WO2022144079A1

WO2022144079A1 - Apparatus and method for the evaluation and prediction of fatigue

Info

Publication number: WO2022144079A1
Application number: PCT/EP2020/088003
Authority: WO
Inventors: Sasan YAZDANI; Nicolas BOURDILLON
Original assignee: Be.Care Sa
Priority date: 2020-12-29
Filing date: 2020-12-29
Publication date: 2022-07-07
Also published as: WO2022144123A1

Abstract

The present invention relates to a method and a data processing apparatus for deriving at least one fatigue indicator of a user. The method or data apparatus comprises the following steps : a) receiving a reference value of the user for each of the plurality of heart rate variability parameters; b) calculating a plurality of relative changes of the reference value of the user for a second plurality of the heart rate variability parameters in a predefined range; c) calculating the fatigue profile for each of the relative change of the reference value of the user for the second plurality of heart rate variability parameters calculated in step b); and d) deriving the at least one fatigue indicator of the user from the fatigue profiles calculated in step c).

Description

APPARATUS AND METHOD FOR THE EVALUATION AND PREDICTION OF FATIGUE

FIELD OF THE INVENTION

The present invention pertains to the field of heart-rate variability analysis to assess fatigue of a user.

More specifically, but not exclusively, the invention concerns a data processing apparatus for deriving at least one fatigue indicator from a group of alternative fatigue profiles identified using heart-rate variability parameters.

BACKGROUND OF THE INVENTION

Heart rate variability (herein after referred to as “HRV”) is the subject of a consensus from the European Society of Cardiology and The North American Society of Pacing and Electrophysiology (Task Force. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Eur Heart J. 1996; 17: 354-381).

HRV is a commonly used method (more than 1,000 scientific publications in 2019) for cardiovascular follow-up in athletes and healthy people. HRV characterizes the state of the user’s sympathetic and parasympathetic nervous systems. It has been demonstrated that athletes and untrained people who train using HRV follow-up achieve better performances than without (see e.g. (i) Schmitt L, Willis SJ, Fardel A, Coulmy N, Millet GP. Live high-train low guided by daily heart rate variability in elite Nordic-skiers. Eur J Appl Physiol. 2018;l 18: 419-428. doi: 10.1007/s00421-017-3784-9; (ii) Javaloyes A, Sarabia JM, Lamberts RP, Moya-Ramon M. Training Prescription Guided by Heart Rate Variability in Cycling. Int J Sports Physiol Perform. 2018; 1-28. doi: 10.1123/ij spp.2018- 0122; (iii) Da Silva DF, Ferraro ZM, Adamo KB, Machado FA. Endurance Running Training Individually Guided by HRV in Untrained Women. J Strength Cond Res. 2019;33: 736-746. doi:10.1519/JSC.0000000000002001; or (iv) Vesterinen V, Nummela A, Heikura I, Laine T, Hynynen E, Botella J, et al. Individual Endurance Training Prescription with Heart Rate Variability. Med Sci Sports Exerc. 2016;48: 1347-1354. doi: 10.1249/MSS.0000000000000910). It is known to use HRV in evaluating the physiological responses to physical training (see e.g. EP3043875B1), in guiding during physical exercise (see e.g. EP2035098B1) and in determining a training indicator (see e.g. EP3340248A1).

The present invention aims at using HRV in calculating a fatigue indicator.

Use of HRV for calculating a fatigue profile is well established in practice and the state-of-the-art (see Schmitt L, Regnard J, Parmentier AL, Mauny F, Mourot L, Coulmy N, et al. Typology of “Fatigue” by Heart Rate Variability Analysis in Elite Nordic-skiers. Int J Sports Med. 2015;36: 999-1007. doi: 10.1055/s-0035-1548885). In this publication, an analysis of HRV on 57 members of French national Nordic Ski teams was performed. The athletes underwent a 15-minute orthostatic test during which their ECG, was recorded. The orthostatic test was comprised of an 8-minute supine recording followed by a 7-minute standing recording. The authors computed time- and frequency-domain HRV parameters to identify four fatigue profiles (referred to as fatigue patterns). For each fatigue profile, the authors described how HRV parameters differ from a non-fatigue reference value of said HRV parameters:

(i) Decreased powers in LF, HF, and total power frequency bands accompanied with an increase in HR in both supine and standing positions.

In this fatigue profile, a decrease in LF and HF powers together with an increase in HR was observed in both supine and standing positions of the orthostatic test, with respect to the non-fatigue reference value. This fatigue profile was most observed in the participating athletes (74% of all fatigue profiles) and was synonymous with a temporary training-induced fatigue following high intensity training.

(ii) Increased supine total power accompanied by a decreased standing total power with increased HR in standing position.

In this fatigue profile the supine and standing HRV parameters behaved contrary to one another. While total power increased in the supine position with mainly unaltered LF power and HR, in the standing position total power decreased mostly due to a decrease in LF together with an increase in HR. This fatigue profile accounted for 14% of all identified fatigue profiles.

(iii) Decreased supine total power accompanied by increased standing total power with decreased HR in standing position.

In comparison with non-fatigue reference value, HR remained unchanged while total power decreased in the supine position. Standing total power increased, especially the HF power, and a minimum of 9% HR decrease was observed. This fatigue profile accounted for 10.6% of all fatigue profiles.

(iv) In supine position, increase of total power, LF power and HF power with decrease of HR. In standing position, decrease of total power, LF power and HF power with increase of HR.

In this fatigue profile, supine total power was largely increased compared to the non-fatigue reference value, mainly due to a large increase in HF power with a decreased HR. In standing position, total power decreased while HR slightly increased. This profile was identified once and accounted for 1.7% of all fatigue profiles.

The work described above assigns a single fatigue profile to a user and does not provide any information on the progress of the user. The user does not know if his fatigue profile is moving towards another fatigue profile, improving or degrading his fatigue.

Consequently, there is still a need for (i) deriving reliable information on fatigue profiles and for (ii) predicting future fatigue profiles.

The invention lies within this context.

SUMMARY OF THE INVENTION

The present invention is directed toward a data processing apparatus and a method for deriving at least one fatigue indicator from a group of alternatives of fatigue profiles.

According to a first aspect of the invention, there is provided a data processing apparatus for deriving at least one fatigue indicator of a user from a plurality of predetermined alternatives of fatigue profiles, each fatigue profile being obtainable from a plurality of HRV parameters; the data processing apparatus comprising a processor configured to perform the following steps: a) receiving a reference value of the user for each of the plurality of HRV parameters; b) calculating a plurality of relative changes of the reference value of the user for a second plurality of the HRV parameters in a predefined range; c) calculating the fatigue profile for each of the relative changes of the reference value of the user for the second plurality of HRV parameters calculated in step b); and d) deriving the at least one fatigue indicator of the user from the fatigue profiles calculated in step c). The data processing apparatus according to the first aspect enables to understand how the fatigue profiles are distributed around the reference values of the user. The exploration of the fatigue profiles around the reference values of the user allows to derive reliable fatigue indicators and to predict future fatigue profiles of the user.

According to a second aspect of the present invention, it is also provided a method for deriving at least one fatigue indicator of a user from a group of predetermined alternatives of fatigue profiles in a data processing apparatus, each fatigue profile being obtainable from a plurality of HRV parameters; the method comprising the following steps: a) receiving a reference value of the user for each of the plurality of HRV parameters; b) calculating a plurality of relative changes of the reference value of the user for a second plurality of the HRV parameters in a predefined range; c) calculating the fatigue profile for each of the relative changes of the reference value of the user for the second plurality of HRV parameters calculated in step b); and d) deriving the at least one fatigue indicator of the user from the fatigue profiles calculated in step c).

According to embodiments of the first aspect or the second aspect, the proposed method or apparatus may also comprise at least one of the following additional features.

According to one embodiment, the reference value of the user for each of the plurality of HRV parameters is based on history of the values of the user for each of the plurality HRV parameters.

According to one embodiment, the predefined range of variation in step b) is based on physiological variations of the second plurality of HRV parameters.

According to one embodiment, the at least one fatigue indicator in step d) is derived by calculating the percentage of each fatigue profile calculated in step c).

According to one embodiment, the at least one fatigue indicator in step d) is derived by:

- repeating steps a) to c) over time; and

- monitoring the change of percentage of each fatigue profile calculated in steps c) over time. According to one embodiment, the at least one fatigue indicator in step d) is derived by comparing the fatigue profiles calculated in step c) with the fatigue profiles of a reference population.

According to one embodiment, step d) comprises the steps of :

- receiving a current value of the user for each of the plurality of HRV parameters ;

- calculating the current fatigue profile of the user from the current value of the user for each of the plurality of HRV parameters and from the reference value of the user for each of the plurality of HRV parameters; and

- calculating at least one fatigue indicator from the current fatigue profile of the user and from the fatigue profiles calculated in step c).

Said embodiment enable to derive a fatigue indicator considering the current fatigue profile of the user. It is thus possible to predict in a reliable manner the future fatigue profile of the user starting from the current fatigue profile of the user.

According to one embodiment, the at least one fatigue indicator in step d) is derived by calculating multidimensional statistics in a multidimensional space representing the current fatigue profile of the user and the fatigue profiles calculated in step c); wherein (i) each HRV parameter of the second plurality of HRV parameters is a dimension or (ii) the relative difference between each HRV parameter of the second plurality of HRV parameters and the respective reference value of each of the HRV parameters of the second plurality of HRV parameters is a dimension.

Said embodiment enables to explore the distribution of the fatigue profiles around the current fatigue profile.

According to one embodiment, the at least one fatigue indicator in step d) is derived by calculating the percentage of each fatigue profile in a second predefined range around the current fatigue profile of the user in the multi-dimensional space.

According to one embodiment, the at least one fatigue indicator in step d) is derived by calculating a score from the vectors connecting, in the multi-dimensional space, the point representing the current fatigue profile of the user and the points representing the center of each fatigue class regrouping all the same fatigue profiles calculated in step c).

- receiving over time a plurality of current values of the user for each of the plurality of heart rate variability parameters; - calculating the current fatigue profiles of the user over time from the current values of the user for each of the plurality of heart rate variability parameters obtained over time and from the reference value of the user for each of the plurality of heart rate variability parameters;

- representing the current fatigue profiles of the user obtained over time in the multidimensional space; and

- calculating the at least one fatigue indicator from the direction of the vectors connecting the points representing successive current fatigue profiles of the user over time in the multi-dimensional space.

According to one embodiment, the plurality of HRV parameters comprises time domain HRV parameters and/or frequency-domain HRV parameters and/or non-linear domain HRV parameters.

According to a third aspect of the invention there is provided a computer program product comprising instructions for implementing the method according to the second aspect of the invention when the program is executed by a computer.

There is also provided according to a fourth aspect of the invention, a non- transitory storage medium readable by a computer storing instructions for implementing the method according to the second aspect of the invention, when executed by the computer.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be more clearly understood from the following description, given by way of example only, with reference to the accompanying drawings, in which:

Figure 1 illustrates a flow diagram of a method for deriving at least one fatigue indicator.

Figure 2 illustrates RR-intervals recording in the standing position and the supine position

Figure 3 illustrates a three-dimensional space representing the fatigue profiles around the point of reference of the user (large light sphere) and the current fatigue profile of the user (large dark sphere). Figure 4 illustrates a sub-three-dimensional space representing the fatigue profiles around the current fatigue profile of the user (large dark sphere). The sub-three- dimensional space is obtained from the three-dimensional space of figure 3.

Figure 5 illustrates the trajectory of successive current fatigue profiles of the user in the three-dimensional space (represented in two-dimensions for visualization purposes).

Figure 6 illustrates a data processing apparatus according to embodiments of the invention. ABBREVIATIONS AND DEFINITIONS

DETAILED DESCRIPTION OF THE INVENTION

Figure 1 is a flow diagram illustrating a method for deriving at least one fatigue indicator in a data processing apparatus according to an embodiment of the invention.

Referring to figure 1, the method starts with block SI. The reference value of the user for a first plurality of HRV parameters is received.

HRV parameters

HRV parameters are a measure of variations in heart rate intervals.

Many heart rate monitors commercially available, for instance wearable heart rate monitors such as chest strap heart rate monitors, measure inter-beat intervals (IB I) such as RR-intervals or pulse rate variability (PRV). Measuring IBI usually involves measuring an electrocardiogram (ECG) or a photoplethysmography (PPG) signal. From an ECG signal, it is known to calculate RR-intervals which in turn are used to calculate HRV parameters. From a PPG signal, it is known to calculate PRV, which in turn, is used to calculate HRV parameters.

Many HRV parameters have been developed over years and it is likely that new HRV parameters will be developed in the next years.

HRV parameters may be extracted from IBI measured at rest or during exercise.

HRV parameters may be extracted from IBI measured in any position, such as in a standing position and/or in a supine position.

Figure 2 illustrates the recording of RR-intervals using a commercially available heart rate monitor in a supine position and in a standing position.

HRV parameters comprises time domain HRV parameters and/or frequencydomain HRV parameters and/or non-linear domain HRV parameters. HRV parameters may comprise HRV parameters in different positions, for instance HRV parameters in supine position and in standing position. The time-domain HRV parameters may be selected from heart rate, SDNN, SDANN, SDNN index (SDNNI), pNN50, HR Max - HR Min, RMSSD, Triangular index and/or TINN.

The frequency-domain HRV parameters may be selected from total power, total power, ULF power, VLF power, LF peak, LF power, HF peak, HF power and/or LF/HF.

The non-linear domain HRV parameters may be selected from SD1, SD2, ApEN, SampEn, D₂, o^, CL₂, REC, DET and/or ShEn.

The HRV parameters are not limited to those listed above but encompass any HRV parameter known to the skilled artisan.

First plurality of HRV parameters

The first plurality of HRV parameters comprises all the HRV parameters required to calculate the fatigue profiles of the group of alternative fatigue profiles.

According to one embodiment, the first plurality of HRV parameters comprises at least one time-domain HRV parameter and/or at least one frequency-domain HRV parameter and/or at least one non-linear domain HRV parameter. In said embodiment, the reference value of the user of at least one time-domain HRV parameter and/or the reference value of the user of at least one frequency-domain HRV parameter and/or the reference value of the user of at least one non-linear domain HRV parameter is received.

According to one embodiment, the first plurality of HRV parameters comprises at least one at least one time-domain HRV parameter, such as heart rate and/or RMSSD; and/or at least one frequency-domain HRV parameter, such as LF power, HF power and/or, total power; and/or at least one non-linear domain HRV parameter such as SD1 and/or SD2.

In this embodiment, the reference value of the user of at least one time-domain HRV parameter such as heart rate and/or RMSSD is received; and/or the reference value of the user of at least one frequency-domain HRV parameter, such as LF power, HF power and/or total power is received; and/or the reference value of the user of at least one nonlinear domain HRV parameter, such as DS1 and/or DS2 is received.

Reference value of the user of a HRV parameter

The reference value of the user of a HRV parameter is determined using the value of the HRV parameter obtained in a reference state of the user. According to one embodiment, a reference state refers to a non-fatigue state as determined using a questionnaire. The questionnaire may be any questionnaire known in the art such as the QSFMS questionnaire (see Schmitt L, Regnard J, Desmarets M, Mauny F, Mourot L, Fouillot JP, Coulmy N, Millet G. Fatigue shifts and scatters heart rate variability in elite endurance athletes. PLoS One 2013; 8: e71588).

According to one embodiment, the reference value of the user is based on the history of values of the user. According to one embodiment, each value of the user of the first plurality of HRV parameters is stored in a memory and the reference value of the user of the first plurality of HRV parameters is updated based on the stored values.

According to one embodiment, if the fatigue profile calculated from the current value of the user for each of the first plurality of HRV parameters is a reference fatigue profile, then the current value of the user for each of the first plurality of HRV parameters is used to update the corresponding reference value of the user. According to one embodiment, the reference value of the user is a mean or median of the current value of the first plurality of HRV parameters for each test for which the fatigue profile calculated from the current value of the user for each of the first plurality of HRV parameters is a reference fatigue profile.

Said embodiment enables to update the reference values of the user over time and thus takes into account the evolution of the user.

Back to figure 1, in block S2, relative changes of the reference value of the user of a second plurality of HRV parameters are calculated in a predetermined range.

Relative changes of a second plurality of HRV parameters

According to one embodiment, the second plurality of HRV parameters is included in the first plurality of HRV parameters. According to one embodiment, the second plurality of HRV parameters is a subset of the first plurality of HRV parameters. According to an alternative embodiment, the second plurality of HRV parameters is the same as the first plurality of HRV parameters. According to one embodiment, the second plurality of HRV parameters comprises or consists of three HRV parameters.

According to one embodiment, the second plurality of HRV parameters comprises at least one time-domain HRV parameter and/or at least one frequency-domain HRV parameter and/or at least one non-linear domain HRV parameter.

According to one embodiment, the second plurality of HRV parameters comprises at least one at least one time-domain HRV parameter, such as heart rate and/or RMSSD; and/or at least one frequency-domain HRV parameter, such as LF power, HF power and/or, total power; and/or at least one non-linear domain HRV parameter such as SD1 and/or SD2.

These HRV parameters are selected because they contribute the most in identification of fatigue profiles.

According to one embodiment, the range is predetermined so that each fatigue profile of the group of alternative fatigue profiles is obtained during block S3. It means that the reference values of the second plurality are modified on a sufficiently wide range and a sufficiently fine pitch so that each fatigue profile is obtained at least once.

According to one embodiment, the range is predetermined based on the physiological values of each HRV parameter of the second plurality of HRV parameters. It means that the range is predetermined so that the HRV parameters covers the whole spectra of physiological values. According to one embodiment, the range is predetermined based on statistically derived values of each HRV parameter.

According to one embodiment, the relative changes are calculated with a pitch depending on the physiological values of each HRV parameter of the second plurality of HRV parameters. According to one embodiment, the pitch is predetermined based on statistically derived values of each HRV parameter.

According to an exemplary embodiment, relative changes of the heart rate are calculated from the reference value of heart rate of the user so that the calculated values covers the physiological values of HR, for instance from 25 bpm to 220 bpm. According to one embodiment, the pitch of relative change of the heart rate is based on statistical analysis of a reference population, for instance a pitch of 0.1.

According to an exemplary embodiment, relative changes of the RMSSD are calculated from the reference value of RMSSD of the user so that the calculated values covers the physiological values of RMSSD, for instance from 0.1 to 10 times the reference value of the user with a pitch of 0.2.

According to an exemplary embodiment, relative changes of the LF power are calculated from the reference value of LF power of the user so that the calculated values covers the physiological values of LF power, for instance from 0.1 to 15 times the reference value of the user with a pitch of 0.2.

According to an exemplary embodiment, relative changes of the HF power are calculated from the reference value of HF power of the user so that the calculated values covers the physiological values of HF power, for instance from 0.1 to 15 times the reference value of the user with a pitch of 0.2.

According to an exemplary embodiment, relative changes of the total power are calculated from the reference value of total power of the user so that the calculated values covers the physiological values of total power, for instance from 0.1 to 15 times the reference value of the user with a pitch of 0.2.

According to an exemplary embodiment, relative changes of the SD1 are calculated from the reference value of SD1 of the user so that the calculated values covers the physiological values of SD1, for instance from 0.1 to 10 times the reference value of the user with a pitch of 0.5.

According to an exemplary embodiment, relative changes of the SD2 are calculated from the reference value of SD2 of the user so that the calculated values covers the physiological values of SD2, for instance from 0.1 to 10 times the reference value of the user with a pitch of 0.5.

In block S3, the fatigue profile associated with each relative change is calculated.

Fatigue profile

A fatigue profile may be calculated from a group of alternative fatigue profiles based on the first plurality of HRV parameters. Once a reference value of the user is known for each of the first plurality of HRV parameters, a fatigue profile may be calculated by comparing a new measure for each of the first plurality of HRV parameters with the reference value of the user for the corresponding HRV parameters.

A value of the first plurality of HRV parameters is compared with the reference value of said first plurality of HRV parameters. Based on the comparison, a fatigue profile may be determined.

For instance :

- a first fatigue profile may be assigned if a decrease in LF power, HF power and total power and an increase in HR is identified in supine and standing positions when comparing the values of HRV parameters of the first plurality with the reference values of the user;

- a second fatigue profile may be assigned if an increase in total power in supine position, a decrease in total power in standing position and an increase of HR in standing position is identified when comparing the values of HRV parameters of the first plurality with the reference values of the user;

- a third fatigue profile may be assigned if a decrease in total power in supine position, an increase in total power in standing position and a decrease in HR in standing position is identified when comparing the values of HRV parameters of the first plurality with the reference values of the user; and

- a fourth fatigue profile may be assigned if an increase in LF power, HF power and total power in supine position, a decrease of HR in supine position, a decrease of LF power, HF power and total power in standing position and an increase of HR in standing position is identified when comparing the values of HRV parameters of the first plurality with the reference values of the user.

According to one embodiment, the group of alternative fatigue profiles comprises a reference fatigue profile. The reference fatigue profile corresponds to a non-fatigue state of the user, as determined for instance from a questionnaire.

According to one embodiment, a fatigue profile is calculated for each relative change of the reference value of the user for the second plurality of HRV parameters by comparing the value of each HRV parameters of the first plurality (taking into account the relative change of at least one HRV parameter of the second plurality) with the reference value of the user of the first plurality of HRV parameters.

For each relative change of each HRV parameter of the second plurality of HRV parameters, a fatigue profile is calculated. The fatigue profiles are calculated by varying all HRV parameters of the second plurality of HRV parameters within the predetermined range.

It means that for each relative change of a HRV parameter of the second plurality of the HRV parameters, the other parameters of the second plurality of HRV parameters vary within the predetermined range and the corresponding fatigue profiles are calculated.

Multi-dimensional space

According to one embodiment, the fatigue profiles calculated in block S3 are represented in a multi-dimensional space. According to one embodiment, each HRV parameter of the second plurality is a dimension in the multi-dimensional space. According to a preferred embodiment, the relative difference with respect to the reference value of the user of each HRV parameters of the second plurality of HRV parameters is a dimension in this multi-dimensional space. According to one embodiment, the multidimensional space is a three-dimensional space. According to one embodiment, the multi-dimensional space is displayed to the user.

According to one embodiment, each fatigue profile of the group of alternative fatigue profiles is displayed in the multi-dimensional space with a different color or symbol.

This embodiment enables to visualize the distribution of the fatigue profiles as a function of the HRV parameters of the second plurality of HRV parameters.

According to one embodiment, multi-dimensional statistics are performed in this multi-dimensional space. According to one embodiment, multidimensional statistics may be performed without displaying the multi-dimensional space.

In block S4, at least one fatigue indicator is derived from the fatigue profiles calculated at in block S3.

The at least one indicator is based on the distribution of fatigue profiles calculated in block S3 around a point of reference. The point of reference is the reference value of each HRV parameter of the second plurality of HRV parameters.

According to one embodiment, a feedback is given to the user depending on the at least one fatigue indicator. The feedback is based on the at least one fatigue indicator and aims at guiding the user towards a reference fatigue profile.

Once the distribution of fatigue profiles around the point of reference is known and represented in the multi-dimensional space, a current fatigue profile of the user may be calculated and represented in the multi-dimensional space. Figure 3 illustrates a three- dimensional space wherein the point of reference is represented with a large light sphere and the current fatigue profile of the user is represented with a large dark sphere.

The position of the current fatigue profile within the distribution of fatigue profiles may be studied to derive at least one fatigue indicator. Said fatigue indicator is reliable and more precise than the current fatigue profile considered alone.

Current fatigue profile

The current fatigue profile is calculated from the current value of the user for each of the first plurality of HRV parameters and from the reference value of the user for each of the first plurality of HRV parameters. The current values are compared to the corresponding reference value. Current value of the user of a HRV parameter

The current value of the user of a HRV parameter refers to the instant’ s value of a HRV parameter as derived from IBI which are measured from a heart rate monitor.

According to one embodiment, the current value of the user for each of the first plurality of HRV parameters is derived from an ECG signal or from a PPG signal or any other means from which the IBI can be derived.

According to an embodiment of the invention, the method comprises, or the apparatus is configured to perform, the step of calculating IBI, such as RR-intervals or PRV signal from an ECG signal of a user or a PPG signal of a user, respectively.

According to an embodiment of the invention, said ECG signal of a user or said PPG signal is recorded in various positions such as in a supine position and in a standing position.

According to an embodiment of the invention, the method comprises, or the apparatus is configured to perform, the step of calculating a value for each of the first plurality of HRV parameters from IBI, such as RR-intervals or PRV signal.

According to an embodiment of the invention, the method comprises, or the apparatus is configured to perform, the step of calculating a current value for each of the first plurality of HRV parameters from IBI, such as RR-intervals or PRV signal in various positions such as in a standing position and/or in a supine position.

In the present invention, at least a first, a second and a third fatigue indicator are calculated from the current fatigue profile and the fatigue profiles calculated in block S3.

First fatigue indicator - Score

As mentioned in the background, one drawback of the known method is the lack of sense of progress. The user does not know towards which fatigue profile he/she is moving before reaching that fatigue profile. The first fatigue indicator according to the invention is a score. This score enables the users to compare and track their changes even when they find themselves in the same fatigue profile.

According to one embodiment, the score is ranging from a first integer a to a second different integer b. for instance from 0 to 100.

According to one embodiment, the score is calculated in the multi-dimensional space. All the same fatigue profiles are considered as a class and the center of each class is calculated in the multi-dimensional space. Then, the score is calculated based on the vectors connecting the point representing the current fatigue profile of the user and the points representing the center of each fatigue class.

According to one embodiment, for each fatigue class (except the fatigue class representing the reference fatigue profile), the projection of (i) the vector connecting the point representing the current fatigue profile and the point representing the center of the fatigue class on (ii) the vector connecting the center of the fatigue class and the center of the fatigue class representing the reference fatigue profile is calculated. Then the score is obtained based on the minimum distance of the projected vectors. For instance, the score is calculated as 100 x the ratio of (i) the minimum distance of the projected vectors to (ii) the distance of the vector connecting the center of the corresponding fatigue class to the center of the fatigue class representing the reference fatigue class.

Second fatigue profile - Prediction

Without willing to be bound to any theory, the applicant discovered that a newly taken measure of HRV parameters of a user falls within a certain distance with respect to the previous measure of said HRV parameters of the user. It means that a new test spans a smaller subspace than the original space representing the physiological variations of HRV parameters. Consequently, by studying the percentage of fatigue profiles in the vicinity of the current fatigue profile, it is possible to predict the next fatigue profile.

Figure 4 illustrate a subspace of the three-dimensional space around the current fatigue profile of the user.

According to one embodiment, a second fatigue indicator is derived by calculating the percentage of each fatigue profile in a second predetermined range for each HRV parameter of the second plurality of HRV parameters around current fatigue profile of the user in the multi-dimensional space.

According to one embodiment, the second predetermined range is customized for the user based on measurements of the variations of the second plurality of HRV parameters carried out before and then stored in a memory. According to one embodiment, the second predetermined range is based on the variations of the second plurality of HRV parameters as measured from a reference population.

According to an exemplary embodiment, the second predetermined range for HR ranges from 0.5 times the current value of the user to 1.5 times the current value of the user. According to an exemplary embodiment, the second predetermined range for RMSSD ranges from 0.8 times the current value of the user to 1.5 times the current value of the user.

According to an exemplary embodiment, the second predetermined range for LF power ranges from 0.5 times the current value of the user to 2 times the current value of the user.

According to an exemplary embodiment, the second predetermined range for HF ranges from 0.5 times the current value of the user to 4 times the current value of the user.

According to an exemplary embodiment, the second predetermined range for total power ranges from 0.5 times current value of the user to 2 times the current value of the user.

According to an exemplary embodiment, the second predetermined range for SD1 ranges from 0.3 times the current value of the user to 3 times the current value of the user.

According to an exemplary embodiment, the second predetermined range for SD2 ranges from 0.3 times the current value of the user to 3 times the current value of the user.

The second fatigue indicator is a reliable indicator of the next fatigue profile of the user.

Third fatigue indicator - Trend analysis

The third fatigue indicator aims at predicting the fatigue profiles over time, not only in the next fatigue profile.

This fatigue indicator is particularly useful to give the user a sense of progression on the long term, especially because the evolution of fatigue profile may be counter intuitive. Figure 5 illustrates the successive current fatigue profile of the user in the multidimensional space. The trajectory of the user suggests moving towards a first fatigue profile before a second fatigue profile. There is thus a need for a reliable prediction over multiple measurements.

According to one embodiment, a plurality of successive current fatigue profiles and associated values of the first plurality of HRV parameters is stored in memory. The successive current fatigue profiles are represented in the multidimensional space. A trajectory is calculated as the direction of the weighted sum of the vectors connecting successive current fatigue profiles in the multidimensional space. According to one embodiment, the third indicator is the next fatigue profile aligned with the trajectory. According to an alternative embodiment, the third fatigue indicator is an estimate of the time to reach the next fatigue profile aligned with the trajectory. An estimate of the time to reach the next fatigue profile may be calculated by dividing the mean or median distance of the vector connecting the successive fatigues profiles by the distance between the current fatigue profile and the next fatigue profile aligned with the trajectory.

Fourth fatigue indicator - Comparison

The fourth fatigue indicator is not based on the current fatigue profile of the user but on the long-time variation of the reference values of the user.

As the reference values of the user evolves over time, the fatigue profiles calculated in block S3 also evolves. Monitoring the change of distribution of the fatigue profiles gives an indication of the evolution of the user.

According to one embodiment, the fourth fatigue indicator is based on the change of percentage of each fatigue profile calculated in block S3 over time when the reference values of the user are updated as mentioned above. This indicator allows to measure the predisposition of a user towards certain fatigue profiles and through time.

Fifth fatigue indicator - Categorization

The fifth indicator is an inter-individual indicator - contrary to the other fatigue indicators which are intra-individual.

According to one embodiment, the fifth fatigue indicator is derived by comparing the fatigue profiles calculated in block S3 with the fatigue profiles of a reference population.

According to one embodiment, the fifth indicator is derived by comparing the percentage of each fatigue profile calculated in block S3 with the percentage of each fatigue profile for a reference population.

Fatigue profdes of a reference population

The distribution of fatigue profiles varies for different lifestyles. It is possible to identify statistically independent patterns of distribution representing various lifestyles, e.g. different sports. As there is a huge variation between HRV parameters, the identification of pattern is not possible for an HRV parameter. However, the identification of pattern is possible for the distribution of fatigue profiles around the point of reference.

According to one embodiment, the fatigue profiles of a reference population refers to the distribution of fatigue profile in a multi-dimensional space for a particular lifestyle. Structure of the data processing apparatus

Figure 6 is a schematic block diagram of the general structure of the data processing apparatus 600. The apparatus comprises a communication bus connected to :

- a central processing unit 601, such as a microprocessor or processor , denoted CPU;

- a random access memory 602, denoted RAM, for storing the executable code of the steps of embodiments of the invention as well as the registers adapted to record variables, data and parameters necessary for implementing the steps according to embodiments of the invention, the memory capacity thereof can be expanded by an optional RAM connected to an expansion port for example;

- a read only memory 603, denoted ROM, for storing computer programs for implementing embodiments of the invention;

- a network interface 604, denoted Net interf, which is typically connected to a communication network over which digital data to be processed are transmitted or received. The network interface 604 can be a single network interface or composed of a set of different network interfaces (for instance wired and wireless interfaces, or different kinds of wired or wireless interfaces). Data are written to the network interface for transmission or are read from the network interface for reception under the control of the software application running in the CPU 601;

- a user interface 605, denoted Usr interf, for receiving inputs from a user or to display information to a user;

- a hard disk 606 denoted HD;

- a communication module 607, denoted COM, for receiving/sending data from/to external devices such as a video source or display.

The executable code may be stored either in read only memory 603, on the hard disk 606 or on a removable digital medium such as for example a disk. According to a variant, the executable code of the programs can be received by means of a communication network, via the network interface 604, in order to be stored in one of the storage means of the apparatus 600, such as the hard disk 606, before being executed.

The central processing unit 601 is adapted to control and direct the execution of the instructions or portions of software code of the program or programs according to embodiments of the invention, which instructions are stored in one of the aforementioned storage means. After powering on, the CPU 601 is capable of executing instructions from main RAM memory 602 relating to a software application after those instructions have been loaded from the program ROM 603 or the hard-disk (HD) 606 for example. Such a software application, when executed by the CPU 601, causes the steps according to embodiments of the invention to be performed.

While various embodiments, have been described and illustrated, the details description and drawings should not be considered as restrictive but merely exemplary and illustrative. Various modifications can be made to the embodiments by those skilled in the art without departing from the scope of the disclosure as defined by the claims.

Claims

- 22 - CLAIMS

1. A data processing apparatus for deriving at least one fatigue indicator of a user from a group of predetermined alternatives of fatigue profiles, each fatigue profile being obtainable from a plurality of heart rate variability parameters; the data processing apparatus comprising a processor configured to perform the following steps : a) receiving a reference value of the user for each of the plurality of heart rate variability parameters; b) calculating in a predefined range a plurality of relative changes of the reference value of the user for a second plurality of the heart rate variability parameters; c) calculating the fatigue profile for each of the relative changes of the reference value of the user for the second plurality of heart rate variability parameters calculated in step b); and d) deriving the at least one fatigue indicator of the user from the fatigue profiles calculated in step c).

2. The data processing apparatus according to claim 1, wherein the reference value of the user for each of the plurality of heart rate variability parameters is based on history of the values of the user for each of the plurality heart rate variability parameters.

3. The data processing apparatus according to claim 1 or claim 2, wherein the predefined range of variation in step b) is based on physiological variations of the second plurality of heart rate variability parameters.

4. The data processing apparatus according to anyone of claims 1 to 3, wherein the at least one fatigue indicator in step d) is derived by calculating the percentage of each fatigue profile calculated in step c).

5. The data processing apparatus according to anyone of claim 1 to 4, wherein the at least one fatigue indicator in step d) is derived by:

- repeating steps a) to c) over time; and

- monitoring the change of percentage of each fatigue profile calculated in steps c) over time.

6. The data processing apparatus according to claim 4, wherein the at least one fatigue indicator in step d) is derived by comparing the fatigue profiles calculated in step c) with the fatigue profiles of a reference population.

7. The data processing apparatus according to anyone of claims 1 to 6, wherein the at least one fatigue indicator in step d) is derived by :

- receiving a current value of the user for each of the plurality of heart rate variability parameters ;

- calculating the current fatigue profile of the user from the current value of the user for each of the plurality of heart rate variability parameters and from the reference value of the user for each of the plurality of heart rate variability parameters; and

8. The data processing apparatus according to claim 7, wherein the at least one fatigue indicator in step d) is derived by calculating multidimensional statistics in a multidimensional space representing the current fatigue profile of the user and the fatigue profiles calculated in step c); wherein (i) each heart rate variability parameter of the second plurality of heart rate variability parameters is a dimension or (ii) the relative difference between each heart rate variability parameter of the second plurality of heart rate variability parameters and the respective reference value of each of the heart rate variability parameters of the second plurality of heart rate variability parameters is a dimension.

9. The data processing apparatus according to claim 8, wherein the at least one fatigue indicator in step d) is derived by calculating the percentage of each fatigue profile in a second predefined range around the current fatigue profile of the user in the multi-dimensional space.

10. The data processing apparatus according to claim 8, wherein the at least one fatigue indicator in step d) is derived by calculating a score from the vectors connecting, in the multi-dimensional space, the point representing the current fatigue profile of the user and the points representing the center of each fatigue class regrouping all the same fatigue profiles calculated in step c).

11. The data processing apparatus according to claim 8, wherein the at least one fatigue indicator in step d) is derived by:

- receiving over time a plurality of current values of the user for each of the plurality of heart rate variability parameters;

- calculating the current fatigue profiles of the user over time from the current values of the user for each of the plurality of heart rate variability parameters obtained over time and from the reference value of the user for each of the plurality of heart rate variability parameters; - representing the current fatigue profiles of the user obtained over time in the multi-dimensional space; and

- calculating the at least one fatigue indicator from the direction of the vectors connecting the points representing successive current fatigue profiles of the user over time in the multi-dimensional space. The data processing apparatus according to anyone of claims 1 to 11, wherein the plurality of heart rate variability parameters comprises time domain heart rate variability parameters and/or frequency-domain heart rate variability parameters and/or non-linear domain heart rate variability parameters. A method for deriving at least one fatigue indicator of a user from a group of predetermined alternatives of fatigue profiles in a data processing apparatus, each fatigue profile being obtainable from a plurality of heart rate variability parameters; the method comprising the following steps : a) receiving a reference value of the user for each of the plurality of heart rate variability parameters; b) calculating a plurality of relative changes of the reference value of the user for a second plurality of the heart rate variability parameters in a predefined range; c) calculating the fatigue profile for each of the relative changes of the reference value of the user for the second plurality of heart rate variability parameters calculated in step b); and d) deriving the at least one fatigue indicator of the user from the fatigue profiles calculated in step c). The method according to claim 13, wherein the at least one fatigue indicator in step d) is derived by :