Method for determining the state of an individual with respect to reference states

Abstract

The invention relates to a method for processing physiological data of an individual, each physiological datum resulting from measurements performed on the individual, the method comprising establishing a score for the individual, the score being a z-score, and projecting the z-score into a space so as to determine a state of the individual.

Description

[0001]Description Title: Method for determining the state of an individual in relation to reference states TECHNICAL FIELD The technical field of the invention is the processing of data, for the purpose of aiding the diagnosis or prognosis of a pathology of an individual. PRIOR ART The biological or medical examinations carried out on patients are intended for a practitioner for the purpose of aiding diagnosis. A common practice consists of comparing physiological characteristics, determined on a patient, with so-called reference physiological characteristics, established on a reference population considered to be healthy. The digitization of examinations allows the creation of large databases for different types of examinations,whether it is the results of biological analysis or data obtained by imaging modalities or other types of measurements carried out on a patient for diagnostic purposes. These can be, for example, electrophysiological measurements such as ECG (Electrocardiogram) or EEG (Electroencephalogram). The creation of databases makes it possible to determine, with greater statistical precision, the reference physiological characteristics. It also allows for better individualization of reference values: contextual data (age, weight, height, sex) or physiological data resulting from other types of examination can be taken into account to refine the reference physiological characteristics. However, the databases are generally large. A dimensionality reduction may be relevant,as described in the publication Attyé et al “Tractlearn: a geodesic learning framework for quantitative brain bundles”. In this publication, the use of a dimensionality reduction algorithm by manifold learning, usually referred to as “manifold learning”, is described for the analysis of brain structures from characteristics resulting from nuclear magnetic resonance imaging (MRI). The inventors propose a method for processing physiological data, exploiting dimensionality reduction for diagnostic assistance purposes. DISCLOSURE OF THE INVENTION A first subject of the invention is a method for processing physiological data of an individual, each physiological data resulting from measurements carried out on the individual, the method comprising the following steps: - a) as a function of the physiological data of the individual, assignment of coordinates, called origin coordinates, in an origin space, to the individual,to each original coordinate being assigned a rank, the original coordinates forming a characteristic vector; - b) projection of the characteristic vector into a first latent space to form a latent vector; - c) application of a regression function to the latent vector resulting from b), so as to estimate an original coordinate, in the original space, for each rank; - d) for at least one rank, comparison of the original coordinates of the individual resulting from c) and a) and determination of a score of the individual as a function of the comparison; - e) formation of a score vector from scores determined for different physiological characteristics; the method being characterized in that it comprises - f) projection of the score vector into a second latent space, so as to obtain a projected score vector, the second latent space comprising at least one state group, corresponding to coordinates, in the latent space,associated with a state of the individual; - g) determining a position of the projected score vector relative to a group of states or to each group of states; steps a) to f) being implemented by computer. The state of the individual may be: - at least one pathological state or a healthy state; - and / or at least two different pathological states,corresponding respectively to at least two different pathologies; - and / or to the same two different states of the same pathology. According to one possibility: - the first latent space is determined from physiological data established on reference individuals of a reference population; - the second latent space is determined from physiological data established on individuals of different states. Step f) may include a projection of the score of the individual and at least one contextual characteristic of the individual. The contextual characteristic may include an age of the individual and / or a sex of the individual, or even morphological data of the individual, such as a height or a weight. The method may include, prior to step a),obtaining physiological data of the individual using at least one sensor. The method may comprise: - forming a two-dimensional or three-dimensional image of the individual; - extracting physiological data from each image of the individual. The image may result from a medical imaging modality such as MRI, CT, scintigraphy, infrared spectroscopy, Raman spectroscopy. The physiological data may result from an analysis modality such as: electrocardiogram, electromyogram, electroencephalogram, magnetoencephalography, oximetry, analysis of the composition of body gas. The method may comprise an analysis of a bodily fluid of the user, previously sampled, the physiological data of the individual resulting from the analysis of the bodily fluid. Another subject of the invention is a system for processing physiological data of an individual, the system comprising: - a sensor, or a set of sensors,configured to measure physiological data of the individual; - a processing unit, configured to receive the physiological data measured by the sensor or set of sensors, and to implement steps a) to f) of a method according to the first subject of the invention. A second subject of the invention is a system for determining a state of an individual, the system comprising: - a sensor, or a set of sensors, configured to measure physiological data of the individual; - a processing unit, configured to receive the physiological data measured by the sensor or set of sensors, and to implement steps a) to f) or a) to g) of a method according to the subject of the invention. Another subject of the invention is a support, connectable to a computer,comprising instructions for implementing steps a) to f) of a method according to the first subject of the invention from physiological data measured on an individual. The invention will be better understood upon reading the description of the exemplary embodiments presented, in the remainder of the description, in connection with the figures listed below. FIGURES Figure 1 shows an example of a system suitable for implementing the invention. Figure 2 diagrammatically shows the main steps for implementing a method according to the invention. Figure 3 diagrammatically shows a projection into a latent space Figure 4A shows groups of states, formed by latent vectors of different individuals whose states are known. The latent vectors are defined in a first latent space. Figure 4B shows groups of states,formed by projected score vectors of different individuals whose states are known. The projected score vectors are defined in a second latent space. Figure 4C is a figure similar to Figure 4B. In Figure 4C, contextual data, in this case the age of the user, has been taken into account in the score vector before its projection into the second latent space. Figure 5 shows groups of states of individuals presenting the same pathology and control individuals. The pathological individuals are divided into two states, depending on the sensitivity of the pathology to a treatment. DISCLOSURE OF PARTICULAR EMBODIMENTS Figure 1 represents an example of a system allowing an implementation of the invention. The system 1 comprises a system 10 for acquiring physiological data of an individual. In this example,the acquisition system is an MRI remnography system. The individual is a human or animal being. Physiological data means a value of a physiological parameter of the individual, likely to vary depending on the individual's state of health. This may be data resulting from an image acquisition system, or from a system for acquiring electrophysiological measurements (for example ECG - Electrocardiogram, EEG - Electroencephalogram, EMG - Electromyogram, MEG - Magnetoencephalography, oximetry or any other analysis of a body gas, in particular a respiratory gas). It may also be data measured on a sample previously taken from the patient. The sample may be, for example, a body fluid, such as blood or urine. The data may be a concentration of a cell or a protein or another molecule, of biological interest,in the body fluid. When the measured data result from an analysis of a body fluid, the physiological data acquisition system may be a biological analysis automaton. The system 1 comprises a processing unit 12, programmed to receive the data measured by the acquisition system and implement the processing steps described below. The processing unit may be a computer or a remote cloud-type computing architecture. One objective of the invention is to provide assistance in determining a possible pathology from the measured physiological data. The physiological data of the individual, measured by the acquisition system 10 and processed by the processing unit 12, are preferably numerous. The number of physiological data is preferably greater than 5 or even 10. It is typically several dozen. As in the publication cited in the prior art,the method implements a manifold learning method using linear or non-linear dimensionality reduction methods. Different data dimensionality reduction methods are available. Generally speaking, a dimensionality reduction method makes it possible to represent original coordinates, in an original space (or real space) ^^ of dimension ^^, into coordinates in an arrival space ^^ of dimension ^^, with ^^ > ^^. The arrival space is usually referred to as the "latent space". As previously indicated, the dimension of the original space ^^, can be a few tens. The dimension of the arrival space ^^ is preferably less than 10. The original coordinates, or characteristics,are formed from physiological data of the individual. The original coordinates form a feature vector. The dimension of the arrival space depends on the number of physiological features and their complexity. Preferably, it is less than 10 so as to facilitate calculations. The transition from the original space ^^ to the latent space ^^ is performed by a projection function ^^ applied to the vector formed from the physiological features, called the feature vector. The transition from the latent space ^^ to the original space ^^ is performed by a backprojection function ^^, ି^applied to a vector formed by the coordinates in the latent space. The latent space ^^, as well as the projection and backprojection functions are defined during a learning phase, during which a set of reference physiological characteristic vectors is available. Each reference physiological characteristic vector includes data measured on reference individuals forming a reference population. The objective of dimensionality reduction is to obtain a representation of the reference physiological characteristics in the latent space, the latter being of restricted dimension compared to the original space. Several dimensionality reduction methods are known to those skilled in the art. For example, principal component analysis (PCA) is an unsupervised and linear dimensionality reduction method.The basis of the latent space is formed by eigenvectors of the covariance matrix of the feature vectors. Nonlinear methods can be implemented, for example a multidimensional scaling (MDS) method, preserving the distance values ​​between all pairs of data in the original space. An example of an MDS method is the Sammon method. Among the nonlinear methods, we can also mention the Isomap (Isometric feature mapping) method, allowing to define a transformation preserving the geodesic distance between the data. Another type of method is the UMAP (Uniform Manifold Approximation and Projection) method. Figure 2 shows the main steps of a method according to the invention. We are first interested in the learning phase, allowing to define latent spaces and projection and backprojection functions.Steps 200 to 260 describe an implementation of a method for processing characteristics of a test individual, for which it is desired to establish a diagnostic aid. By diagnostic aid, we mean a provision of information relating to the state of health of the individual, forming an aid to a practitioner to establish a diagnosis. Steps 100 to 190 are learning steps of the method. Step 100: Formation of reference vectors. During this step, a set of physiological data is available. measured on reference individuals ^^^^^^ ^ forming a reference population. The reference population is made up of individuals whose condition is considered known, and preferably healthy. For each reference individual, based on the measured physiological data, characteristics are established ^^ ^,^ forming a feature vector ^^ ^defined in the original space. The index ^^ corresponds to a rank assigned to each physiological data, with 1 ≤ ^^ ≤ ^^, ^^ denoting the number of measured physiological data. In this example, each characteristic is a centered and reduced measured physiological data: where ^^ ^ is an average of the values ​​^^ ^,^ for all ^^ reference individuals and ^^ ^ is the standard deviation of the characteristics ^^ ^,^ of the same rank ^^. The characteristics of each reference individual form a reference vector ^^ ^ , of dimension ^^, assigned to the individual ^^^^^^ ^ . Each feature ^^ ^,^ corresponds to an origin coordinate of the reference individual ^^^^^^ ^ in the original space ^^. For example, each measured value can be an analysis of a parameter resulting from a blood test or an image. For each reference individual ^^^^^^ ^, we establish a reference vector ^^ ^ comprising the measured values, for the reference individual, centered and reduced by taking into account the entire reference population. Step 110: dimensionality reduction During this step, a dimensionality reduction algorithm is applied to all or part of the reference characteristics of the reference vectors ^^ ^ . The dimensionality reduction algorithm defines a first projection function ^^ ^ linear or non-linear. This is for example a UMAP type algorithm. The projection function ^^ ^ allows you to project the vector ^^ ^ in a first latent space ^^ ^ of dimension ^^ ^ less than the dimension of the original space ^^. In the first latent space^^^, we obtain a projected vector ^^^ ൌ ^^^൫^^^൯, whose dimension ^^^ is less than the dimension of the vector ^^ ^ . The first latent space ^^^ is usually called "variety" by those skilled in the art. The dimension ^^ ^ of the projected vector ^^ ^ is usually between 1 and 10. Each term ^^ ^,^, of the projected vector is a coordinate of the reference individual ^^^^^^ ^ in the latent space. The index ℎ corresponds to a rank assigned to each term ^^ ^,^ . ℎ is an integer between 1 and ^^ ^ , being the dimension of the first latent space ^^ ^ .The dimension is usually between 1 and 10. Step 120: formation of the latent vector The terms ^^ ^,^ of the projected vector ^^ ^ can form a latent vector ^ ^ ^ of dimension ^^ ^ . According to one possibility, for each rank ℎ, the terms ^^ ^,^ of the projected vector ^^ ^ can be centered and reduced, so as to obtain coordinates ^^ ^,^ reduced centered in latent space: where ^^ ^is an average of the terms ^^ ^,^ for all ^^ reference individuals for the same rank ℎ, and ^^ ^ is the standard deviation of the terms ^^ ^,^ of the same rank ℎ for the set of ^^ reference individuals. For each reference individual, the set of reduced centered coordinates ^^ ^,^ forms a latent vector ^ ^ ^ of dimension ^^ ^ .Thus, the latent vector ^^^includes either the terms ^^^,^ of the projected vector ^^^, in which case ^^^ ൌ ^^^, or the reduced centered terms ^^ ^,^ as described in connection with (2). A backprojection (or regression) function ^^ ^ ି^ allows to estimate an original vector^^^ such that has the same dimension as ^^ ^ . Each term of the vector is an estimated feature ^^ ^,^ in the original space ^^, from the latent vector ^ ^ ^. The rear projection function ^^ ^ ି^can be established from a Nadaraya-Watson kernel estimator. Such an estimator is commonly used to establish nonparametric regression models. It is constructed from a kernel function ^^, for example Gaussian, and a window of size ℎ. When the projected vector ^^^ is not centered and reduced, ^^^ ൌ ^^^ and ^^^ ൌ ^^ି^^ ൫^^^൯ ൌ ^^^ ି^ ൫^^ ^ ൯. According to one possibility, described later, a regression function is determined for each rank ^^. Thus, from the latent vector ^ ^ ^, we estimate a characteristic of rank ^^in the original space ^^ such that: is a regression function determined for the rank characteristic ^^. According to this variant, two regression functions, applied to the same latent vector, and allowing the estimation of two different original coordinates, are different. 130: Backprojection and estimation of the characteristic vector in the original space ^^. The application of the backprojection function ^^ ^ ି^ to the latent vector ^ ^ ^ allows to obtain an estimate of a vector defined in the original space ^^, as described in step is such that: ^^^ ൌ ^^^ ^ ^^^,^4^can be considered as a digital twin of ^^ ^ ^^ ^ is a deviation vector, of the same dimension as and ^^ ^, . ^^ ^ corresponds to a residual of the regression to the model. Plus the reference individual ^^^^^^ ^ is representative of the reference population, plus the norm of ^^ ^,^is weak. The deviation vector ^^ ^ has residues ^^ ^,^ respectively associated with each rank characteristic ^^: ^^^,^ ൌ ^^^,^ ^ ^^^,^^5^The distribution of the values ​​of ^^ ^,^ varies depending on the reference individual ^^^^^^ ^ and the rank of the characteristic ^^. Plus the dimension ^^ ^ increases, the more the average of the residues ^^ ^,^ of each individual is low. It is considered that beyond a dimension of 3 or 4, the gain in terms of average residue decreases or no longer evolves significantly. Each residue ^^ ^,^, can be used to determine a z-score for the individual ^^^^^^ ^ , relative to the characteristic ^^. The term "z-score" is a common term for those skilled in the art. The determination of the first latent space and the first projection function are carried out in such a way as to minimize the residuals for the reference population. The first latent space ^^ ^and the first projection function ^^ ^ are established by implementing a dimensionality reduction technique. This involves preserving or maximizing variability between individuals, or allowing separation of individuals by class, while reducing the dimension. Steps 100 to 130 correspond to an initial learning process, during which the first latent space ^^ ^ is defined, as is the first projection function ^^ ^ and the backprojection function Steps 140 to 180 describe a second training, in which the first training is used to define a second latent space ^^ ଶ , different from the first latent space ^^ ^ , in which at least one state group ^^ ^ is defined. The or each state group is associated with a state, chosen from a healthy state or a pathological state, the pathology being known. Each state group ^^ ^is a set of coordinates, in the second latent space, which correspond to the same state s. More generally, the state of the individual can correspond to: a type of pathology, a characteristic of a pathology (for example Alzheimer's disease with atrophied or non-atrophied hippocampus), or a state of evolution of a pathology or a severity of a pathology or an inflammatory state. The pathological state can also represent a prediction of evolution of a pathology, or a typology of a pathology, for example a sensitivity of a pathology to a treatment. 140: formation of a vector of characteristics The implementation of step 140 assumes the taking into account of a learning population, comprising individuals ^^ considered as having a known state ^^. The state is either a known pathological state or a healthy state.Preferably, the learning population is formed of individuals in a state chosen from at least two different pathological states: healthy state and pathological state, or at least two different pathological states, that is to say: two different pathologies, or two different typologies of the same pathology (different clinical scores for the same pathology, different levels of severity or advancement of the same pathology, sensitivity to a different treatment of the same pathology). For each individual in the learning population, we have a vector of characteristics ^^. ^,^ as defined in step 100. Each feature is centered reduced according to (1) if this option was chosen in step 100, to define the first latent space. : Projection into the first latent space ^^ ^ of each vector ^^ ^,^In this step, the respective feature vectors of each individual ^^ having a state ^^ are projected into the first latent space ^^ ^ , so as to form a latent vector ^^ ^,^ , as described in connection with step 120. For each individual ^^, the first projection function ^^^ is implemented, so as to obtain a projected vector ^^^,^ ൌ ^^^൫^^^,^൯. The vector ^^^,^ is obtained from ^^ ^,^ as previously described. backprojection During this step, the regression function is implemented so as to estimate, from each latent vector ^^^,^, the original vector ^^^,^ such that ^^^,^ ൌ The regression function is the one determined during step 120. Step 170 During this step, we establish, for each characteristic ^^ of each individual ^^, presenting the state ^^, a Z-score ^^ ^,^,^ for each characteristic ^^. such that: Generally speaking, the z-score involves a comparison between ^^ ^,^,^ and ^^ ^,^,^ . In the example of formula (6), the comparison is a subtraction normalized by the standard deviation ^^ ఌ^ which corresponds to the standard deviation of the residual values ​​^^ ^,^ determined for characteristic i for each reference individual. Step 180: formation of a vector of scores. During this step, from the z-scores calculated during step 170, a vector of scores is established ^^ ^,^ for each individual ^^ considered to be in state ^^. The dimension of the score vector corresponds to the number of z-scores that are considered relevant for the next step. The score vector can include terms representing contextual characteristics, for example the user's weight or height or age or gender. Thus, the dimension of the score vector is ^^' can be equal to or different from ^^. Step 190: formation of a second latent space and determination of a second projection function During this step, a second latent space ^^ is determined ଶ , as well as a second projection function ^^ ଶ allowing a projection of a vector of scores, defined in the original space R, into the second latent space. The dimension ^^ ଶ of the second latent space ^^ ଶ can be equal to or different from the dimension of the first latent space ^^ ^ . The second projection function ^^ଶ allows to form a projected score vector ^^^,^ ൌ ^^ଶ൫^^^,^൯. (7). The second latent space ^^ ଶ is such that the projected score vectors of individuals ^^ presenting the same state ^^ are grouped, in the second latent space, so as to form a state group ^^ ^ .As for the first latent space, the second latent space is sized to capture the variability of the training population in a limited dimension space, preferably less than 5, or even less than or equal to 3. The second latent space is then defined so as to allow, in a restricted dimension space, a better possible separation between individuals of different states. Steps 150 to 190 are carried out using physiological data measured with individuals having at least two different and known states: healthy state / pathological state (presence of a pathology), two different pathological states, as previously described, two different types of the same pathology, different clinical scores for the same pathology. Steps 140 to 190 can be carried out using physiological data measured with training individuals having different pathologies.In this case, the second latent space is defined so as to separate each pathology as best as possible. The second latent space is then defined so as to optimize a separation between healthy individuals and individuals presenting each pathology. More generally, during steps 140 to 190, the learning population is formed of individuals whose state ^^, pathological and / or healthy, is known. The second latent space is determined so as to separate the individuals as best as possible according to their state. In the second latent space, the coordinates corresponding to each state are respectively grouped together to form state groups ^^. ^ . Steps 200 to 250 correspond to an implementation of the invention on a test individual. The test individual is assigned a vector of physiological data ^^ ^,௧^^௧It is understood that for each rank, the physiological data of the test individual are of the same nature as the physiological data established for each reference individual or each individual of the learning population. Step 200: From the measured physiological data ^^ ^,௧^^௧ , we establish characteristics ^^ ^,௧^^௧ forming a feature vector ^^ ௧^^௧ defined in the original space ^^. In this example, each feature is a measured physiological data centered and reduced: ^^ ^ and ^^ ^ were defined in step 100. Step 210: Projection of the feature vector into the first latent space In this step, the vector ^^ ௧^^௧ is projected into the first latent space ^^ ^defined in the first learning step (steps 100 to 130), so as to obtain a projected vector^^௧^^௧ ൌ ^^^^^^௧^^௧^. The projection is carried out using the first projection function ^^^previously established following the first learning. The vector ^^௧^^௧ has coordinates ^^^,௧^^௧, with is the dimension of the first latent space ^^ ^ . In Figure 3, the main vectors implemented in steps 200 to 260 are shown. Step 220: obtaining latent coordinates Optionally, the projected vector ^^ ௧^^௧ can be centered and reduced, so as to obtain coordinates ^^ ^,௧^^௧ reduced centered in the first latent space: where ^^ ^ is an average of the terms ^^ ^,^ for the set of ^^ reference individuals for the same rank ℎ, and ^^ ^ is the standard deviation of the terms ^^ ^,^ of the same rank ℎ for all ^^ reference individuals. ^^ ^is determined during the first learning phase previously described. See expression (2) of step 120. When the projected vector ^^ ௧^^௧ is not centered and reduced, each latent coordinate ^^ ^,௧^^௧ is such that ^^^,௧^^௧ ൌ ^^^,௧^^௧. Thus, ^^௧^^௧ ൌ ^^௧^^௧The coordinates ^^ ^,௧^^௧ form a vector ^^ ௧^^௧ of dimension in the first latent space ^^ ^ . Step 230: Backprojection During step 230, the regression function corresponding to the latent space ^^ ^ is applied to the vector ^^௧^^௧ to estimate a vector ^^௧^^௧ such that ^^௧^^௧ ൌ ^^^ ି^^ ^^௧^^௧ ^ . Chaque term of the vector ^^ ௧^^௧ is an estimated feature ^^ ^,௧^^௧ ,in the original space ^^, from the vector ^^ ௧^^௧ coordinates in the first latent space ^^ ^ . Step 240: Calculating a score During this step, we determine the z-score ^^ ^,௧^^௧, of the test individual, and this for each rank characteristic ^^: ^^ ఌ^ was previously defined. According to one possibility, the z-score ^^ ^,௧^^௧ is determined according to the expression: ^^′ ఌ^ is determined according to a so-called “leave one out” procedure, according to which the iterative steps 110 and 140 are carried out by successively removing a reference individual of rank ^^′. This makes it possible to form ^^ first latent spaces ^^ ^,^ ᇲ .Each first latent space ^^ ^,^ ᇲ is defined from the reference population, without taking into account the reference individual of rank ^^′. Subsequently, steps 200 to 230 are implemented using the individual of rank ^^' as a test individual. Thus, for each individual of rank ^^', we obtain a residual such as^^^,^ᇲ ൌ ^^^,^ᇲ െ ^^^,^ᇲ (5').^^′ ఌ^, then corresponds to the standard deviation of the residuals ^^ ^,^ ᇲcorresponding respectively to the different individuals of rank ^^' considered. The reference individuals rank j' can be chosen from among the reference individuals closest to the test individual, for example the 30 closest. The different z-scores, respectively established for each characteristic i, allow the formation of a vector of scores ^^ ௧^^௧ , each term of which is either a z-score ^^ ^,௧^^௧ , or a contextual characteristic, for example chosen the age and / or sex of the individual and / or weight or height or other morphological criteria, for example body mass index. : projection of the score vector ^^ ௧^^௧ . During this step, the score vector ^^ ௧^^௧ is projected into the second latent space^^ଶ so as to obtain a projected score vector ^^௧^^௧ ൌ ^^ଶ^^^௧^^௧^. The projection is performed using the second projection function ^^ ଶpreviously established during step 190. 260 As previously described, the second latent space ^^ ଶ has state groups ^^ ௌ grouping coordinates corresponding respectively to different states of the user: healthy state and / or one or more pathological states. When the coordinates of the projected score vector ^^ ௧^^௧ are part of a state group ^^ ௌ , an indication is obtained that the test individual is likely to be affected by the state ^^ associated with this group. The method comprises determining a position of the vector ^^ ௧^^௧ compared to each state group ^^ ௌ . The diagnosis of the individual, depending on the position, is up to the practitioner. Thus, the invention can be limited to providing information allowing the practitioner to make a diagnosis. The dimension of the or each group of states ^^ ௌ . corresponds to the dimension ^^ ଶof the second latent space ^^ ଶ .In the examples described below, the second latent space has a dimension equal to 2. Each state group is then formed by a surface or a union of several surfaces. When the dimension of the second latent space is equal to 3, each state group ^^ ௌ is a volume or a union of volumes. The second latent space can contain several state groups ^^ ௌ , each state group being associated with a different state (e.g. a different pathology) from that associated with another state group. The second latent space may include a state group considered to be representative of a healthy state of the test individual. When the coordinates of the vector ^^ ௧^^௧are included in the latter, the test individual is considered healthy. Experimental tests. During a first series of tests, the physiological data determined were volumes of brain structures observed on anatomical images obtained by MRI, with T1 weighting.The following were considered for training: - 425 training individuals with Alzheimer's disease (AD), whose characteristics were extracted from the international cohorts AIBL (Australian Imaging Biomarker and Lifesyle), MIRIAD (Minimal Interval Resonance Imaging in Alzheimer's Disease) and ADNI (Alzheimer's Disease Neuroimaging Initiative); - 296 training individuals with frontotemporal dementia (FTD), whose characteristics were extracted from the international cohorts NFID (Northern Finland Intellectual Disease) and 4RTNI (R Repeat Tauopathy Neuroimaging Initiative) - 537 training individuals with Parkinson's disease (PD), whose characteristics were extracted from the international cohort PPMI (Parkinson's Progression Markers Initiative). The interest of these cohorts is that they were established from several MRI devices, with different T1 sequence settings.We also considered 398 healthy training individuals (C: control). These were used as reference individuals, to form the first latent space during the first training, and as training individuals, to form the second latent space during the second training. We implemented steps 100 to 200 by taking into account: - for steps 100 to 130: healthy individuals; - for steps 140 to 200: healthy individuals and pathological individuals. For each individual, we considered 196 features from T1 MRI imaging sequences. Figure 4A shows the first latent space, of dimensions 2, in which we grouped the latent vectors ^^. ^,^ individuals, reduced non-centered. Figure 4B shows the second latent space, of dimensions 2, in which the projected score vectors have been grouped ^^ ^,^ individuals. In each figure, we have represented groups of states ^^ௌ , gathering the vectors ^^ ^,^ (figure 4A) and ^^ ^,^ (Figure 4B) of individuals of the same state ^^. We observe that in Figure 4B, the groups of states are better separated than in Figure 4A. This shows that if we want to separate the individuals according to their state, the vector ^^ ^,^ is more relevant than the vector ^^ ^,^ . Figure 4C shows an embodiment in which we have taken into account, in the score vector ^^ ^,^ of each individual, an age of each individual. Then the score vector ^^ ^,^ was projected into the second latent space, so as to form a projected score vector ^^ ^,^ . We observe that in Figure 4C, the state groups ^^ ^are better separated than in Figure 4B. Thus, taking into account contextual data can improve the separation of individuals according to their state. In Figure 4C, each set is delimited by an outline annotated according to the percentage of individuals of the same state that it delimits: for example AD 80corresponds to a contour containing 80% of individuals with Alzheimer's disease. In a second series of tests, 196 cortical structures were analyzed for control subjects (group C), as well as for subjects with multiple sclerosis (MS), with different levels of effectiveness for a given first-line treatment: MS group 1 corresponds to individuals for whom the first-line treatment works moderately. MS group 2 corresponds to individuals for whom the first-line treatment does not work. In this example, there are 40 individuals in each MS1 and MS2 group and 100 individuals in the control group C. Figure 5 shows a projection, in the second latent space, obtained with all the individuals taken into account: the control populations C, MS 1 and MS 2 can be separated.Thus, the implementation of the invention makes it possible to estimate, for an individual presenting a pathology, in this case multiple sclerosis, the type of pathology, according to a sensitivity to a treatment. The invention makes it possible to form, in the second latent space, an “atlas” of states, in which the projected score vectors are grouped according to a state of a user. The position of the projected score vector forms an assistance to a practitioner to evaluate the state of the user and form a diagnosis. A notable advantage of the method is that it allows measured physiological data to be taken into account, by limiting a bias linked to the measuring instrument.The first reduced space makes it possible to capture the effects linked to inter-individual variability as well as to the variability linked to the measuring instrument, in particular when the reference population is sufficiently large: the characteristics of each individual are measured with different devices, which makes it possible to take into account the instrument variability, which corresponds to the variability affecting the measurement parameters of each instrument. The projection into the second latent space, from z-scores, makes it possible to limit the influence of the instrument variability. This makes it possible to obtain a clearer separation according to the state of each individual: the pathology is thus visualized more specifically and the different pathological states can be better separated. The invention can be implemented using physiological data resulting from different types of analysis, for diagnostic aid purposes.

Claims

CLAIMS 1. Method for processing physiological data of an individual, each physiological data resulting from measurements carried out on the individual, the method comprising the following steps: - a) as a function of the physiological data of the individual (^^ ^,௧^^௧ ), assignment of coordinates, called origin coordinates (^^ ^,௧^^௧ ), in an original space (^^), to the individual, each original coordinate being assigned a rank (^^), the original coordinates forming a vector of characteristics (^^ ௧^^௧ ) ; - b) projection of the feature vector into a first latent space to form a latent vector (^^ ௧^^௧ ) ; - c) application of a regression function to the resulting latent vector of b), so as to estimate an original coordinate (^^ ^,௧^^௧ ), in the original space (^^), for each rank (^^); - d) pour au moins un rang (^^), comparaison (^^^,௧^^௧ െ ^^^,௧^^௧) des coordonnées d’origine the individual resulting from c) and a) and determination of a score for the individual (^^ ^,௧^^௧) depending on the comparison; - e) formation of a vector of scores (^^ ௧^^௧ ) from scores determined for different physiological characteristics; the method being characterized in that it comprises - f) projection of the score vector (^^ ௧^^௧ ) in a second latent space, so as to obtain a projected score vector (^^ ௧^^௧ ), the second latent space containing at least one state group ^^^ ௌ^, corresponding to coordinates, in the latent space, associated with a state of the individual; - g) determining a position of the projected score vector relative to a group of states or to each group of states; steps a) to f) being implemented by computer.

2. Method according to claim 1, in which the state of the individual is chosen from: - at least one pathological state or a healthy state; - and / or at least two different pathological states, corresponding respectively to at least two different pathologies; - and / or to the same two different states of the same pathology.

3. Method according to claim 1 or claim 2, wherein - the first latent space is determined from physiological data established on reference individuals (^^), of a reference population; - the second latent space is determined from physiological data established on individuals (^^) of different states (^^).

4. Method according to any one of the preceding claims, wherein step f) comprises a projection of the score of the individual and at least one contextual characteristic of the individual.

5. Method according to claim 4, wherein the contextual characteristic comprises an age of the individual and / or a sex of the individual and / or morphological data of the individual.

6. Method according to any one of the preceding claims, comprising, prior to step a), obtaining the physiological data of the individual using at least one sensor. 7.Method according to claim 6, comprising - forming a two-dimensional or three-dimensional image of the individual; - extracting physiological data from each image of the individual.

8. Method according to claim 6, wherein the image results from a medical imaging modality such as MRI, CT scan, scintigraphy, infrared spectroscopy, Raman spectroscopy.

9. Method according to claim 6, wherein the physiological data results from an analysis modality such as: electrocardiogram, electromyogram, electroencephalogram, magnetoencephalography, oximetry, analysis of the composition of body gas.

10. Method according to claim 6, comprising an analysis of a bodily fluid of the user, previously taken, the physiological data of the individual resulting from the analysis of the bodily fluid. 11.A method according to any preceding claim, wherein the latent vector is obtained by a projection of the feature vector into a first space. latent space followed by processing to center and reduce the coordinates of the projected feature vector.

12. Method according to any one of the preceding claims, wherein the second latent space is different from the first latent space.

13. Method according to any one of the preceding claims, wherein - the first latent space has been previously obtained during a learning phase, by applying a dimensionality reduction algorithm to the physiological data of reference individuals considered to be healthy - the second latent space has been previously obtained, during the learning phase, by applying a dimensionality reduction algorithm to z-scores determined for individuals of different and known states: healthy state and pathological state or different pathological states or different types of the same pathology. 14.System (1) for processing physiological data of an individual, the system comprising: - a sensor (10), or a set of sensors, configured to measure physiological data of the individual; - a processing unit (12), configured to receive the physiological data measured by the sensor or the set of sensors, and to implement steps a) to f) of a method according to any one of the preceding claims.

15. Support, connectable to a computer, comprising instructions for implementing steps a) to f) of a method according to any one of claims 1 to 13 from physiological data measured on an individual.

Images