Dassault Systèmes Patente

The disclosure notably relates to a computer-implemented method for forming a dataset of medical images. The dataset is configured to be used for training a model. The trained model is thereby configured to output a prediction from an input medical image. The dataset-forming method comprises obtaining a first set of medical images. The first set consists of annotated medical images. The first set is unbalanced with respect to a meta-property relative to each medical image. The dataset-forming method further comprises obtaining a second set of medical images. The second set comprises unannotated medical images. The dataset-forming method further comprises outputting a third set of medical images. The third set comprises the first set and unannotated medical images of the second set. The third set is balanced with respect to the meta-property. The dataset-forming method is an improved solution for forming a dataset of medical images.

The disclosure notably relates to a computer-implemented method for processing a medical image of a body of a patient on a horizontal support. The method comprises obtaining the medical image. The medical image comprises voxels. Each voxel is associated with a value. The value represents tissue density at the position represented by the voxel. The method further comprises discarding each voxel of the medical image associated with a value smaller than a predefined threshold. The discarding splits the image into connected components separated by discarded voxels. The method further comprises identifying, among the connected components, a connected component having a largest number of voxels. The method further comprises computing a curve separating the identified connected component from the remaining part of the medical image. The voxels bounded by the curve correspond to the body of the patient.

The disclosure notably relates to a computer-implemented method for generating, from one or more medical images each of an organ of a patient, a medical report compliant with a predefined set of medical report rules. The method comprises, automatically, determining positions and diameters of one or more lesions in the one or more medical images. The method further comprising, automatically, determining, based on determined positions and diameters of the lesions, a list of target lesions in accordance with the predefined set of medical report rules. The method further comprises, automatically, generating the medical report based on the determined target lesions and their positions and diameters.

The disclosure concerns a method for machine-learning a function configured constraints prediction for a sketch representing at least a portion of a mechanical part. The method comprises providing a training dataset of training examples, each comprising a graph representing a sketch. The graph comprises object nodes each representing a geometric object, geometry nodes each representing a geometric property, topological edges each representing a topological link, geometrical edges each connecting a geometry node to one object node, and constraint edges each representing a constraint and labeled with a label of a predefined set of labels each representing a constraint of a predefined constraints set. The method comprises training the function to take as input a graph representing a sketch and to output constraints each of a constraint edge of the graph and each belonging to the predefined constraints set.

The disclosure notably relates to a computer-implemented method for computing a score representing a percentage of occurrence of a clinical event. The method comprises obtaining a database including a list of P characteristics within a vector X, associated with the clinical event. The method comprises obtaining, for a given patient, the modalities corresponding to a subset of the characteristics in the list (p ∈ X*, wherein X* ⊂ X). The method comprises extracting, from the database, the modality weight associated with each modality obtained for the patient. The method comprises computing the score (score_patient) for the given patient by summing the extracted modalities weights, using the formula score_patient = Σ_{pj ∈}_X* α(p_j ). The method comprises computing a dispersion range associated with the computed score. Such a method forms an improved solution for computing a score representing a percentage of occurrence of a clinical event for a patient.

The disclosure notably relates to a computer-implemented method for generating a template image of an organ. The method comprises obtaining a dataset comprising a set of medical images, and an autoencoder pre-trained on a set of medical images. The method comprises determining the template image, by minimizing in the latent space a loss which is a function of the latent vector and, for each medical image of the dataset, parameters of a diffeomorphic deformation applied to the template image to obtain the medical image. The loss includes a first term penalizing a discrepancy between a posterior probability of the parameters of the diffeomorphic deformation, given the medical image, and a Gaussian prior probability of the parameters of the diffeomorphic deformation. The loss includes a second term penalizing a negative log-likelihood of the set of medical images of the dataset, given the latent vector and the parameters of the diffeomorphic deformation.

The disclosure notably relates to a computer-implemented method for encoding a time series of values of a physiological measure of a patient. The method comprises obtaining the time series of values of the physiological measure of the patient. The method further comprises, for each time point of the time series of values, obtaining a slope coefficient of the time series of values at the time point. The obtaining a slope coefficient is based on the time point, a predetermined number of previous time points, and a predetermined number of next time points. The method further comprises, for each time point of the time series of values, comparing the slope coefficient to a reference slope coefficient. The method is an improved solution for encoding a time series of values of a physiological measure of a patient.

The disclosure notably relates to a computer-implemented method for machine-learning a neural network comprising a Spike Neural Network for predicting a time occurrence of a clinical event. The method comprises providing a training dataset of training examples. Each training example comprises one or more time series of values of respective physiological measures of a patient over a past time period and a corresponding ground truth of a time occurrence of the clinical event. The method further comprises training the neural network based on the training dataset. The neural network is trained for taking as input one or more time series of values of the respective physiological measures of a patient over a past time period and for outputting a prediction of probabilities over a time period of the occurrence of the clinical event. The method is an improved solution for predicting a time occurrence of a clinical event.

A method for generation of a sequence of synthetic ultrasound images of an organ. The method comprises providing video diffusion model trained to take as input a sequence of semantically labelled 2D representations of the organ and a corresponding sequence of noise images, to generate a corresponding denoised sequence of ultrasound images respecting the labels. The method comprises providing several consecutive sequences of labelled 2D representations of the organ and corresponding consecutive sequences of noise images, iteratively applying the video diffusion model to each couple formed by one said consecutive sequence and one corresponding noise images sequence, including, at each iteration, using, for each denoising step of the backward process of the video diffusion model, a part of the denoised synthetic ultrasound images generated at the previous iteration for that denoising step in replacement of a part of images of the given couple for that denoising step.

The disclosure notably relates to a computer-implemented method for measuring an anatomical target from a set of medical images representing the anatomical target. The method comprises: obtaining at least one medical image of the anatomical target; obtaining an annotation associated with the obtained at least one medical image of the anatomical target, the annotation comprising a measurement of the anatomical target. The method further comprises obtaining a trained model configured for outputting, based on the obtained annotation, a segmentation mask of the anatomical target represented on the obtained at least one medical image. The method comprises applying the obtained trained model to the obtained at least one medical image and the obtained annotation. Thereby, the method identifies a segmentation mask of the anatomical target represented on the obtained at least one medical image. The method comprises computing a measurement of the anatomical target on the identified segmentation mask.

The disclosure notably relates to a computer-implemented method for annotating a set of medical images of at least one anatomical target of a same patient. The method comprises: obtaining at least one query from the set of medical images, each query being an unannotated 2D slice of a medical image; obtaining a support from the set of medical images, the support comprising a set of pairs, each pair comprising a 2D slice of the medical image and a respective 2D annotation; obtaining a pre-trained few-shot model being configured: for taking as input at least one query and a support; and for outputting, based on the inputted support, a 2D annotation associated with the inputted at least one query. The method also comprises predicting, by using the pre-trained few-shot model, a 2D annotation associated with the obtained at least one query.

The invention relates to a method for updating an exploded path (25) within a 3D scene (11), comprising:- displaying a 3D scene (11) provided with a three-axis system (15) and including an exploded path (25) extending from a starting point (SP) to a finishing point (FP), the exploded path (25) being a connected series of at least three line segments (25A, 25B, 25C, 25D) each aligned with one of the axes (X, Y, Z) of the three-axis system (15) according to a periodic direction pattern;and, upon instruction to apply a translation to a line segment (25C):- applying the translation to this line segment (25C) and to the following or preceding line segment (25B, 25D) to preserve the periodic direction pattern, and, if necessary,- adding a linking line segment to relink the exploded path (25) to the starting or finishing point (SP, FP).

The disclosure notably relates to a computer-implemented method of machine-learning, for learning a generative function configured to generate a B-rep given a conditioning signal representing a geometry. The generative function comprises a vertex neural network, an edge neural network, and a face neural network. The edge and face neural network each further comprise a topological model and a geometrical model. Each of the neural network is configured to generate the respective parts of the B-rep (vertices, edges, and faces) respecting the conditioning signal. This provides an improved solution for designing B-reps given a conditioning signal representing a geometry.

The disclosure notably relates to a computer-implemented method for forming a gap on a surface. The surface represents a surface portion of a product to be manufactured. The method comprises providing specifications of the gap. The specifications comprise a radius of a connection between the surface and flanges of the gap. The specifications further comprise a profile of the gap. The specifications further comprise a shape defining the gap on the surface. The method further comprises sweeping the profile along the shape. The method further comprises connecting the result of the sweep to the surface by applying a fillet operation. The fillet operation has as radius the provided radius.

The disclosure relates to a machine-learning method for learning a function configured for predicting a next feature in a feature tree. The method comprises providing a dataset of examples each comprising a graph representing at least a part of a feature tree. The graph comprises nodes each representing a feature and each labeled with a label of a set of labels each indicating a feature type of a predetermined set. The graph comprises edges each connecting nodes and representing a parent-child relationship between the features represented by the nodes. The graph comprises ground truth data indicating a next feature according to one or more selected features each corresponding to a graph node. The method further comprises training the function to take as input a graph representing at least a part of a feature tree and to output a prediction of one or more next features in the feature tree.

The invention relates to a method for previewing a 3D scene video having duration and frame rate comprising: a) receiving (200) at least one set of rendering parameter values of the 3D scene video, each value being associated with a respective timestamp, a respective interpolation function being associated with each pair of values having consecutive timestamps; b) determining (210) a number N of timecodes, where N is an integer based on the product of the duration and frame rate. The method further comprises iteratively, in a non-linear rendering order that is not chronological for at least some of the N timecodes: c) computing (250) a respective value for each set of rendering parameter values by using the interpolation function associated with the pair of values whose respective timestamps bound the current timecode, and d) rendering (280) a frame based on each rendering parameter value determined in operation c).

The disclosure notably relates to a computer-implemented method of designing a 3D modeled object representing a mechanical product. The 3D modeled object has a feature tree and an original value of a set of semantic parameters. The method comprises by a user graphically interacting with the 3D representation, and in particular, graphically selecting a semantic point of an outer surface of the 3D modeled object and graphically selecting one or more target 3D positions of the semantic point. The method further comprises, automatically, iteratively modifying the value of the set of semantic parameters. Thereby the method obtains an updated value of the set of semantic parameters. The method further comprises, automatically by the CAD system, updating the 3D representation of the 3D modeled object based on the updated value of the set of semantic parameters. This provides an improved solution for updating 3D modeled objects based on graphical interactions.

The disclosure notably relates to a computer-implemented data structure representing an adjacency matrix associated with a predicate of a set of quads of an RDF graph database, wherein the adjacency matrix being partitioned into a grid of rectangular partitions, the partitions having a same dimension; the adjacency matrix comprising a set of adjacency sub-matrices, each adjacency sub-matrix corresponding to a respective partition of the adjacency matrix; each adjacency sub-matrix being identified by its respective position in the grid.

The disclosure relates to a method for executing linear recursive queries. The method comprises obtaining a first part of the query that defines as input one or more initial conditions of query elements in a second part of the query on the RDF graph database. The method also comprises obtaining at least two clauses specifying that a second part of the query will be executed recursively. The first clause defines how an output of an execution of the second part of the query is used as an input of a next execution of the second part of the query. The second clause defines the second part of the query and comprises the query elements describing how the output of an execution is queried using (i) the input of the one or more initial conditions for an initial execution or (ii) the output of a previous execution for the next execution.

The disclosure notably relates to a computer-implemented data structure for storing RDF dataset comprising a set of RDF quads. The data structure comprises a first data structure. The first data structure comprises a first non-graph database with ACID properties for temporarily storing one or more modifications to be applied to the RDF dataset. The data structure also comprises a second data structure. The second data structure comprises a second non-graph database with ACID properties for storing a state of the RDF dataset as a binary representation. The binary representation is stored as binary data in a native structure of the second non-graph database.

The disclosure notably relates to a computer-implemented method for predicting a cardiovascular behavior of a newborn to birth from his/her fetus physiological parameters. The method comprises obtaining physiological parameters of the fetus, obtaining a non-calibrated surrogate cardiovascular model modeling a cardiovascular system of the fetus and modelling at least one physiological change triggered by the birth, calibrating the non-calibrated surrogate cardiovascular model with a data assimilation algorithm using the obtained physiological parameters of the fetus on the non-calibrated surrogate cardiovascular model, thereby obtaining a calibrated surrogate cardiovascular model of the fetus, and predicting the cardiovascular behavior of the newborn by triggering the at least one physiological change of the calibrated surrogate cardiovascular model of the fetus, thereby obtaining a calibrated surrogate cardiovascular model of the newborn.

The disclosure notably relates to a computer-implemented method of machine-learning for assembling mechanical parts, based on a mating score and a mating axis. The method comprises providing a dataset of pairs of B-Reps, each pair comprising at least one B-Rep representing an assembly of mechanical parts, being labelled with mating compatibility data and, if the B-Reps of the pair are compatible according to the mating compatibility data, mating axis compatibility data. The method also comprises training a neural network based on the dataset, configured for taking as input a pair of B-Reps. The neural network also outputs a mating score of a pair of single embeddings, each single embedding representing a B-Rep, the mating score representing a score of mating compatibility between the mechanical parts represented by the pair, and if the B-Reps are compatible according to the mating score, data defining a mating axis.

The disclosure notably relates to a computer-implemented method of using a machine-learning function preconfigured for taking an input 3D layout and a given noise level, and for predicting an output 3D layout. The function is preconfigured with a conditioning drop-out with respect to at least one layout parameter. The method further comprises obtaining a set of conditioning inputs and, for each conditioning input, determining one or more conditioning candidate 3D layouts and determining a plurality of perturbed conditioning candidate 3D layouts. The method further comprises, applying the preconfigured function to each perturbed conditioning candidate, in which the one layout parameter is dropped out, computing reconstruction errors, and averaging the reconstruction errors, thereby obtaining a score. This forms an improved solution for predicting 3D layouts.

The disclosure notably relates to a computer-implemented method of machine-learning. The method comprises obtaining a dataset of ground truth 3D layouts. The machine-learning method further comprises obtaining a probability distribution of noise levels. The machine-learning method also comprises, for each ground truth 3D layout, obtaining a respective perturbed 3D layout. The machine-learning method moreover comprises training a function. The function is configured for taking an input 3D layout and a given noise level, and for predicting an output 3D layout. The training is performed over the dataset based on a loss which penalizes a dissimilarity metric between each ground truth 3D layout and a respective predicted 3D layout.

Zusammenfassung wird geladen …

The invention notably relates to a computer-implemented method for forming a dataset configured for learning a neural network. The neural network is configured for inference, from a freehand drawing representing a 3D shape, of a solid CAD feature representing the 3D shape. The method comprises generating one or more solid CAD feature includes each representing a respective 3D shape. The method also comprises, for each solid CAD feature, determining one or more respective freehand drawings each representing the respective 3D shape, and inserting in the dataset, one or more training samples. Each training sample includes the solid CAD feature and a respective freehand drawing. The method forms an improved solution for inference, from a freehand drawing representing a 3D shape, of a 3D modeled object representing the 3D shape.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

<div notation="docdba" type="abstract" xml:base="/api/emi/section/d225e46b-e181-4fb3-a1b0-64f6dd337976/image" xml:lang="en"><p>The invention notably relates to a computer-implemented method for simulating evolution of a tumor associated to an oncogene. The method comprises providing (S10) a plurality of pieces of data, each corresponding to a given cell of the tumor, and includes a degree of activation of the oncogene in the given cell. The method further comprises providing (S20) a model configured to take an input piece of data and to output an information on proliferation of the respective given cell corresponding to the input piece of data. The information on proliferation depends on the degree of activation of the oncogene. The method further comprises running (S30) the model on one or more pieces of data of the plurality of pieces of data and updating (S40) the plurality of pieces of data based on the result of the running (S30). Such a method improves the simulation of the evolution of a tumor.</p></div>

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

The disclosure relates to a method for generating a CAD feature tree from a discrete geometrical representation of apart. The method comprises providing the representation. The method comprises providing a set of CAD features. The method comprises optimizing an objective function rewarding a fitting of the discrete geometrical representation by a candidate sequence of CAD features and penalizing a complexity of the candidate sequence. The complexity of the candidate sequence is a function of the candidate sequence that increases when adding a feature to the candidate sequence. The optimization comprises determining one or more sequences of CAD features from the set of CAD features, including performing a depth-first search in a tree of candidate sequences of CAD features based on rewarding a fitting of the discrete geometrical representation by a candidate sequence and on penalizing a complexity of a candidate sequence.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …