embodiment, the second learning function 25 comprises a neural network, preferably of the nnU-Net type.

[0083] In one embodiment, the second trained learning function 25 is configured to produce at least one series of presence scores 14A, 14B, 14C, each presence score being representative of the presence on the acquired image 11 or the synthetic anatomical image 12 of a predetermined structural anomaly different from the other presence scores.

[0084] In an embodiment illustrated in [Fig.2], the training 110 of the second learning function 25 is carried out by a database of acquired medical images (such as MRI images) 21 and / or synthetic anatomical images matched to a series of presence scores 22 for each predetermined structural anomaly.

[0085] In one embodiment, training 110 includes a preprocessing step of anatomical and / or synthetic anatomical images comprising the extraction of the organ envelope and training. In one embodiment, training is performed using pairs of CT scan images simultaneously with and without the extraction of said envelope.

[0086] In one embodiment, the second learning function 25 is configured to generate as output the total volume 15 of the organ such as a lung volume or a vascular volume.

[0087] The training 110 of the second learning function 25 may include, in particular, receiving as input to the network a set of data (for example, the anatomical image with or without the organ's envelope) which is propagated through the layers of the network. At the output of the network, a predictive result is generated. The training of the second learning function includes calculating a loss function to measure the difference between the predicted output and the presence scores of each anomaly and / or the organ volume received as input to the network. In a mode In terms of implementation, the loss function can include, but is not limited to, the mean squared error (MSE) and / or the Dice Loss. A learning agent is then implemented to adjust the pois by minimizing the loss function while preserving the network's ability to generalize to new data.

[0088] In one embodiment of training the second learning function, the Dice loss can be optimized to retrain specifically on the proportion of training data having the worst similarity performance between the ground truth and the prediction issued by the second learning function.

[0089] In another embodiment of the training of the second function A learning reinforcement (LR) agent is proposed for deep supervision and uses a one-layer long short-term memory (LSTM) network followed by a fully connected layer to dynamically optimize the deep supervision weights. The LSTM input is a sequence containing losses and Dice scores for all N supervision levels. The LR agent integrates into the learning function training process as follows: the learning function is trained during one iteration, and then the losses and Dice scores are calculated and fed into the LR agent. The agent generates weights w, which are applied to the losses in deep supervision. A reward is calculated based on the improvement in losses and Dice scores. The LR agent is updated using the loss function.This formulation encourages the agent to adjust the weights to maximize the reward, resulting in an improvement in both loss and Dice scores at all levels of supervision.

[0090] This deep supervision enhancement agent uses a function of reward which can be formulated as follows: reward = ^jL prev jL^^ currj " D1C Gprev^

[0091] In which: - N is the number of deep supervision levels - Lprev is the loss value calculated during the previous iteration for each deep supervision level (i). This value represents the model error before the supervision level weights are adjusted by the reinforcement agent. - Lcuit j is the current value of the loss calculated after the dynamic adjustment of the supervision level weights by the agent. This value reflects the new error after applying the changes made by the agent. - Diceprevest is the value of the Dice score calculated in the previous iteration for each level of deep supervision (i). This value represents the model performance in terms of overlap between prediction and ground truth before the agent adjusts the weights of the supervision levels. - Dicecuir is the current value of the Dice score, calculated after the reinforcement agent has dynamically adjusted the weights of the deep supervision levels. This value reflects the new performance of the model after the agent's intervention.

[0092] The deep supervision reinforcement agent is updated with the following loss function (Lagent) W^). reward

[0093] Where w is the importance weight assigned by the learning agent to each level (i) of deep supervision during training.

[0094] This formulation encourages the agent to adjust the weights to maximize the reward, which represents an improvement in losses and Dice scores at all levels of supervision.

[0095] The example described below more precisely describes an embodiment of the invention where the organ comprises the lungs of a subject and includes: processing the CT scan image to extract the lung envelope, receiving said processed image by the second trained learning function, and automatically generating by the trained learning function a presence score on at least three labels: bronchiectasis, peribronchial thickening, the presence of bronchial or bronchiolar mucus, consolidations / atelectasis, and lung volume. Preferably, the method includes a step of normalizing each presence score by the total lung volume. This normalization makes it beneficial to compare values ​​between different subjects with different lung volumes.

[0096] In one embodiment, the method further includes the generation of a bronchial tree and / or a vascular tree of the organ from synthetic anatomical images.

[0097] In one embodiment, the method includes generating a graphical representation allowing for a detailed and usable visualization of structural anomalies on the generated bronchial tree and / or vascular tree. In one execution mode, said graphical representation is stored in memory and / or displayed on an AFF display.

[0098] From the generated synthetic anatomical image 12 (for example, a synthetic CT-scan image from the previous analysis), a bronchial volume and / or a vascular volume 15 is extracted. These volumes 15 are determined by dedicated segmentation algorithms, based on anatomical characteristics specific to bronchial and vascular structures, such as their density, shape, or their spatial distribution. The precise extraction of these volumes allows for the beneficial contextualization of abnormality scores and the normalization of their interpretation. Bronchial volumes can, for example, represent the analyzed airway region, while vascular volumes provide a reference for adjacent areas or correlated abnormalities.

[0099] In one embodiment, the anomalies include airway anomalies. In one embodiment, the structural anomalies include the following anomalies: • Bronchiectasis, • Thickening of the bronchial walls, and / or • An accumulation of bronchial mucus, and / or • An accumulation of bronchiolar mucus, and / or • Condensation, and / or • Atelectasis.

[0100] In one embodiment, the second learner function 25 is configured to generate, for each or part of the anomalies listed above, an attendance score of 14A, 14B, 14C or a normalized attendance score of 16A, 16B, 16C.

[0101] In an alternative embodiment illustrated in [Fig. 1], for each presence score 14A, 14B, 14C generated by the second learning function 25, a normalization step 400 is performed by dividing the presence score 14A, 14B, 14C by the extracted bronchial and / or vascular volume 15. This normalization 400 makes it possible to generate normalized presence scores 16A, 16B, 16C that are comparable between different patients or images, regardless of inter-individual anatomical or physiological variations. Thus, this step ensures that the detected abnormalities are assessed proportionally to the size of the surrounding structures, providing a relative and clinically relevant measurement.

[0102] Provisioning of predefined knowledge to the latent space

[0103] A particular aspect of the invention is to allow the use of specific knowledge of the second trained learning function 25 to improve the training of the first learning function 26.

[0104] In particular, one aspect of the invention aims to execute a training step 120 the first learning function 26 using data 30 produced by the loss function implemented during the execution of the second learning function 25.

[0105] One benefit of this feature is to enable the training of the first learning function 26 in such a way as to orient it to generate synthetic anatomical images 12 allowing efficient identification of predetermined structural anomalies.

[0106] In an execution mode, additional information 30 is provided to the latent space of an nnU-Net in order to guide its loss function; several approaches can be used. These methods consist of integrating additional information 30 or constraints into the model training, either by modifying the architecture or by adding regularizations in the optimization.

[0107] In one embodiment, this additional information 30 constitutes specific knowledge enabling supervised training to improve the generation of synthetic anatomical images 12 towards images enabling the identification of anomalies.

[0108] In one execution mode, the first learning function 26 is implemented using data produced by the loss function implemented during the execution of the second learning function 25. In one embodiment, said produced data includes segmentation information extracted from the loss function of the second learning function.

[0109] For example, the loss function of the first learning function can be oriented so as to guide the latent space of the neural network implementing the first learning function to produce representations of anomalies aligned with the representation of said anomalies on CT-scan images.

[0110] In one embodiment, the various coefficients of the loss function implemented to execute the first learning function are fixed or generated based on data extracted from the loss function implemented by the second learning function. These coefficients may include the coefficients of the mathematical formula of the loss function, particularly when the latter includes a combination of both dice loss and cross-entropy loss. One benefit is to enable the generation of synthetic anatomical images whose performance has been optimized to enhance the performance of presence scores by the second learning function.

[0111] In a particular embodiment, the method includes a training step 120 of the first learning function 26 using a discriminator 30. This discriminator operates in the latent space generated by said first learning function 26, evaluating whether the features extracted by the latter respect predetermined properties, these properties being derived from the second learning function 25 previously trained.

[0112] More specifically, the discriminator is configured to receive as input the latent representations generated by the first learning function and to compare these representations to a target distribution or to criteria extracted from the latent features of the second learning function. These predetermined properties may include, but are not limited to, spatial structures and contextual relations.

[0113]

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123] or statistical distributions related to the learning task. The discriminator thus acts as a regularization mechanism to guide the learning of the first learning function towards an optimization consistent with these properties. During training, an adversarial loss function is used to improve the quality of the generated latent representations. The loss includes a term corresponding to the error calculated by the discriminator, which evaluates whether the generated features conform to the target properties. This Ladversariai loss function can be formulated as follows: Adversarial Ez ~ first learning function [lOgD(z)] 4“ Ez ~ second learning function [log( 1 D[z))] Where D(z) is the output of the discriminator applied to the latent space z. Where E is the expected value of the evaluated values ​​for the latent representations (z) generated by the first or second learning function The first learning function is then optimized to maximize the similarity between its latent representations and those of the second learning function, while the discriminator is simultaneously trained to distinguish the representations of the two sources. This mechanism promotes convergence where the features generated by the first learning function capture properties relevant to the target task, while respecting the constraints imposed by the second learning function. This process ensures greater consistency of latent representations and improved overall performance of the first learning function in its respective application. Within the scope of the invention, the concept of structural or anatomical image may also include three-dimensional or multi-dimensional representations obtained by processing raw imaging data, allowing detailed and usable visualization of internal structures, often used as a reference for the analysis of functional or combined images. System According to one aspect, system 1 according to the invention includes software and hardware means for implementing the process as described above. An embodiment of system 1 according to the invention is now described with reference to [Fig.3]. The system includes a REC receiver. In one embodiment, the REC receiver is intended to be connected to a medical imaging acquisition device such as an MRI device, so as to receive images acquired by said acquisition device.

[0124] The REC receiver can be connected to the acquisition device by a wired or wireless connection, for example by a Bluetooth connection or a WLFI connection or any other data exchange protocol known to those skilled in the art.

[0125] The REC receiver may include or be associated with one or more memory units for temporarily storing received images. The REC receiver is directly or indirectly connected to the AFF display to transmit the acquired images to the AFF display.

[0126] The REC receiver may include an input processor configured to preprocess the received images, including steps such as intensity normalization, artifact correction, and resampling to ensure compatibility with learning functions.

[0127] System 1 further includes a CALC calculation device.

[0128] The CALC calculation device includes software and / or hardware means to implement the conversion 200 and detection 300 steps of the invention described above.

[0129] In one embodiment, the CALC calculation device preferably comprises the conversion module M1 for implementing the conversion step 200 of the acquired image 11 into a synthetic anatomical image 12 and a detection module M2 for detecting a predetermined structural anatomy and / or calculating a presence score for at least one predetermined structural anatomy. The two modules M1 and M2 may be independent or integrated into a single module.

[0130] According to an alternative, the learning functions 25, 26 are implemented by remote electronic equipment, such as a remote server. In this case, the computing device includes an interface for exchanging data with the remote equipment in order to transmit data and retrieve the result of the processed data, for example the generated attendance score(s).

[0131] Finally, in the present invention, it is understood that when the learning functions 25, 26 are implemented wholly or partly by remote equipment, the CALC computing device can be interpreted as the system comprising on the one hand the local device described in this application and the remote means enabling the learning functions to be implemented.

[0132] The CALC computing device further comprises at least one processor or computer associated with a memory module (MEM) to perform at least some of the steps of the method according to the invention. For example, a first processor can be configured to perform the steps of conversion, detection, and display generation. In one embodiment, the at least one processor or computer comprises means for transmitting and receiving information with a remote device, via example via an internet network, allowing the steps of the method according to the invention to be implemented.

[0133] In some cases, the processor or computer can communicate with one or more external devices via the network. The processor or computer can be connected to the network via a wired connection (e.g., via an Ethernet cable) and / or a wireless connection (e.g., via a Wi-Fi network). These external devices can include servers, workstations, and / or databases. The processor or computer can communicate with these devices to, for example, offload computationally intensive tasks. For example, the processor or computer can send a medical image acquired via the network to the server for analysis and receive the results of the analysis from the server. In addition (or alternatively), the processor or computer can communicate with these devices to access information that is not available locally and / or update a central information repository.

[0134] Device 1 may also include a plurality of processors, each associated with one or more memories, and configured to perform such steps together. In one embodiment, the processor(s) may be remote and connected to the display via a data network.

[0135] The device 1 further comprises one or more memories for storing or recording the sequences generated by the method according to the invention and / or for storing the computer programs which, when executed by one or more processors, implement the method according to the invention. In one embodiment, the device further comprises an EMM transmitter connected to said MEM memory for transmitting data from the second learning function, such as attendance scores or normalized attendance scores, from said MEM memory to a data network.

[0136] Device 1 may further include communication means such as EMM transmitters and REC receivers for exchanging information with an ACQ acquisition device and / or a remote device.

[0137] The AFF display may include means for receiving the different information received by the different means REC, CALC, MEM, of device 1 to generate a final image to be displayed.

Claims

Demands

1. A computer-implemented method (1000) for detecting a structural anomaly of a subject's organ from a functional image comprising the following steps: • receiving (100) an image acquired (11) by a non-irradiating image acquisition (QAC) device; • converting (200) said acquired image (11) into at least one synthetic anatomical image (12) by a conversion module; • detecting (300) one or more structural anomalies of a subject's organ on the synthetic anatomical image (12) by a detection module.

2. A method according to claim 1, wherein: • the conversion step (200) is implemented by a first trained learning function (26) configured to receive as input the acquired image (11) and generate as output the synthetic anatomical image (12) and • the detection step (300) is implemented by a second trained learning function (25) configured to receive as input the synthetic anatomical image (12) and generate as output at least one presence score (14A, 14B, 14C) of at least one predetermined structural anomaly; wherein the training of the first learning function (120) is implemented using data produced (22) by a loss function implemented during the execution of the second trained learning function (25).

3. Method according to claim 2 wherein the data produced (22) include segmentation information extracted from the loss function of the second trained learning function (25).

4. A method according to claim 2, wherein the first learning function (26) is trained using a discriminator which acts on a latent space of said first learning function to evaluate whether the features generated by the first learning function (26) comply with predetermined properties extracted from the second trained learning function (25).

5. A method according to claim 2 or claim 3, wherein the first learning function is trained using a loss function that incorporates a learning reinforcement agent over deep supervision.

6. A method according to claim 2 or claim 3, wherein the first learning function (26) is trained using a loss function that incorporates an adversarial or corrective term; and wherein the adversarial or corrective term guides the latent space of the first learning function (26) to produce representations aligned with target information extracted from the second trained learning function (25).

7. A method according to any one of claims 1 to 6 characterized in that it comprises: • the generation of a presence score (14A, 14B, 14C) by the second trained learning function (25) for each anomaly present on the synthetic anatomical image (12), said score being a function of the integral of the number of voxels and / or pixels classified for each predetermined anomaly; • the extraction of a bronchial volume (15) and / or a vascular volume from the synthetic anatomical image; and • the normalization (400) of each presence score generated by the extracted bronchial and / or vascular volume to generate, for each predetermined anomaly, a normalized presence score (16A, 16B, 16C).

8. System (1) for detecting structural anomalies comprising software and / or hardware means for implementing the method according to any one of claims 1 to 7.

9. Computer program product comprising code instructions which, when implemented by a computer (PRO), cause the system (1) according to claim 8 to carry out the steps of the process (1000) according to any one of claims 1 to 7.

10. Computer-readable data carrier (MEM) on which the computer program product according to claim 9 is recorded.

Images