EP4104073A1

EP4104073A1 - Method for generating a composite nutritional index, and associated system

Info

Publication number: EP4104073A1
Application number: EP21707891.4A
Authority: EP
Inventors: Eric LAMEIGNÈRE
Original assignee: Fablife
Current assignee: Fablife
Priority date: 2020-02-11
Filing date: 2021-02-11
Publication date: 2022-12-21
Also published as: US20230089697A1; WO2021160720A1

Abstract

Disclosed is a method for generating a composite nutritional index (INC₁) comprising:  Selecting (SEL) an individual (U₁);  Acquiring (ACQ₁) a first set of phenotype data of the individual (U₁) characterising phenotypic descriptors (PHE₁);  Acquiring (ACQ₂) a second set of data (ENS₂) of a genotype characterising genotypic descriptors (GEN₁) of the individual (U₁);  Applying a set of predefined rules (R1_CIB) comprising:  Generating a set of personalised phenotypic and genotypic indices (IP_kp, IG_i) of an individual;  Calculating a target value (V_CIB) of a daily intake of the at least one nutrient (NUT_i) based on the application of an inference engine configured from:  Determining a composite nutritional index (INC₁) comprising an operation of associating a plurality of a target value (V_CIB) of a daily intake of the at least one nutrient (NUT_i) with at least one metabolic function (F_M).

Description

Title: PROCESS FOR GENERATING A COMPOSITE NUTRITIONAL INDEX, ASSOCIATED SYSTEM

The field of the invention relates to the field of methods and systems aimed at generating quantifications of daily nutrient intakes as a function of personalized parameters of an individual.

Currently there are recommendations to develop reference values for nutrient intakes for a population of individuals. Current methods generally define values taking into account some parameters of an individual such as gender or age or condition, for example, a pregnant woman.

However, current methods do not make it possible to take into account classes of individuals with physiological or genetic specificities in order to generate personalized / individualized nutritional recommendations. A first obstacle is the multitude of factors that can possibly be taken into account. Indeed, it is difficult to prioritize factors among themselves and to establish cause and effect relationships according to a person's physiological or medical history and the genetics of a subject. The methods aiming to integrate a larger panel of parameters even come up against incompatibilities of recommendations which can sometimes be contradictory or outside the norm depending on the physiological signs considered and in particular the pathological situations.

There is a need to automatically and quickly generate nutrient intake quantifications for an individual defining personalized or individualized recommendations. This need should preferably take into account not only a wide range of parameters and also the specificities of the individual to determine appropriate daily intake values that are compatible with one another.

There is therefore a need to define a solution which can respond to this problem. The present invention aims to resolve the aforementioned drawbacks. According to one aspect, the invention relates to a method for generating a composite nutritional index comprising:

^■ Selection of an individual;

^■ Acquisition of a first set of phenotype data of said individual characterizing phenotypic descriptors, said data comprising at least one age datum, one gender datum and at least one set of data characterizing physiological signs of said individual;

^■ Acquisition of a second set of data of a genotype characterizing genotypic descriptors of said individual, said data comprising information characterizing mutations and / or variations of at least one gene;

^■ Application of a set of predefined rules comprising: o at least a first subset of rules aimed at generating at least one phenotypic index from a calculation of a score of a quantification of a phenotypic descriptor, said index being standardized; at least a second subset of rules aimed at generating at least one genotypic index from a calculation of a score of a quantification of a genotypic descriptor, said index being normalized;

^■ Generation of a set of personalized phenotypic and genotypic indices of an individual;

^■ Calculation of target values of daily intakes of a plurality of nutrients from the application of an inference engine configured from: o a knowledge base comprising a repository of predefined values of phenotypic indices and / or genotypic and at least one set of conditional rules applied to said predefined values of phenotypic indices and genotypic indices and; o a fact base comprising all the phenotypic and genotypic indices of said individual calculated from the data acquired, ^■ Determination of a composite nutritional index comprising an operation of associating a selection of target values of daily intakes of all the nutrients with at least one metabolic function. The method is preferably implemented by computer.

One advantage is that it automatically calculates the daily intake of nutrients in an adapted and individualized manner for a given individual. Another advantage is to group these contributions according to predefined metabolic functions. This grouping allows for better readability for the individual. Grouping the daily intake target values also improves the individual response to each metabolic function. Another advantage is the synergy of metabolic effects obtained when daily intakes are followed by an individual.

Finally, the use of an inference engine in this case makes it possible to obtain a convergence of multicriteria values, each criterion having dependencies according to other criteria. Advantageously, the invention makes it possible to take into account criteria corresponding to genotypic and phenotypic data by dispensing with a data model linking the various phenotypic and genotypic factors together. According to one embodiment, the method comprises the steps of:

^■ receiving a plurality of individual reference values of a daily intake of a plurality of nutrients;

^■ generation of at least one differential indicator representing for each nutrient an individual reference value and a target value of daily intake of said nutrient.

An advantage of the invention is to take into account individualized reference values. The individualized reference values take into account an individual's phenotypic and genotypic context. An advantage over existing solutions and to better adapt the target values to a given individual.

According to one embodiment, the individual reference values of daily intakes of said nutrients are:

^■ directly extracted from a knowledge base referencing predefined daily intakes of nutrients and / or; ^■ calculated from reference rules automatically calculating daily nutrient intakes from data on phenotypes of said individual and from predefined values referenced in a knowledge base. According to one embodiment, the step of determining a composite nutritional index comprises a plurality of groupings of target values of daily nutrient intakes, each grouping contributing to improving a given metabolic function of said individual.

One advantage is to deliver an intake recommendation for a given metabolic function. This grouping makes it possible to pool an action by a user aimed at responding to a metabolic function. Finally, this solution makes it possible to optimize a nutritional action to achieve a metabolic goal.

According to one embodiment, the nutrients are macronutrients or micronutrients, said macronutrients being associated with a metabolic function quantifying an energy supply from said individual.

According to one embodiment, at least one reference value of a daily intake of an overall energy quantity of at least one macronutrient is calculated for an individual from a first set of phenotypic descriptors comprising an age, a gender. . In this case, at least one target value of a daily intake of an overall energy quantity of said macronutrient is calculated for said individual from a first set of phenotypic descriptors comprising an age, a gender and a second set of. phenotypic and / or genotypic descriptors.

According to one embodiment, at least one reference rule comprises, for the calculation of a reference value of a daily intake of a given nutrient, at least one operation handling a first set of quantifications of phenotypic descriptors. In this case, at least one target rule comprises, for the calculation of a target value of a daily intake of the nutrient, at least one operation manipulating, in addition to the first set of quantifications of phenotypic descriptors, a second set of quantifications of phenotypic and / or genotypic descriptors. An advantage of this modeling is to compare a target value calculated by a target rule taking into account different criteria with a reference value obtained by a reference rule. Thus, the reference values can also be individualized in order to optimize the individualization of the calculation of the target values.

According to one embodiment, at least one reference rule comprises, for the calculation of a reference value of a daily intake of a nutrient, at least one operation considering a first set of quantifications of phenotypic descriptors. According to this case, at least one target rule comprises, for the calculation of a target value of a daily intake of the nutrient, at least one operation aimed at defining a fixed value of a daily intake of said nutrient as a function of at least a threshold value reached by at least one quantification of a phenotypic and / genotype descriptor. One advantage is to take into account a threshold value to generate a target value which remains acceptable from a nutritional recommendation point of view. This solution also makes it possible to generate target values for nutritional supplements when a threshold is reached. According to one embodiment, at least one phenotypic descriptor is calculated from a sum of scores, each score quantifying a physiological state of the individual.

One advantage is to quantify and standardize the indices generated in order to carry out operations in particular on a large panel of users. According to one embodiment, at least one step of generating a list of recipes is carried out, said list of recipes being extracted from a database of recipes comprising a set of recipes each comprising a list of ingredients, each ingredient being associated with a list of macronutrients and micronutrients, said nutrients each being quantified for a recipe according to a value and at least one temporal datum quantifying a period during which the nutrients are present in the body, said extraction operation correlating the target values daily nutrient intakes with the recipe base to produce a recipe list for a plurality of days. One advantage is that it automatically determines a food program that meets daily target intakes which can be smoothed over a predefined period.

According to one embodiment, the recipe base is filtered from a selection of predetermined ingredients, said recipes generated in the list not including the filtered ingredients.

According to another aspect, the invention relates to a computer program product comprising a computer and a memory, said program comprising program code instructions executed on a computer for carrying out the steps of the method of the invention.

According to another aspect, the invention relates to a system comprising at least a computer, a memory and an interface for implementing the method of the invention. The system of the invention advantageously comprises at least one terminal or a user's computer for recording the data acquired by means of an interface. In addition, the system of the invention advantageously comprises at least one communication interface for transmitting the acquired data to at least one remote server. The system comprises at least a remote database and a remote computer, such as that of a remote server, making it possible to perform operations aimed at generating personalized phenotypic and genotypic indices, target values of daily intakes of a plurality. of nutrients and a composite nutritional index.

According to one embodiment, the system of the invention comprises a memory for recording a repository comprising at least predefined data of thresholds, ranges of values, scale of values and predefined calculation rules, the system comprising, in furthermore, an interface for acquiring data of a first and a second set of data of at least one individual and a memory for storing said data, the system comprising a computer for executing a set of rules and an engine inference to produce a composite nutritional index from the method of the invention, the system also comprising a display to display said composite nutritional index.

Other characteristics and advantages of the invention will emerge on reading the detailed description which follows, with reference to the appended figures, which illustrate: Fig. 1: the main steps of an embodiment of the method of the invention;

FIG. 3: an example of processing of the data acquired from the method of the invention in order to generate normalized input data at the inputs of the inference engine,

Fig. 3: an example of an architecture of the system of the invention.

Definitions

In the present description, a phenotype comprises all the objective and quantifiable data of an individual, such as his age, his weight, his height, etc., and the state data which can be collected and processed by the method and the system of the invention. Condition data includes data describing physiological signs such as symptoms and data relating to an individual's physiological state or physiological activity. Data on physiological status or physiological activity may result from the acquisition of information aimed at describing an individual's habits, a lifestyle and / or a lifestyle. Phenotype data is subject to change over time and can be updated during the process to generate an updated composite nutritional index. The nutritional index is therefore for a given subject an index dependent on the "time" variable.

We call a phenotypic descriptor, an objective datum or a state datum of a phenotype which can be quantified. The quantification operation can be simple when the descriptor is already dealing with quantified data of an individual such as anthropometric data. Among these data we find for example an age, a weight, a height. The quantification operation can be more complex and result from a mathematical operation, such as the calculation of the body mass index, noted BMI. This usually manipulates basic phenotype data already quantified itself, for example height and weight regarding BMI. Finally, the quantification operation can correspond to an operation aimed at qualifying and measuring on a predefined scale the manifestation of a physiological sign relating to a physiological activity, or to a biological sign of an individual. The phenotypic descriptors can also include one or more level (s) of biomarkers taken from different biological fluids of an individual or more broadly a biological constant measured or acquired from an individual. The quantification of the descriptors is possibly modified over time by various measurements carried out successively. Thus, a variation of a quantization of a descriptor can also constitute a new quantization of a descriptor. As an example, a drop in weight can be quantified. The decrease in blood pressure over time is also to be quantified. Other characteristics corresponding to variations of a quantization can constitute quantized descriptor quantities. The variations are preferably quantified over predefined periods of time.

When the quantification of a value is normalized on a predefined scale of values, it is also referred to as a "qualified value".

Data characterizing phenotypic descriptors are specially formatted to be stored in physical memory. The memory can be, for example, a database whose architecture makes it possible to use the data characterizing these descriptors. Thus, data such as weight, BMI, age, height, gender, a biological constant, biomarker values, etc., can preferably reflect the architecture of the database used so that these data can be extracted and used. during calculations by one of the steps of the method of the invention.

A phenotypic index, denoted IPN, is called a standardized value of a quantification of one or more phenotypic descriptors, whether combined or not. The phenotypic indices are calculated from a scale of predefined values. The scale of values is defined in a given repository of a knowledge base. Said scale of values is generally associated, within the framework of the invention, with conditions aimed at discriminating, by means of an inference engine, the result of at least one rule applied to a phenotypic index. When different phenotypic descriptors are grouped together within the same phenotypic index, one or more rule (s) can be applied to generate a score of the phenotypic index according to a plurality of quantifications of the descriptors. phenotypic. In some special cases, the phenotypic descriptor and the phenotypic index may be the same. In the present description, a genotype comprises all the data describing all or part of the genetic capital of an individual. More particularly, in the context of the implementation of the invention, the sequence of a selection of genes of an individual is exploited. In the context of the invention, the genotype can therefore designate the sequence of a selection of genes as well as a description of said genes referencing in particular variations or genetic mutations relating thereto.

A genotypic descriptor is called a datum of the genotype which can be quantified or qualified. The quantification process may include identifying or not identifying genetic mutations, such as genetic variations, and enumerating them.

The genetic mutation can be point or related to a variation in the sequence of a gene involving several phenomes. Indeed, a variation in the sequence of a gene can be of different types, in particular:

^■ mutation by substitution, corresponding to a replacement of one (or more) nucleotides by another (or more others);

^■ insertion mutation, corresponding to the addition of one or more nucleotides;

^■ mutation by deletion, corresponding to a loss of one or more nucleotides

^■ inversion mutation, corresponding to a permutation of two neighboring deoxyribonucleotides.

A genotypic descriptor can also qualify and quantify a polymorphism in which a variability in the number of copies of the same gene or of a chromosomal segment in the genome is observed, also known in English terminology as "copy number variation", whose acronym is CNV.

The data characterizing genotypic descriptors are specially formatted to be recorded in a physical memory. The memory can be, for example, a database whose architecture makes it possible to use the data characterizing the descriptors. Thus, the mutations, their number, their type, etc., can preferentially reflect the architecture of the database used so that these data can be extracted and used during calculations by one of the steps of the method of the invention.

A genotypic index, denoted IG _N , is called a standardized value of a quantification of a genotypic descriptor, and it can therefore be a qualified value. The genotypic indices are calculated, for each copy of the genome present in each individual, from a scale of predefined values relating to predefined states such as a “mutated” state or a “wild” state of a gene. According to one embodiment, in addition to the determination of a “mutated” or “wild” state, the genotypic index can integrate the more or less deleterious consequence on the physiology of the individual. The scale of values is defined in a given repository of a knowledge base. Said scale of values is generally associated, within the framework of the invention, with conditions aimed at discriminating, by means of the inference engine, the result of at least one rule applied to a genotypic index.

In some special cases, the genotypic descriptor and the genotypic index are identical.

An inference engine, computer program, software or application implementing a deductive reasoning simulation algorithm is called. Deductive reasoning can in particular generate convergent conclusions of results at the output of the inference engine from, on the one hand, a fact base comprising data of individuals and, on the other hand, a database of facts. knowledge comprising a set of rules, conditions and a repository of predefined values. Among the known inference engines are engines like CLIPS, GEOMETRIX, PROLOG, KADVISER, etc. The invention is not limited to a given implementation of an inference engine. Any type of inference engine compatible with operations implemented by applying rules applied to N inputs with P outputs can also be implemented in the method of the invention.

In the remainder of the description, a repository is a subset of knowledge base data comprising all the reference data as well as predefined rules.

This data may include, for example: ^■ predefined thresholds, so as to produce one or more comparables when a rule aims to compare an input variable with a threshold,

^■ limit values delimiting the definition framework of a variable,

^■ ranges of values defining, for example, a scale of values to determine a standardized value,

^■ reference daily intake values when they are used directly,

^■ absolute values making it possible to compare phenotypic or genotypic index values.

The predefined rules of the knowledge base notably include rules that can be played for:

^■ quantify phenotypic descriptors from responses to questionnaires or;

^■ consolidate physiological or biometric values from predefined calculation rules or,

^■ normalize values of phenotypic or genotypic descriptors, for example, which compile different state values.

Finally, other rules from the repository are used when running the inference engine and are applied to the fact base data relating to a new topic.

Certain so-called reference rules make it possible to calculate reference values for each new individual from the reference system. Other rules, called targets, make it possible to calculate target values for each new individual from the repository and the inference engine.

It is then possible to generate a composite index representing a target value associated with a reference value.

Profile, Phenotype

The method of the invention makes it possible to receive a first set of data ENSi and a second set ENS ₂ of data from a user. The first ENS ₁ data set characterizes data specific to the user's _{PHE 1 phenotype.} The second ENS _{2 set} of data characterizes data specific to the user's _{GEN 1 genotype.} Those ENSi and ENS ₂ datasets can be acquired in a single step. The ENS ₁ and ENS ₂ sets are functionally differentiated, for example for the processing of the data acquired, but may be inseparable for the user who will deliver these data from an interface of a system such as that generated by a computer.

The first set of ENS ₁ data comprises data which can be entered via a man-machine interface, called an HMI, making it possible to configure data specific to the user. According to one example, certain data of the ENS _{1 set} are automatically acquired from a communication interface making it possible to receive data and to decode them. According to one example, data is automatically acquired from a connected object. We can cite, for example, a connected watch, a connected blood pressure monitor, a connected scale or even a connected glucometer. Different acquisition modes can possibly be used in the context of the invention. For example, data can be received from a wired or wireless interface. The data is possibly stored on a remote server. The data can alternatively be stored in a memory of the same equipment carrying out the calculations making it possible to carry out the steps of the method of the invention.

Among the data of the ENS _{1 set} describing the PHE ₁ phenotype data of a user, there is for example data of profiles of an individual such as his age or his gender, namely his gender / sex {FEMALE, MAN }, its size, or possibly characteristics of culture or place of residence. The data of the ENS ₁ set also includes anthropometric and physiological data of an individual such as his weight, or one or more concentrations of biomarkers taken from different biological fluids or more broadly any biological constant measured or acquired from an individual. The data from the ENS _{1 set} may also include data characterizing muscle mass, heart rate, data from an electroencephalogram, fat mass, water level, bone mass, blood sugar, cholesterol, triglycerides, etc.

According to one example, the method of the invention makes it possible to categorize data of phenotypes into a subset so as to facilitate data acquisition and use. The ENSi set can, for example, comprise data of behavioral phenotypes, data of emotional or stress phenotypes, data of biological phenotypes, data of physiological phenotypes. Genotype

The second ENS ₂ data set comprises data collected after an analysis of the genome of a selection of genes of an individual, the genotype.

When this data is digitally accessible, a system communications interface is configured to receive the data in digital format. According to an alternative, the data of the second set are entered via an interface of the system.

According to an exemplary embodiment, a selection of 20 to 30 genes is processed so as to extract information characterizing the individual state of each gene. In a preferred mode, a set of data is extracted, said data characterizing each gene of the selection. A first datum G ₁ enumerating the genes having at least one mutation or one variation is used. A second piece of data G ₂ can also be used when it characterizes a number of modifications of a gene, that is to say a number of mutations or variations of a gene. In this case, it is considered that a modification of a gene corresponds to a modification of generic information, that is to say to a variation of the sequence. Finally, a third piece of data G ₃ describing the type of mutations or variation is possibly used according to the embodiments of the invention. According to one embodiment, rules comprising conditions on a set of mutations or variations of a set of genes make it possible to produce a fourth additional datum G _4.

Rules of the knowledge base are configured in such a way as to produce scores according to the values of the first, second, third and fourth data G ₁ , G ₂ , G3, G ₄ of a gene or a group of genes of the selection.

The method of the invention makes it possible to verify the type of genetic mutations such as substitution, insertion or deletion mutations or the variability of the number of copies of the same gene or of a chromosomal segment. According to one embodiment, the invention comprises also an analysis of types of mutations such as duplication, translocation or inversion. The method of the invention also makes it possible to take into account the presence of genes in the homozygous or heterozygous state of an individual and the number of associated mutations. According to an exemplary embodiment, the genotypic indices comprise a standardized quantification for each point mutation considered among the following three cases:

^■ 0 mutation, also noted wild / wild homozygote or in the Anglo-Saxon terminology wild-typ and wild-type), · 1 mutation, also noted wild / mutated heterozygote or in the Anglo-Saxon wild-type / mutant terminology,

^■ 2 mutations, also noted homozygous double mutated).

This quantified value can be associated, for example, with a quality of the mutation such as a variation of the polymorphism type of a single nucleotide, called PSN, or even a deletion, an insertion or a variability in the number of copies.

The method of the invention comprises a step making it possible to describe these data characterizing a genotype of a predefined selection of genes of an individual. To this end, genotypic descriptors make it possible to quantify the various data Gi, G2, G3, G4, etc., or even to generate a score characterizing this genotype information as a function of the knowledge base.

The method then comprises a step aimed at generating a genotypic index IG i. The IGi genotypic index is used in particular as input data in the inference engine in the same way as the IP phenotypic indices. The IGi genotypic index is a standardized value that can be exploited by rules of the inference engine.

The passage from genotypic descriptors to phenotypic indices can be ensured by the application of predefined rules, said rules being applied before the execution of the inference engine. For example, the presence of one or more mutations in the same individual allows, through the application of one or more rules, to modulate the phenotypic profile of the latter.

Rules The method of the invention comprises a database of the knowledge base, in which rules are recorded and possibly updated. The rules define operations making it possible to check conditions on the data of the sets of the first set ENSi and of the second set ENS _2.

Certain RREF rules are applied to calculate daily intake of NUT nutrients, and individual VIR reference values from other RCIB rules are applied to calculate VCIB target values of daily NUTi nutrients that correspond to a personalized intake for a individual according to the data of the first and second set ENS ₁ , ENS _2.

According to a first case, certain RCIB rules making it possible to calculate VCIB target values are defined on the basis of RREF rules and an enrichment of said RREF rules. An enrichment includes, for example, taking into account an additional factor or a weighting coefficient calculated from the inclusion of a phenotypic descriptor and / or a genotypic descriptor.

In order to illustrate this example, the method of the invention can be configured to define an RCIB rule which consists in applying an incremental or subtractive value to a reference score either of a phenotypic descriptor, ie a genotypic descriptor, or to a. phenotypic index, ie a genotypic index. The incremental or subtractive value is for example considered when at least one condition is verified.

According to one example, this incremental value may correspond to an additional percentage of a daily reference nutritional intake. VCIB = 1, 2-VIR is obtained when an additional daily intake of a nutrient must be 20% greater than a reference value VREF of this intake. In this example, the personalized daily intake, called target, is calculated independently of other phenotypic and genotypic index values at the input of the inference engine.

The method of the invention makes it possible in particular, according to another variant, to consider a modification of the phenotypic or genotypic index as a function of a reference value of the index considered, for example by incrementing this value. Said new phenotypic index value is then taken into account by the inference engine which can converge towards a new daily intake value VCIB for a class of nutrients. In the latter case, there is not systematically an independent cause and effect relationship between the value of the phenotypic index and the value of a daily intake of a nutrient. Moreover, according to certain embodiments, the method of the invention comprises scenarios in which the values of reference phenotypic and genotypic indices are calculated from a subset of phenotypic or genotypic descriptors. These benchmarks can be used to calculate daily intake values of benchmark nutrients. One contribution of the invention is the use of an inference engine to calculate target values of daily nutrient intakes which differ from reference values of daily nutrient intakes.

According to one embodiment, descriptors are used to quantify other phenotypic or genotypic descriptors. We are talking about consolidation rules. In fact, in the latter case, the quantified values of the descriptors come from a calculation and are therefore consolidated insofar as the value used of the phenotypic descriptor or of the phenotypic index is calculated from an operation manipulating data acquired from a subject.

According to an example, if you are a man over 18 years old, an individual reference value VIR (GR, AGE) of a given nutrient NUTi, such as the daily intake of Chromium, is calculated by a function of these two parameters: Age: AGE and gender: GR. Numerical charts, ranges of values or thresholds pre-stored in a memory can be used to apply the predefined rules.

The calculation of a VCIB target value, corresponding to a daily intake of a nutrient and therefore to the individual VIR reference value of the same nutrient, can take into account a larger panel of phenotypic and genotypic descriptors. According to the example for which a reference value has been previously calculated, the method of the invention makes it possible to calculate a personalized VCIB target value VCIB (GR, AGE, BMI, OB, SPO, CHLO, ANX) = VIR +/- 10.

In this last example, GR denotes gender, AGE denotes age, BMI denotes body mass index, OB denotes a category of obesity, SPO designates a quantification of daily sporting effort, CHLO designates a quantification of a cholesterol level, ANX designates a quantification of an individual's level of anxiety. These variables are descriptors or phenotypic indices depending on the case. In this example, the inference engine makes it possible to take into account all the parameters and to play them in order to determine a value VCIB which results from a combination of causes quantified by the aforementioned descriptors.

According to a second case, the RCIB rules are defined by an algorithm which makes it possible to calculate a VCIB target value from the phenotypic and genotypic indices. This algorithm can be set so that the target value does not result directly or indirectly from an RREF rule but from a new rule to calculate individual values of VIR references. For example, an RCIB rule corresponds to the generation of a fixed value of a daily intake of a nutrient when one or more condition (s) is / are satisfied. The condition is, for example, taking into account a parameter of the phenotype or genotype of the individual.

To illustrate this example, an RCIB rule involves applying a fixed value to a daily amount of a given nutrient when an individual's phenotype indicates that the individual is a pregnant woman. In the latter case, the "pregnant woman" parameter corresponds to a phenotypic descriptor or a phenotypic index making it possible to generate a condition on the result of a calculation rule making it possible to generate a value for a daily intake of a nutrient. The method of the invention is based in particular on a preliminary configuration step aimed at determining quantifications of ranges of values of phenotypic and / or genotypic descriptors. These quantifications make it possible to define the conditions of association of the results obtained by the application of RREF OR RCIB calculation rules to calculate individual values of VIR references OR VCIB target values.

These quantifications are preferably normalized by the quantization transposition of the ranges of values from the descriptors to the indices in order to homogenize a processing based on different rules using an inference engine. Quantification of phenotypic parameters According to an exemplary embodiment, a first family of quantifications of phenotypic descriptors is calculated from values acquired or received and directly exploited from an individual. The height and / or the weight and / or the waist circumference of an individual can be examples of descriptors of this family of quantification. These values are in fact directly exploited by rules as a function of other predefined values from the repository. For example, the individual is a man with a BMI greater than a first threshold and having a cholesterol level greater than a second threshold, then a rule makes it possible to define a value with respect to a list of 'daily intake of a subset of nutrients.

According to another example, a second family of descriptor quantifications results from the acquisition of a set of physiological states acquired from an individual which are processed so as to generate a physiological quantification. This physiological quantification can be used to generate a quantification of the descriptor or a quantification of a phenotypic index to be exploited by other calculation rules of the inference engine. By way of example, a quantification of a physiological state of fatigue can be generated from a list of answers to predefined questions comprising, for example, a quantification of a volume of sleep, an evaluation of a stress. daily, an evaluation of a daily physical effort, a quantification of a level of drowsiness, etc.

According to an exemplary embodiment, the values quantifying certain physiological states such as stress are normalized on a scale of values comprising a number of predefined values to generate a phenotypic index that can be used by the inference engine.

Example of a quantification of a sports activity According to an example, a list of questions makes it possible to evaluate, on a scale, a composite score of a physiological state. Each answer to a question increases the composite score. For example, a first question generates an evaluation from 0 to 10, a second question generates an evaluation from 11 to 20 and so on up to the tenth question which generates an incremental evaluation of 91 to 100. The incremented composite score has, for example, a value of 56/100. Such a composite score can then be normalized on a scale of 0 to 3 or a scale of 0 to 5 depending on the range of values defining the phenotypic index scale. An advantage is to obtain a normalization of the physiological states which can then be treated by a homogenized treatment in the inference engine. Another advantage is to be able to dynamically compare the evolution of the calculated score. According to one example, the evaluation of a physiological state relating to the sporting activity can comprise a list of questions making it possible to evaluate the type of sport, its frequency, its intensity, etc. In order to determine an individual's sporting activity factor, to quantify it and to standardize it, the method comprises a step aimed at transposing the scores of the descriptors within a predefined scale. The level of sporting activity can be quantified, for example, among 3 values of a standardized scale.

According to another example, a quantification of the psychological stress of an individual, a quantification of the health of an individual, a quantification of the feeling of well-being and the morale of an individual, a quantification of fatigue, a quantification of the quality of memory and concentration and quantification of oxidative stress can be assessed. Assessments include a method of calculating a score, for example, an incremental method. This method offers the possibility of taking into account different factors cumulating their quantified effects. A normalized value can then be calculated to determine an input to the inference engine.

Inference engine

According to one embodiment, the calculation of a given VCIB target value of a daily intake of a nutrient is obtained from the application of a set of rules applied to a set of values quantifying phenotypic indices and / or genotypic of an individual. In such a case, a plurality of rules aim to carry out operations validating or invalidating conditions, such as exceeded thresholds, lowered thresholds, values included in predefined ranges or even state value control, etc. Conversely, the result of an applied rule can influence a plurality of daily intakes of a given set of nutrients.

The model of the invention results in a model comprising a number N of inputs comprising the processing of a plurality of values quantifying the phenotypic and genotypic indices of an individual and generating at the output of this model has a number P of outputs corresponding to a plurality of daily inputs of individualized nutrients.

According to one embodiment, the invention comprises the application of an optimization function aiming to determine one or more application sequences of the inference engine in order to obtain VCIB target values of the various personalized nutritional daily intakes converging in one. time and a limited computational cost.

According to one example, the rules can be applied in a given order and the values obtained from the daily intake of a given nutrient result from the application of said set of rules according to the first order of application of the rules. In order to infer a target value from a plurality of daily intakes, according to an exemplary embodiment, all the rules are applied according to a second schedule. The set of rules making it possible to calculate the VCIB target values of the daily nutrient intakes can then be reapplied a certain number of times until a set of VCIB target values optimized with respect to each other is inferred with regard to the 'set of rules taken into account. The set of VCIB target values generated is in this case the result of a convergence of a set of applied rules so that the effects of rule dependencies between them on the calculated target values are minimized.

Different optimization functions can be configured to define an inference engine. The use of an inference engine is based in particular on a fact base and a knowledge base. The knowledge base corresponds to the database comprising rules and conditions for determining membership criteria for the calculated values. The fact base corresponds to the entries corresponding to the quantified values derived from the phenotypic and / or genotypic descriptors of an individual.

The inference engine makes it possible to take into account, for example, different quantifications of indices derived from phenotypic and / or genotypic descriptors derived from different physiological and genetic parameters to determine daily intakes of macronutrients, such as lipids, proteins and carbohydrates. . The various physiological parameters influencing the distribution value of the daily intakes of macronutrients are taken into account in the inference engine in order to converge towards optimized daily intake values.

At the output of the inference engine, a list of daily intakes of nutrients, macro- and micronutrients, is generated. The invention allows these different inputs to be grouped together by major metabolic functions. Thus a first metabolic function comprises in particular energy metabolism, a second function relates to lipid metabolism, a third function relates to the metabolism of amino acids, a fourth function includes the functions ensuring the oxidative balance and finally a fifth function relates to the functions relating to life quality.

This categorization is detailed below through exemplary embodiments, however other categorizations can be implemented in the method of the invention. The categories of major metabolic functions can in particular be organized and configured according to a given indication such as fertility, sleep, physical recovery and / or physical activity, etc.

The first metabolic function includes, for example, the metabolism of macronutrients such as carbohydrates, fats, proteins, vitamins B1 and B3, daily intake VCIB target values and individual VIR reference values for each of these nutrients.

The second metabolic function includes, for example, the metabolism of saturated fatty acids, oleic acid, amino acids, linolenic acid, omega-3s such as IΈRA or DHA, the VCIB target values of daily intakes and the individual VIR reference values for each of these nutrients.

The third metabolic function includes, for example, the metabolism of vitamins B2, vitamins B6, vitamins B9, vitamins B12, the daily intake VCIB target values and the individual VIR reference values for each of these nutrients. The fourth metabolic function includes, for example, the metabolism of vitamins A, vitamins C, vitamins E, selenium and zinc, daily intake VCIB target values and individual VIR reference values for each of these nutrients. The fourth metabolic function includes, for example, magnesium and vitamins D, the daily intake VCIB target values and the individual VIR reference values for each of these nutrients.

An advantage of this association between each subset of nutrients grouped together with a metabolic function is to provide an appropriate recommendation and benefiting from a synergy of effects and metabolic consistency. These grouped nutrient subsets can be reconfigured according to a given indication, that is, an indication of fertility, sport or another. In another configuration, these groupings and associations produce a different adapted synergy which produces recommendations adapted to the indication.

Finally, each subset can be displayed by means of a graphical interface so as to present composite indicators comprising the individual reference value VIR with which the target value VCIB is associated on a common scale of values. One interest is to represent for a given individual a personality recommendation vis-à-vis a non-personalized reference value.

Food recipes

The invention allows from the composite indicator IC thus generated to offer a set of recipes to an individual. "Recipe" means a dish or a set of ingredients making up the dishes.

According to one embodiment, the system of the invention comprises a database of recipes. The recipes each include, for example, an identifier, a name, a list of ingredients, each ingredient is associated with its nutritional value according to said ingredients, the quantity of ingredients and possibly the preparation such as the cooking method. Each ingredient for a given amount can be segmented into a quantified list of macronutrients and micronutrients as well as the calories it contains or provides. Knowing the nutritional composition of a plurality of ingredients and therefore of recipes, the method of the invention makes it possible, on the basis of a daily nutritional recommendation, to generate, conversely, a set of recipes for an individual.

The method of the invention comprises a calculation step aimed at optimizing the distribution of revenue over a given period, for example a period corresponding to "the week". One advantage of this optimization is that it “smooths out” the nutrient supply of food according to the recipes, that is to say to distribute the nutrient supply over time over a given period. For example, if an iron intake is recommended for 10 mg / day for a man according to his phenotype and genotype, according to the period of digestion, absorption, assimilation and persistence in the body of iron, recipes can be split to ensure an average weekly intake corresponding to a recommendation for the week, not just the day. This is also the case for certain nutrients / foods having more or less long persistence in the body such as a fat-soluble vitamin or another water-soluble.

In another example, the intake of vitamin C, a water soluble vitamin, cannot be stored, so it persists for a shorter time than minerals such as iron or other fat soluble vitamins. The recipes indicated then include ingredients that make it possible to provide, on a day-to-day basis, as close as possible to the desired daily nutritional recommendation, an adapted recommendation. On the other hand, like fat-soluble vitamins, the intake of recipes for these nutrients is smoothed over periods of up to several days, over this period the average daily intake remains that calculated in the previous step.

The BDr recipe database therefore includes specific data determining for each nutrient the length of time during which the daily nutrient intake can be considered to generate dish recommendations over a given period. On the other hand, the nutrient intake of each recipe can be determined automatically from a reference database. For example, the French reference base is the CIQUAL database updated by the National Agency for Food, Environmental and Occupational Health Safety (ANSES). Several countries are developing and maintaining their own baselines. One or more reference databases can be used concomitantly to automatically calculate the cumulative quantitative contributions of each ingredient contained in each recipe according to the countries or cultural preferences. This calculation can be carried out for example by taking into account one or more weighting coefficients of the quantities of macroelements. Others criteria can be taken into account. According to another embodiment, the taking into account of a country can be implemented according to another technique.

Individualization of dietary constraints

The method of the invention makes it possible to take into account certain constraints or eating habits of a subject. One interest is to filter ingredients and therefore recipes containing them in the generated recommendation. Dietary constraints can be due to an allergy, intolerance, an individual's taste or be cultural. Thus, the recommended recipes include not only a personalized daily nutrient allowance but also ingredients or groups of ingredients filtered so that all recipes can be consumed by the individual.

The method of the invention therefore makes it possible to use the database of recipes with data stored in a memory corresponding to the personalized profile of an individual in order to generate proposals for a list of recipes taking into account the constraints and eating habits of an individual. topic.

Ingredients with a given nutritional value and being filtered due to a dietary constraint are automatically replaced by one or more other ingredients so as to provide a new dish that the subject can consume. The method of the invention may include a step of filtering all the recipes in the BDr recipe database comprising ingredients that are discarded by a subject. The recommended nutrient intakes are then provided by recipes that have not been filtered. Coverage index

According to an exemplary embodiment, the selection of proposed recipes comprises an indicator of coverage of the daily nutritional recommendation aimed at informing an individual if the recipe is in phase with the composite index generated. An advantage is to provide a list of alternative recipes that are less effective than the recommended main recipe (s). This possibility gives an individual a choice of alternative recipes that are less optimized but offer a wider choice. According to one example, the user of the interface can choose, using the interface, recipes covering 80% of the recommended daily allowance or 90% or even 100% for a strict diet. In one example, this is a third-party user, such as a doctor or a nutritionist, who accesses the recipes from the interface and their coverage rate vis-à-vis the recommendation for a given patient.

Food stalemate

When a recommended nutritional intake cannot be ensured by the ingredients of a recipe in the database or when too few recipes correspond to a given intake of a nutrient, the method of the invention makes it possible to calculate , for an individual, a supplementation of nutritional supplements. Supplementation can be calculated quantitatively and qualitatively nutrient by nutrient. The supplementation is personalized in particular by a selection of supplements for a given individual. According to one example, the recommended nutrient intake can be considered "uninsured" when this intake is not assured over an average amount of daily intakes over a given period. This is the case with vitamin D which can be supplemented to fill a deficiency over a given period. The individualized composition of the supplements are determined based on a nutrient deficit in recipes or to ensure variability in dishes to supplement the supply of a given nutrient for an individual at a given time.

Embodiments of the invention Figure 1 shows the main steps of an embodiment of the method of the invention. An individual SEL selection step is performed. The selection step includes at least an identification of the subject. Identification can be made by means of an interface allowing the selection of a name, an identifier or any other data digitally attached to an identifier of an individual. The selection of an individual includes, for example, the selection of a digital profile describing the digital characteristics of a predefined individual. The method of the invention is of particular interest due to the generation of a list of personalized daily nutrient intakes. Consequently, the method is preferably carried out for a single individual at a time.

A first ACQi acquisition step is carried out to collect the phenotype data of the selected individual. A second ACCh acquisition step is carried out to collect the genotype data of an individual. These steps advantageously make it possible to collect data organized in the form of a descriptor in order to quantify the data of phenotype and genotype. According to another example, the descriptors are a preliminary calculation step to the determination of the phenotypic and genotypic indices. Acquisitions can be made at the same time or at different times. The method includes steps for generating phenotypic and genotypic indices of an individual from the descriptors. It happens that the values of the descriptors correspond to those of the indices. This is the case when the acquired or processed digital values are already normalized. The phenotypic and genotypic indices are the input data for an M1 inference engine.

The method comprises a step of determining a composite nutritional index DETJNCi to produce a composite nutritional index INCi. This index includes a list of daily nutrient intakes including macronutrients and micronutrients. The step of determining the INDCi composite nutritional index is performed using an inference engine.

Figure 2 details an example case in which the first ENSi comprises four physiological states E11, E12, E13, E14. Each of these states represents a part of a phenotype noted PHEi of the individual Ui. The four states can be processed by rules from a knowledge base of a REFi repository in order to calculate scores associated with the states. The scores are noted Score (E11), Score (E12), Score (E13), Score (E14). In this example, the score calculations from the states are performed in a block denoted DESCi in Figure 1. In this example, the Score (E11) and Score (E12) scores are used to calculate a phenotypic index IP1 (E11, E12), and the scores Score (E13) and Score (E14) are used to respectively calculate the indices IP2 (E13) and IP3 (E14), for example from a rule for calculating the knowledge base. The calculations of the phenotypic indices are carried out in a block denoted NORMi in FIG. 1. It is then understood that scores of descriptors can be combined to produce phenotypic indices such as IP1.

The values of the phenotypic indices are thus transmitted to a block denoted M1 corresponding to the algorithm configured to define the inference engine. FIG. 1 comprises another data processing chain for receiving and processing the data of the ENS _{2 set} corresponding to the genotype data GEN ₁ of the same individual.

FIG. 1 comprises a first block denoted DESC ₁ aimed at acquiring data of genotype G1, G2, G3 and G4. The genotype data is then processed to calculate Score (G1), Score (G2), Score (G3) and Score (G4). It can be, for example, the number of mutations of a gene, the type of mutation of this same gene, etc.

A second block denoted NORM1 makes it possible to calculate genotypic indices IG, to define inputs of the inference engine M1. When the Scores (Gi) already correspond to a standardized score, the IG genotypic index may be identical to the score previously calculated.

The inference engine M1 is then applied to all the input data, that is to say the phenotypic and genotypic indices by means of rules defined in the reference reference REF _1. The inference engine M1 makes it possible to determining convergent VCIB target values of the target daily nutritional intake of an individual.

FIG. 3 represents an example of the architecture of the system of the invention. The repository is denoted REF ₁ in FIG. 3. It comprises at least one database or a file comprising predefined thresholds, ranges of predefined values, predefined rules as well as scales of values making it possible to standardize the scores. It can also include reference values such as reference values of descriptors or indices or else individual values of daily intake VIR references for typical profiles. The reference frame REF ₁ is used by a calculation block denoted Ki. The calculation block K ₁ can include one or more calculator (s), such as microcontrollers, microprocessors or any other means making it possible to perform digital calculations. The calculation block K ₁ makes it possible to quantify the descriptors from a first level of rules, to normalize these values to generate indices from a second level of rules and to play the inference engine from a third level of rules. The inference engine is here denoted Ml.

According to one embodiment, a user database, denoted DBu, makes it possible to select a user U ₁ recorded in said based. Each user profile Ui can be associated with the PHEi phenotype and the GENi genotype of said user Ui after the acquisition of these data.

One advantage of capitalizing a BDu user base is in particular to generate a statistical engine, for example, from artificial intelligence. Such a configuration makes it possible to establish metrics over time specific to the success or failure of recommendations based on the calculation of a composite nutritional index of a set of individuals. The recommendations for a new individual can thus be modulated or adapted according to (efficiency) the success or failure rates of a set of profiles close to said new profile.

According to one embodiment, the composite nutritional index INDCi is calculated at different periods which may or may not be regular. The advantage of calculating the INDCi composite nutritional index at different times is to adapt the individualized recommendations by taking into account the changes in conditions that may occur as a result of the follow-up of the recommendation by the individual Ui.

A second calculator K ₂ is represented in FIG. 3. It makes it possible to use a database of recipes BDr to produce a _{meal plan PLAN 1} in a period defined for a given user U _1. According to one embodiment, the calculator K ₂ can be the same as the computer K ₁ depending on the architecture of the system chosen.

The calculator K ₂ uses in particular a database of recipes BDr, a list of constraints and dietary habits IMP ₁ predefined by the user U ₁ , a database of dietary supplements (not shown) and rules making it possible to diversify the diet in time, that is to say to distribute the nutritional contributions over time scales depending on the specific data of absorption, assimilation and persistence in the body of food.

According to an exemplary embodiment, these food diversification rules over time can be integrated into the recipe base BDr in order to enrich the recipe data. These rules are denoted REP ₁ in Figure 3.

The dietary constraints are extracted here from the BDu user base storing data on user profiles and preferences. Here the constraints are noted IMP ₁ .

Claims

1. Process for generating a composite nutritional index (INCi) comprising:

^■ Selection (SEL) of an individual (Ui);

^■ Acquisition (ACQi) of a first set of phenotype data of said individual (Ui) characterizing phenotypic descriptors (PHE-i), said data comprising at least one age datum (AG), one gender datum (GR) and at least one set of data characterizing physiological signs (SP) of said individual;

^■ Acquisition (ACCh) of a second set of data (ENS2) of a genotype characterizing genotypic descriptors (GEN1) of said individual (Ui), said data comprising information characterizing mutations and / or variations of at least one uncomfortable ;

^■ Application by a calculator of a set of predefined rules (RCIB) comprising: o at least a first subset of rules (R1C _IB ) aimed at generating at least one phenotypic index (PI,) from a calculation of a score (Score (Ekp)) of a quantification of a phenotypic descriptor, said index being normalized; o at least a second subset of rules (R2CIB) aimed at generating at least one genotypic index (GI,) from a calculation of a score (Score (Gi)) of a quantification of a genotypic descriptor, said index being normalized;

^■ Generation of a set of personalized phenotypic and genotypic indices (PI ,, IG,) of an individual;

^■ Calculation of target values (VCIB) of daily intakes of a plurality of nutrients (NUT,) from the application of an inference engine configured from: o a knowledge base comprising a repository of predefined values of phenotypic and / or genotypic indices and at least one set of rules conditional (CR) applied to said predefined values of phenotypic indices (IPREF) and genotypic indices (IGREF) and; o a fact base comprising all the phenotypic and genotypic indices (PI ,, IG,) of said individual calculated from the data acquired,

^■ Determination of a composite nutritional index (INC ₁ ) comprising an operation of associating a selection of target values (VCIB) of daily intakes of all nutrients (NUT,) with at least one metabolic function (F _M ).

2. Method for generating a composite nutritional index (INCi) according to claim 1, characterized in that the method comprises the steps of:

^■ receiving a plurality of individual reference values (VIR) of a daily intake of a plurality of nutrients (NUT,);

^■ generation of at least one differential indicator representing for each nutrient an individual reference value (VIR) and a target value (VCIB) of daily intake of said nutrient (NUTi).

3. Method for generating a composite nutritional index (INCi) according to claim 2, characterized in that the individual reference values (VIR) of daily intakes of said nutrients (NUT,) are:

^■ directly extracted from a knowledge base referencing predefined daily intakes of nutrients and / or;

^■ calculated from reference rules (RREF) automatically calculating daily nutrient intakes from phenotype data (PHi) of said individual and predefined values referenced in a knowledge base.

4. Method for generating a composite nutritional index (INCi) according to any one of claims 1 to 3, characterized in that the step of determining a composite nutritional index (INCi) comprises a plurality of groupings of target values ( VCIB) daily intake of nutrients, each grouping contributing to improving a given metabolic function of said individual.

5. Method according to any one of claims 1 to 4, characterized in that the nutrients are macronutrients or micronutrients, said macronutrients being associated with a metabolic function quantifying an energy supply of said individual.

6. Method according to claim 5, characterized in that:

^■ at least a reference value (VREF) of a daily intake of a total energy amount of at least one macronutrient is calculated for an individual from a first set of phenotypic descriptors having an age, a gender and;

^■ at least one target value (VCIB) of a daily intake of a total energy amount of said macronutrient is calculated for said individual from a first set of phenotypic descriptors having an age, a gender, and a second set of phenotypic and / or genotypic descriptors.

7. Method according to any one of claims 2 to 6, characterized in that:

^■ at least one reference rule (RREF) comprises, for the calculation of a reference value (VREF) of a daily intake of a given nutrient (NUT _D ), at least one operation handling a first set (ENSi) quantifications of phenotypic descriptors (PHEi);

^■ at least one target rule (RCIB) comprises, for the calculation of a target value (VCIB) of a daily nutrient intake (NUT _D ), at least one handling operation, in addition to the first set (ENSi) of quantifications of phenotypic descriptors

(PHEi), a second set (ENS2) of quantifications of phenotypic and / or genotypic descriptors (PHEi, GEN1).

8. Method according to any one of claims 2 to 7, characterized in that: at least one reference rule (RREF) comprises, for the calculation of a reference value (VREF) of a daily intake of a nutrient (NUT), at least one operation considering a first set of quantifications of phenotypic descriptors ( PHEi); at least one target rule (R _CIB ) comprises, for the calculation of a target value (V _CIB ) of a daily intake of the nutrient (NUT), at least one operation aiming to define a fixed value of a daily intake of said nutrient (NUT) as a function of at least one threshold value reached by at least one quantification of a phenotypic and / genotype descriptor.

9. Method according to any one of claims 1 to 8, characterized in that a phenotypic descriptor is calculated from a sum of scores, each score quantifying a physiological state of the individual.

10. Method according to any one of claims 1 to 9, characterized in that a step of generating a list of recipes is carried out, said list of recipes being extracted from a database of recipes (BDr) comprising a set of recipes each comprising a list of ingredients, each ingredient being associated with a list of macronutrients and micronutrients, said nutrients each being quantified for a recipe according to a value (energy, mass) and at least one quantifying temporal datum (REPi) a period of time during which the nutrients are present in the body, said extraction operation correlating the target values (VCIB) of daily nutrient intakes with the recipe base (BDr) to produce a list of recipes for a plurality of days.

11. The method of claim 10, characterized in that the recipe base is filtered from a selection of predetermined ingredients, said recipes generated in the list not including the filtered ingredients.

12. System comprising a memory for recording a repository

(REFi) comprising at least predefined threshold data, ranges of values, scale of values and predefined calculation rules, the system further comprising an interface for acquiring data from a first (ENSi) and a second (ENS2) set of data from at least one individual and a memory for storing said data, the system comprising a computer (K1) for executing a set of rules and an inference engine in order to produce a composite nutritional index (INC1) from the method of the any one of claims 1 to 11, the system also comprising a display for displaying said composite nutritional index (INC1).