# US 10,765,328 B2 · Method and system to determine anaerobic threshold of a person non-invasively from freely performed exercise and to provide feedback on training intensity > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,765,328 B2 - Document ID: US10765328B2 - Status: Granted - Kind: B2 - Application number: 15/086,337 - Filed: 2016-03-31 - Published: 2017-02-23 - Granted: 2020-09-08 - Earliest claimed: 2015-08-18 - Assignees: Firstbeat Analytics Oy - Applicants: Firstbeat Analytics Oy - Inventors: Sami Saalasti; Kaisa Hämäläinen; Tero Myllymäki - Web page: https://patentgrep.com/patent/US10765328B2 ## CPC - A61B 5/024: Measuring pulse rate or heart rate - A61B 5/0205: Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition - A61B 5/02405: {Determining heart rate variability} - A61B 5/0833: {Measuring rate of oxygen consumption} - A61B 5/222: {combined with detection or measurement of physiological parameters, e.g. heart rate} - A61B 5/486: {Biofeedback (using electroencephalography [EEG] A61B5/375)} - A61B 5/4866: {Evaluating metabolism (using breath test A61B5/083)} - A61B 5/7221: {Determining signal validity, reliability or quality (preventing, reducing or removing noise induced by motion artefacts A61B5/7207; noise originating from a therapeutic or surgical apparatus A61B5/7217)} - A61B 5/7225: {Details of analogue processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation (input circuits for detecting, measuring, or recording bioelectric or biomagnetic signals A61B5/30; specific diagnostic methods using bioelectric or biomagnetic signals A61B5/316)} - A61B 5/7239: {using differentiation including higher order derivatives} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 5/742: {using visual displays (displays for heart-related electrical signals, e.g. ECG, A61B5/339)} - G16H 50/30: for calculating health indices; for individual health risk assessment - A61B 5/7275: {Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor} - A61B 2503/10: Athletes - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract A method and system for determining anaerobic threshold intensity (AnT) of a user in a freely performed physical exercise. A physiological response of a user is measured by heart rate and measured heart rate values are recorded as heart rate data. An external workload values are recorded and are each associated with one measured heart rate values to form a plurality of data points. The data points are filtered to form accepted data points, which are classified within a plurality of heart rate segments representing a heart rate within an anaerobic threshold (AnT) of the user. A data point with highest probability is stored for each segment. A first probability factor for each accepted data point is calculated. The calculated first probability factor is compared to a stored probability factor in each segment, and the higher probability factor is retained. AnT is calculated using the stored probabilities in each segment. ## Claims ### Claim 1 A method for determining anaerobic threshold intensity (AnT) of a user in a freely performed physical exercise, comprising: continuously measuring, by a body-mounted heart rate sensor, a physiological response of a user by heart rate, wherein a plurality of measured heart rate values are recorded with time stamps as heart rate data forming a recorded heart rate of the user; continuously measuring an external workload in terms of speed and inclination or measured power, wherein a plurality of measured workload values are recorded and each measured workload value is associated with one of the plurality of measured heart rate values to form a plurality of data points; calculating, by a processor, heart rate variability (HRV) values from intervals between successive heartbeats and storing each calculated HRV value associating to a data point; filtering, by the processor, one or more data points based on a predetermined criteria to form a plurality of accepted data points; classifying, by the processor, accepted data points within a plurality of heart rate segments (i) representing a heart rate within an anaerobic threshold (AnT) of the user, wherein at least one segment is below a probable heart rate value of AnT, and at least one segment is above the probable heart rate value of AnT; calculating, by the processor, a first probability factor (P1) for each of the plurality of accepted data points based on the measured external workload value and a calculated heart rate variability value of each data point; comparing, by the processor, the calculated first probability factor with at least one stored probability factor in a respective segment, and if the first probability factor is higher than the at least one stored probability factor, the stored probability factor is updated by the first probability factor; determining if a second probability factor (P2) needs to be created or updated, and calculating, by the processor, the second probability factor of each stored data point for each respective segment based on a deviation between change in recorded external workload and expected change in external workload; determining if a third probability factor (P3) needs to be created or updated, and calculating, by the processor, the third probability factor of each stored data point for each respective segment based on the recorded heart rate of the user in proportion to the maximum heart rate of the user; calculating, by the processor, a total probability (P) of each recorded heart rate in each segment using the first probability factor, the second probability factor, and the third probability factor wherein the total probability is derived from a sum of the first probability factor, the second probability factor, and the third probability factor, the sum being raised by the power of a constant; calculating, by the processor, an estimate of the user's AnT as a weighted value of heart rate values recorded for each segment, wherein each recorded heart rate value in each segment is multiplied by the total probability stored for that segment when all weighted values are added together; and outputting, to an output device, the estimate of the user's AnT. ### Claim 2 (depends on 1) The method of claim 1, wherein the first probability factor is calculated based on a first relationship between current measured external workload and highest measured external workload, and a second relationship between current measured heart rate variability level and lowest measured heart rate variability level; and wherein the method further comprises: generating and maintaining an index-based register, and storing, in the index-based register, the plurality of heart rate segments (i); and associating, in the index-based register, at a first index value, a particular heart rate segment (i) and the first probability factor following, in real-time, the measurement of the particular heart rate segment (i). ### Claim 3 (depends on 2) The method of claim 2, wherein the first probability factor (P1) is calculated according to the equation: P1=(1500−100*(vmax−v)−600*(MAD−MADmin))/15, wherein v=current measured external workload as at least one of current speed or velocity or power output, vmax=highest measured external workload as at least one of the highest speed or velocity or power output, MAD=current measured heart rate variability level, and MADmin=lowest measured heart rate variability level. ### Claim 4 (depends on 1) The method of claim 1, wherein the second probability factor and the third probability factor are based on population reference values. ### Claim 5 (depends on 1) The method of claim 1, wherein the second probability factor (P2) for a particular heart rate segment (i) is calculated according to the equation: ∑ i = 0 j - 1 ⁢ ( ( V j - V i ⁢ ( pHR j - pHR i ) * ⁢ maximalMET * 1.5625 - 1 ) / MaxFunc ) ( j ) wherein j is a respective segment index, i runs from 0 to j, Vi is a user's recorded external workload in segment (i), pHRi is a user's recorded heart rate in segment (i) relative to a maximum heart rate, and maximal MET is a user's maximal exercise capacity, and MaxFunc is a set limit function reducing the effect of old datapoints. ### Claim 6 (depends on 1) The method of claim 1, wherein the third probability factor (P3) is calculated according to the equation: P3(i)=100*(101−phr(i))/11, when phr(i) is greater than 90%; or P3(i)=100*(11+phr(i)−90)/11 when phr(i) is less than or equal to 90%; wherein phr(i) is a user's recorded heart rate proportional to the maximum heart rate of the user in segment (i). ### Claim 7 (depends on 1) The method of claim 1, wherein the estimate of the user's AnT as a weighted value of the recorded heart rates (HR) of each segment (i) is calculated according to the equation: AnT ⁡ ( t ) = ∑ ( HR ⁡ [ i ] * ( P ⁡ [ i ] y ) ∑ ( P ⁡ [ i ] y ) wherein P[i] is the total probability of each segment (i), and y is a chosen power value. ### Claim 8 (depends on 1) The method of claim 1, wherein the estimate of the user's AnT can be displayed to the user during the exercise or after the exercise. ### Claim 9 (depends on 1) The method of claim 1, wherein the estimate of the user's AnT can be displayed to the user as a value describing external workload. ### Claim 10 (depends on 1) The method of claim 1, wherein the plurality of heart rate segments include 5-30 segments. ### Claim 11 (depends on 10) The method of claim 10, wherein the size of each segment is 0.5-3% of the user's maximum heart rate. ### Claim 12 A system for determining anaerobic threshold intensity (AnT) of a user in a freely performed physical exercise, comprising: means for continuously measuring a physiological response of a user by heart rate, said means disposed on the body of the user, wherein a plurality of measured heart rate values are recorded with time stamps as heart rate data forming a recorded heart rate of the user; means for continuously measuring an external workload in terms of speed and inclination or measured power, wherein a plurality of measured workload values are recorded and each measured workload value is associated with one of the plurality of measured heart rate values to form a plurality of data points; means for calculating heart rate variability (HRV) values from intervals between successive heartbeats and storing each calculated HRV value associating to a data point; means for filtering one or more data points based on a predetermined criteria to form a plurality of accepted data points; means for classifying accepted data points within a plurality of heart rate segments (i) representing a heart rate within an anaerobic threshold (AnT) of the user, wherein at least one segment is below a probable heart rate value of AnT, and at least one segment is above the probable heart rate value of AnT, means for calculating a first probability factor (P1) for each of the plurality of accepted data points based on the measured external workload value and a calculated heart rate variability value of each data point; means for comparing the calculated first probability factor with at least one stored probability factor in a respective segment, and if the first probability factor is higher than the at least one stored probability factor, the stored probability factor is updated by the probability factor; means for determining whether a second probability factor (P2) needs to be created or updated, and for calculating the second probability factor of each stored data point for each respective segment based on a deviation between change in recorded external workload and expected change in external workload; means for determining whether a third probability factor (P3) needs to be created or updated, and for calculating the third probability factor of each stored data point for each respective segment based on the recorded heart rate of the user in proportion to the maximum heart rate of the user; means for calculating a total probability (P) of each recorded heart rate in each segment using the first probability factor, the second probability factor, and the third probability factor wherein the total probability is derived from a sum of the first probability factor, the second probability factor, and the third probability factor, the sum being raised by the power of a constant; means for calculating an estimate of the user's AnT as a weighted value of heart rate values recorded for each segment, wherein each recorded heart rate value in each segment is multiplied by the total probability stored for that segment when all weighted values are added together; means for outputting the estimate of the user's AnT to an output device. ### Claim 13 (depends on 12) The system of claim 12, wherein the second probability factor and the third probability factor are only calculated when the estimate of the user's AnT is required. ### Claim 14 (depends on 12) The system of claim 12, wherein the first probability factor is calculated based on a first relationship between current measured external workload and highest measured external workload, and a second relationship between current measured heart rate variability and lowest measured heart rate variability level. ### Claim 15 (depends on 12) The system of claim 12, wherein the estimate of the user's AnT can be displayed to the user during the exercise or after the exercise. ### Claim 16 (depends on 12) The system of claim 12, wherein the plurality of heart rate segments include between 5-30 consecutive 0.5-3% segments of the maximum heart rate of the user. ### Claim 17 A method for determining anaerobic threshold intensity (AnT) of a user in a freely performed physical exercise, comprising: continuously measuring, by a body-mounted heart rate sensor, a physiological response of a user by heart rate, wherein a plurality of measured heart rate values are recorded with time stamps as heart rate data forming a recorded heart rate of the user; continuously measuring an external workload in terms of speed and inclination or measured power, wherein a plurality of measured workload values are recorded and each measured workload value is associated with one of the plurality of measured heart rate values to form a plurality of data points; calculating, by a processor, heart rate variability (HRV) values from intervals between successive heartbeats and storing each calculated heart rate variability value associating to a data point; filtering, by a processor, one or more data points based on a predetermined criteria to form a plurality of accepted data points; classifying, by a processor, accepted data points within a plurality of heart rate segments (i) representing a heart rate within an anaerobic threshold (AnT) of the user, wherein at least one segment is below a probable heart rate value of AnT, and at least one segment is above the probable heart rate value of AnT, wherein a data point with highest probability is stored for each segment; calculating, by a processor, a first probability factor (P1) for each of the plurality of accepted data points based on at least the measured external workload value and a calculated heart rate variability value of each data point; retrieving, by the processor, from a set of stored probability data, at least one stored probability factor for a respective segment, the set of stored probability data further comprising a second probability factor (P2) and a third probability factor (P3); comparing, by a processor, the calculated first probability factor with the at least one stored probability factor in the respective segment, and if the first probability factor is higher than the at least one stored probability factor, the stored probability factor is updated by the first probability factor, and at least one further calculation is triggered to update stored probability data other than with the first probability factor; and calculating, by a processor, AnT using the stored probabilities in each segment; and outputting, to an output device, the estimate of the user's AnT to an output device. ### Claim 18 (depends on 17) The method of claim 17, further comprising: triggering a calculation of the at least one second probability factor (P2) based on chosen parameters of said data point if the first probability factor replaces the at least one stored probability factor; calculating a total probability of each recorded heart rate in each segment using the first probability factor and the at least one second probability factor; calculating an estimate of the user's AnT as a weighted value of heart rate value recorded for each segment, wherein each recorded heart rate value in each segment is multiplied by the total probability of that segment when all weighted values are added together. ### Claim 19 (depends on 18) The method of claim 18, wherein at least one second probability factor is only calculated when the estimate of the user's AnT is required. ### Claim 20 (depends on 17) The method of claim 17, wherein the first probability factor is calculated based on a first relationship between current external workload and highest measured external workload, and a second relationship between current and lowest measured heart rate variability level. ### Claim 21 (depends on 17) The method of claim 17, wherein the first probability factor (P1) is calculated according to the equation: wherein v=current measured external workload as at least one of current speed or velocity or power output, vmax=highest measured external workload as at least one of the highest speed or velocity or power output, MAD=current measured heart rate variability level, and MADmin=lowest measured heart rate variability level. ### Claim 22 (depends on 21) The method of claim 21, wherein the estimate of the user's AnT as a weighted value of the highest recorded heart rates (HR) of each segment (i) is calculated according to the equation: AnT ⁡ ( t ) = ∑ ( HR ⁡ [ i ] * ( P ⁡ [ i ] y ) ∑ ( P ⁡ [ i ] y ) wherein P[i] is the total probability of each segment (i) and y is a chosen power value. ### Claim 23 (depends on 17) The method of claim 17, wherein the at least one second probability factor is based on population reference values. ### Claim 24 (depends on 23) The method of claim 23, wherein the population reference values are based on a population average of heart rate level corresponding to AnT. ### Claim 25 (depends on 23) The method of claim 23, wherein the population reference values are based on a population average of heart rate level corresponding to AnT. ### Claim 26 (depends on 23) The method of claim 23, wherein the population reference values are based on a linear dependency between heart rate and external workload. ### Claim 27 (depends on 17) The method of claim 17, wherein the plurality of heart rate segments includes 5-30 segments. ### Claim 28 (depends on 27) The method of claim 27, wherein a size of each segment is 0.5-3% of the user's maximum heart rate. ### Claim 29 (depends on 17) The method of claim 17, wherein a final AnT estimate is calculated from at least two AnT estimates weighted by a chosen criteria. ### Claim 30 (depends on 17) The method of claim 17, wherein an AnT estimate is given to a user only if exercise heart rate reaches a chosen limit.