Method and apparatus for assisting in determining antioxidant dosage for individuals suffering from oxidative stress - Patent Application 20070122999

Abstract

The present invention relates to a method (100) for assisting in determining the dosage of an antioxidant to be administered to a patient suffering from oxidative stress, said method (100) comprising at least one iteration of a prediction step (108) comprising the steps of: measuring (110) the levels of at least the following input parameters for a blood sample previously taken from said patient: cholesterol level, zinc level, copper level, vitamin C level, vitamin E level, and selenium level; and estimating (130) the dosage of at least one antioxidant to be administered to said patient via a previously trained estimation model executed by a computation unit, said estimation model taking as input said at least one input parameter. The present invention also relates to a device implementing such a method.

Description

[Technical Field] 【0001】 The present invention relates to a method for assisting in determining the dosage of antioxidants to be administered to individuals suffering from oxidative stress, in particular individuals suffering from facioscapulohumeral dystrophy (FSHD), and also to a device for implementing such a method. 【0002】 The field of the invention is generally that of aiding in determining the dosage of antioxidants to be administered to individuals suffering from oxidative stress, and more specifically to individuals suffering from FSHD. [Background technology] 【0003】 Antioxidants are known to be used in the treatment of FSHD. Currently, the dosage of antioxidants administered to individuals with FSHD is standardized. There are no predefined rules for quickly determining the most appropriate dosage. Specifically, for each antioxidant, a standardized starting dose is administered to the patient. The patient's response to the starting dose is then monitored. Depending on the patient's response, the starting dose is adjusted by a predetermined increment, particularly a fixed value. This adjustment procedure is repeated as many times as necessary depending on the patient's condition. 【0004】 It is understood that this method of determining and adjusting the dosage of antioxidants administered to patients with FSHD is not optimized and does not allow for individual determination of the dosage most suitable for each patient. 【0005】 Furthermore, with current solutions, determining the appropriate dosage for a patient is a matter of trial and error, which can be harmful to the patient. 【0006】 Furthermore, current solutions require a significant amount of observation time to determine the appropriate dosage for the patient, and current solutions are time-consuming, not very ergonomic, or ineffective when the patient's condition changes, which can be monitored and quantified using scores such as the Brooke and Vignos score for FSHD, or more commonly the MFM score (which stands for "Motor Function Measure"). Summary of the Invention [Problem to be solved by the invention] 【0007】 SUMMARY OF THE INVENTION It is an object of the present invention to ameliorate at least one of the drawbacks of the prior art. 【0008】 Another object of the present invention is to propose a solution for determining in an individualized manner the dosage of antioxidants to be administered to patients suffering from oxidative stress, in particular FSHD. 【0009】 Another object of the present invention is to propose a solution for determining the dosage of antioxidants to be administered to patients suffering from oxidative stress, in particular FSHD, in a faster and more ergonomic way for the patient. 【0010】 A further object of the present invention is to propose a solution for determining in a more accurate way the dosage of antioxidants to be administered to patients suffering from oxidative stress, in particular FSHD. 【0011】 A further object of the present invention is to propose a solution for determining in a more effective way the dosage of antioxidants to be administered to patients suffering from oxidative stress, in particular FSHD. [Means for solving the problem] 【0012】 The present invention proposes to achieve at least one of the above-mentioned objectives by means of a method to aid in determining the dosage of an antioxidant to be administered to a patient suffering from oxidative stress, in particular to a patient suffering from FSHD, said method comprising at least one iteration of a prediction step comprising the following steps: - a blood sample previously taken from said patient, which contains: Cholesterol levels Zinc levels Copper level Vitamin C levels Vitamin E levels; and measuring the level of at least one input parameter of selenium level; - estimating the dosage of at least one antioxidant to be administered to said patient by means of a previously trained model executed by a computational unit, said model taking as input said at least one input parameter. 【0013】 The present invention therefore provides assistance in determining antioxidant dosages for individuals suffering from oxidative stress, more specifically FSHD, using a prediction model pre-trained and implemented by a computational unit, where the dosage is estimated as a function of at least one patient-specific input parameter, the values ​​of which are measured on blood samples from said patient, and the prediction model thus sets a dosage that corresponds to a standardized dosage, thus allowing a dosage to be estimated that is individualized for said patient, unlike current solutions that are not individualized regardless of the patient. 【0014】 Furthermore, dosage estimation according to the present invention is performed by an estimation model that is pre-trained for dosage estimation and that takes as input parameters measured in a patient's blood sample. Thus, the present invention makes it possible to estimate the dosage of antioxidant(s) for patients suffering from oxidative stress, in particular FSHD, in a faster and more ergonomic way for patients compared to current solutions, since the present invention does not require successive iterations to determine the correct dosage for the patient by trial and error. 【0015】 Furthermore, the present invention allows for determining a more accurate and more effective dosage of antioxidant(s) at the upper limit for each antioxidant in a healthy population for patients suffering from oxidative stress, in particular FSHD, since it is not based on the use of predetermined increments but on precise values ​​that depend on input parameters measured in patient blood samples. 【0016】 In a non-limiting manner, this solution can be adapted to patients suffering from condition-specific oxidative stress, where the intervals and types of vitamins and / or trace elements are specific to the condition being targeted. 【0017】 Preferentially, the estimation model estimates the dosage of several antioxidants. 【0018】 Preferably, the prediction model is capable of performing dosage estimation for at least one, and in particular for each, of the following antioxidants: - Zinc - copper - Vitamin C - Vitamin E, and - Selenium. 【0019】 In a preferred embodiment, the dosage may include values ​​for each of the parameters listed above. 【0020】 For at least one antioxidant, the prediction model can be trained to provide a value within a predetermined range of values ​​for said antioxidant, for example, the prediction model can be trained to predict a value within said range of values, such that the value provided can be any value within said range of values. 【0021】 Alternatively or additionally, for at least one antioxidant, the prediction model may be trained to select one of several predetermined discrete values ​​for said antioxidant, in which case the prediction model always returns one of these predetermined values. 【0022】 In some embodiments, for zinc, the prediction model can be trained to provide a dosage between 15 mg / day and 45 mg / day. In a non-limiting example, for zinc, the prediction model can be trained to select a dosage from the following predetermined discrete values: 15 mg / day, 30 mg / day, 45 mg / day. 【0023】 In some embodiments, for copper, the prediction model can be trained to provide a dosage between 0 and 0.3 mg / day. In a non-limiting exemplary embodiment, for copper, the prediction model can be trained to select a dosage from the following predetermined discrete values: 0 mg / day and 0.3 mg / day. 【0024】 In some embodiments, for vitamin E, the prediction model can be trained to provide a dosage between 125 mg / day and 500 mg / day. In a non-limiting exemplary embodiment, for vitamin E, the prediction model can be trained to select a dosage from the following predetermined discrete values: 125 mg / day, 250 mg / day, 500 mg / day. 【0025】 In some embodiments, the prediction model can be trained to provide a dosage between 125 mg / day and 500 mg / day for vitamin C. In a non-limiting exemplary embodiment, for vitamin C, the prediction model can be trained to select a dosage from the following predetermined discrete values: 125 mg / day, 250 mg / day, 500 mg / day. 【0026】 As mentioned above, the predicting step may further include measuring selenium levels in the patient's blood, and the estimation of the dosage by the predictive model may further be performed as a function of said measured selenium levels. 【0027】 Therefore, the selenium level is an additional input parameter that is given as an input to the estimation model. 【0028】 Selenium levels can be measured from a patient's blood sample. 【0029】 The predictive model can also provide the dosage of selenium to be administered to the patient. 【0030】 In the case of Selenium, the estimation model can be trained to provide values ​​within a predetermined range, for example, the estimation model can be trained to estimate a value within said range of values, such that the value provided can be any value within said range of values. 【0031】 Alternatively or additionally, for Selenium, the estimation model can be trained to select one of several predetermined discrete values, in which case the estimation model always returns one of these predetermined values. 【0032】 In some embodiments, for selenium, the estimation model can be trained to provide a dosage between 0 and 200 pg / day. In a non-limiting exemplary embodiment, for selenium, the estimation model can be trained to select a dosage between the following predetermined discrete values: 0 and 200 μg / day. 【0033】 In some embodiments, the predicting step may further comprise measuring iron levels in the patient's blood, and the estimation of the dosage by the predictive model is further performed as a function of said measured iron levels. 【0034】 In this case, the iron level is an additional input parameter that is given as an input to the estimation model. 【0035】 Iron levels can be measured from a patient's blood sample. 【0036】 Vitamin C helps the human body absorb iron. Therefore, the vitamin C dosage should take into account the patient's iron level as a priority, so as not to excessively increase the patient's iron level. Therefore, taking into account the patient's blood iron level, it is possible to adjust the vitamin C dosage. 【0037】 The method according to the invention may further comprise measuring the level of diabetes in the patient's blood, and the estimation of the dosage by the prediction model is further carried out as a function of said level of diabetes. 【0038】 The level of diabetes can be measured by measuring the level of glycated hemoglobin. 【0039】 The diabetes level can be measured from a patient's blood sample. 【0040】 A measurement of the diabetes level can be performed before the first iteration of the prediction stage to determine whether the patient has diabetes. 【0041】 Alternatively or additionally, diabetes levels can be measured during each iteration of the prediction step. 【0042】 Determining whether a patient has diabetes can lead to adjustments in selenium dosage. For diabetic patients, the selenium dosage can range from 0 μg / day to 100 μg / day. For non-diabetic patients, the selenium dosage can range from 0 μg / day to 200 μg / day. 【0043】 Furthermore, a predictive model can be trained to determine antioxidant dosages to achieve the following ratio in the patient's blood at the end of the treatment period, for example: 0.7<(Vitamin C) / (Vitamin E)<0.9 0.8<(copper) / (zinc)<1 7mg / g<(Vitamin E) / (Cholesterol)<10mg / g 【0044】 In some embodiments, the estimation model may be a decision tree. 【0045】 Alternatively, the estimation model may be a neural network that performs classification or regression, such as a convolutional neural network. 【0046】 In this case, the estimation model can be trained on a training base that includes multiple training sets, each of which can be an input vector containing the value of at least one parameter; an output vector comprising a dosage of at least one antioxidant. 【0047】 The training set can be measured for a large number of patients suffering from oxidative stress, and more specifically, FSHD. 【0048】 Training can be performed using any known training algorithm, such as backpropagation. 【0049】 The method according to the invention can advantageously be implemented to determine the dosage of an antioxidant to be administered to a patient suffering from facioscapulohumeral muscular dystrophy (FSHD). 【0050】 In a further aspect of the present invention, there is provided a computer implemented estimation model comprising executable instructions which, when executed by a computing device, implement the estimation steps of the method according to the present invention. 【0051】 The estimation model can be in any computer language, such as machine language, C, C++, JAVA, Python, etc. 【0052】 According to another aspect of the present invention, there is provided an apparatus for aiding in determining the dosage of an antioxidant to be administered to a patient suffering from oxidative stress, in particular FSHD, comprising: - a blood sample previously taken from said patient, having the following input parameters: Cholesterol levels Zinc levels Copper level Vitamin C levels Vitamin E levels Selenium levels, and optionally at least one means for measuring at least one of the iron level and / or the glycosylated hemoglobin level; - a pre-trained estimation model for estimating a dosage of at least one antioxidant to be administered to said patient as a function of said measurements. 【0053】 The device according to the invention may comprise a computing unit for executing the estimation model, which may be any type of device such as a server, computer, tablet, calculator, processor, computer chip, etc. 【0054】 The means for measuring the at least one input parameter can be any known means. 【0055】 For example, at least one input parameter may be measured in a laboratory. 【0056】 The device according to the invention can be advantageously used to determine the dosage of antioxidants to be administered to patients suffering from facioscapulohumeral muscular dystrophy (FSHD). 【0057】 Other advantages and features will become apparent upon examination of the detailed description of the embodiments, which is by no means limiting, with reference to the accompanying drawings. [Brief explanation of the drawings] 【0058】 [Figure 1] 1 is a schematic diagram of a non-limiting embodiment of a method according to the present invention. [Figure 2] 1 is a schematic diagram of a non-limiting embodiment of an apparatus according to the present invention; [Figure 3] FIG. 1 is a schematic diagram of a non-limiting example of a decision tree that can be used as an estimation model in the context of the present invention. DETAILED DESCRIPTION OF THE INVENTION 【0059】 It should be noted that the methods described below are not limited thereto. In particular, it is possible to imagine a variant of the present invention that includes only a selection of the described features, separated from the other features subsequently described, if this selection of features provides a technical advantage or is sufficient to distinguish the present invention from the prior art. This selection includes at least one feature, preferably a functional feature, without structural details, or only a portion of the structural details, if only this portion provides a technical advantage or is sufficient to distinguish the present invention from the prior art. 【0060】 In particular, all the described variations and embodiments can be combined with each other unless there are technical obstacles to such combinations. 【0061】 In the figures and the remainder of the description, elements common to several figures retain the same reference numbers. 【0062】 FIG. 1 is a schematic diagram of a non-limiting embodiment of the method according to the present invention. 【0063】 The method 100 of FIG. 1 is executed by a computing unit and can be used to assist a practitioner in determining the dosage of an antioxidant to be administered to a patient suffering from oxidative stress, particularly FSHD, using a pre-trained / determined estimation model for dosage estimation of at least one antioxidant. 【0064】 In a preliminary step 102 that is not part of the method 100, a blood sample is taken from a patient. The blood sample can be taken in any conventional manner. 【0065】 Method 100 includes an optional step 104 for measuring glycated hemoglobin. Measuring glycated hemoglobin is a well-known procedure and can be performed in a conventional manner, for example in a laboratory. 【0066】 The hemoglobin level is compared to a predetermined threshold in optional step 106 to determine whether the patient also suffers from diabetes. The output of this comparison can be used as an input parameter indicating whether the patient has diabetes. For example, this input parameter can take on a binary value. - On the other hand, for example "0", indicating that the patient is non-diabetic. - On the other hand, for example 1, indicating that the patient is diabetic. 【0067】 The method 100 includes a dose prediction step 108 based on the patient's biological input parameters. 【0068】 The prediction stage 108 includes a step 110 for measuring at least one biological parameter of the blood sample, called an input parameter. 【0069】 In particular, the measuring step 110 includes measuring the following of the blood sample: - measuring 112 the cholesterol level; - measuring zinc levels 114; - measuring the copper level 116; - measuring vitamin C levels; - measuring vitamin E levels 120; - measuring 122 the selenium level. 【0070】 Optionally, measuring step 110 can also include a step 124 of measuring the level of iron in the blood sample. 【0071】 Each of the values ​​measured in the measuring step 110 constitutes an input parameter for a pre-trained or predetermined estimation model for estimating the antioxidant dosage to be administered to the patient. 【0072】 For example, taking into account optional steps, the set of input parameters may form an input vector denoted VE having eight values. VE={D, Ch, Zc, Cu, C, E, Se, Fe} where D indicates whether the patient has diabetes, and Ch, Zc, Cu, C, E, Se, and Fe are the measured values ​​of cholesterol, zinc, copper, vitamin C, vitamin E, selenium, and iron, respectively, in a blood sample taken from the patient. 【0073】 The prediction stage 108 includes a step 130 in which the input vector V is provided as input to a predetermined estimation model for estimating the dosage of at least one antioxidant, in particular a combination of at least two antioxidants, to be administered to a patient. 【0074】 Preferentially, the estimation model is a decision tree. Alternatively, the estimation model may be a neural network or a polynomial model. 【0075】 In response to an input vector Ve, the estimation model provides an output vector, e.g., Vs, that includes at least one dosage value. In a non-limiting example, the output vector includes: - Zinc dosage - Copper dosage - Vitamin C dosage - Vitamin E dosage, and - Selenium dosage. 【0076】 A predictive model can be trained or determined to determine the antioxidant dosage to achieve the following ratio in the patient's blood, for example, at the end of a treatment period: 0.7<(Vitamin C) / (Vitamin E)<0.9 0.8<(copper) / (zinc)<1 7mg / g<(Vitamin E) / (Cholesterol)<10mg / g 【0077】 Furthermore, a predictive model can be determined or trained to provide antioxidant dosages as follows: - The zinc dosage is selected from the following predetermined discrete values: 15 mg / day, 30 mg / day, and 45 mg / day. - Cu dosage is selected from the following predetermined discrete values: 0 mg / day and 0.3 mg / day. - Vitamin E dosage is selected from the following predetermined discrete values: 125 mg / day, 250 mg / day, 500 mg / day. - Vitamin C dosage is selected from the following predetermined discrete values: 125 mg / day, 250 mg / day, 500 mg / day. 【0078】 Furthermore, a predictive model can be determined or trained to provide selenium dosage as follows: - If the patient is diabetic: the selenium dosage is selected between 0 and 200 μg / day, but can be either 0 μg / day or 200 μg / day. - If the patient is non-diabetic: the selenium dosage is selected between 0 and 100 μg / day, but can be either 0 μg / day or 100 μg / day. 【0079】 Of course, all the above values ​​are given as non-limiting examples. 【0080】 The prediction step 108 can be repeated as many times as necessary for the same patient, with the aim of always adapting the antioxidant dosage to the patient's condition. 【0081】 For example, the predicting step 108 can be repeated at a predetermined frequency. Alternatively or additionally, the predicting step 108 can be repeated at the request of the patient or at the discretion of the practitioner. 【0082】 In this way, the patient's condition can be monitored and the dosage of antioxidant can be quickly adapted to changes in the patient's condition. 【0083】 In method 100, diabetes-related steps 104 and 106 are performed before the first iteration of prediction stage 108. Alternatively, these steps may be part of prediction stage 108 and may be repeated at each iteration of prediction stage 108. 【0084】 FIG. 2 is a schematic diagram of a non-limiting embodiment of an apparatus according to the present invention. 【0085】 The apparatus 200 of FIG. 2 may be used to assist a practitioner in determining the dosage of antioxidants to be administered to a patient suffering from oxidative stress, in particular FSHD, using a pre-trained / determined estimation model for dosage estimation of at least one antioxidant, executed by a computational unit. 【0086】 The apparatus 200 of FIG. 2 can be used to implement the method according to the invention, in particular the method 100 of FIG. 【0087】 The device 200 comprises at least one means 202 for measuring at least one biological parameter of a patient's blood sample, such as, for example, the parameters described above with reference to the method 100, i.e., zinc level, copper level, vitamin C level, vitamin E level, selenium level, glycated hemoglobin level, and iron level. At least one of these parameters can be measured by a dedicated measuring device. At least two of these parameters can be measured by the same measuring device. 【0088】 The apparatus 200 further comprises a calculation unit 204 for executing a pre-trained or predetermined estimation model 206 for estimating a dosage of at least one antioxidant, such as, for example, the antioxidants described above with reference to process 100, i.e., a zinc dosage, a copper dosage, a vitamin C dosage, a vitamin E dosage, and a selenium dosage, as a function of input parameter(s) measured by the at least one measurement means. 【0089】 The computing unit 204 can be any type of device, such as a server, computer, tablet, calculator, processor, computer chip, etc. 【0090】 The estimation model 206 can be a decision tree or a neural network. 【0091】 The input parameters may be presented to the estimation model 206 in the form of an input vector, denoted VE, containing a value for each input parameter. The dosage values ​​may be provided by the estimation model in the form of an output vector, denoted Vs, containing a dosage value for each antioxidant. 【0092】 FIG. 3 is a schematic illustration of a non-limiting schematic example of a decision tree that can be used as an estimation model in the context of the present invention. 【0093】 The decision tree shown in FIG. 3 can be used in a method according to the invention, in particular in the method 100 of FIG. 1 or in the device 200 of FIG. 2, respectively. 【0094】 3 begins by taking as input a measurement of vitamin E. The measurement of vitamin E is used to select one of several branches of the decision tree, each corresponding to a value or range of values ​​of vitamin E. 【0095】 The decision tree then takes into account the measured cholesterol value. The measured cholesterol value is used to select one of several branches of the decision tree, each corresponding to a cholesterol value or range of values. The same is done with the measured values ​​of vitamin C, then copper, then zinc, and finally selenium, leading to a terminal branch 302i, i.e., one of several terminal branches 302i-302n. Each 302i terminal branch corresponds to at least one of the following parameters, in particular a dosage for each: - Zinc - copper - Vitamin C - Vitamin E, and - Selenium. 【0096】 Of course, the decision tree 300 can take into account other input parameters such as iron levels and glycated hemoglobin levels. 【0097】 Additionally, the dosages corresponding to each terminal branch of the decision tree may include dosages for parameters other than the indicated parameter, such as selenium levels. 【0098】 Furthermore, the decision tree 300 can be organized differently so that the input parameters are considered in a different order than that shown above. 【0099】 Of course, the invention is not limited to the examples just described.

Claims

[Claim 1] A method (100) for assisting in determining the dosage of antioxidants to be administered to a patient suffering from oxidative stress, the method (100) comprising at least one iteration of a prediction step (108) including the following steps: - For blood samples previously collected from the aforementioned patient, the following input parameters Cholesterol levels • Zinc level Copper level Vitamin C levels • Vitamin E levels; and - A step (110) of measuring at least one of the selenium levels, A method (100) comprising the steps of: - estimating the dosage of at least one antioxidant to be administered to the patient by a previously trained model (206) run by a computing unit (204), wherein the model (206) takes the at least one input parameter as input. [Claim 2] The aforementioned estimation model (206) identifies the following antioxidants - Zinc - Copper - Vitamin C - Vitamin E, and - The method according to claim 1 (100), characterized in that an estimation of the dosage is performed for at least one of the seleniums, particularly for each of them. [Claim 3] The aforementioned estimation model (206) - Zinc dosage of 15 mg / day to 45 mg / day; - Copper dosage of 0-0.3 mg / day; - Vitamin E dosage of 125 mg / day to 500 mg / day; - Vitamin C dosage of 125 mg / day to 500 mg / day; - The method according to claim 2 (100), characterized in that it is trained to provide a selenium dose of 0 to 200 μg / day. [Claim 4] The method according to claim 1 (100), characterized in that the prediction step (108) further includes measuring the iron level in the patient's blood (124), and the estimation of the dosage by the estimation model (206) is further performed as a function of the measured iron level. [Claim 5] The method according to claim 1 (100), further comprising measuring the level of diabetes in the patient's blood, particularly the level of glycated hemoglobin (104), wherein the estimation of the dosage by the estimation model (206) is further performed as a function of the level of diabetes (100). [Claim 6] The method according to claim 1 (100), characterized in that the estimation model (206) is a decision tree. [Claim 7] The method according to claim 1 (100), characterized in that it is used to determine the dosage of an antioxidant administered to a patient suffering from facioscapulohumeral dystrophy (FSHD). [Claim 8] A computer-implemented estimation model (206) that, when executed by a computer, includes an executable instruction that implements the estimation step (130) of the method (100) described in claim 1. [Claim 9] A device (200) for assisting in determining the dosage of antioxidants administered to patients suffering from oxidative stress, - Regarding blood samples previously taken from the aforementioned patient, the following applies: Cholesterol levels • Zinc level Copper level Vitamin C levels Vitamin E levels ・ Selenium level, and - In some cases, at least one means (202) for measuring at least one level of the input parameters of iron level and / or glycated hemoglobin level, and Apparatus (200) comprising: an estimation model (206) pre-trained to estimate the dosage of at least one antioxidant to be administered to the patient as a function of the measured values; [Claim 10] Use of the apparatus (200) according to claim 9 for determining the dosage of an antioxidant administered to a patient suffering from facioscapulohumeral muscular dystrophy (FSHD).

Images