Equine monitoring method

Abstract

The present invention relates to an equine monitoring system in which, in order to acquire the basic data for using the disclosed system, a Bluetooth H10 sensor (1) is used, since it is a compact device that permits the precise and reliable acquisition of heart rate data. The device comprises an adjustable elastic strap (2) that is positioned around the animal's thorax (3) in line with the heart, and an electronic module (4) that is connected to the elastic strap (2). The module (4) contains a heart rate sensor (not shown), which detects the heartbeats and transmits the data via Bluetooth to mobile devices (5), such as smartphones, tablets or smartwatches.

Description

METHOD FOR MONITORING EQUINES FIELD OF THE INVENTION

[0001] In general, the present invention belongs to the technological sector of large animal veterinary medicine, more specifically equines, and it is, objectively, a method for monitoring equines through the combined use of Heart Rate Variability (HRV) and Global Positioning System (GPS) data, acquired from equines (athletes or not), using portable Bluetooth-type electro-measurement sensors during physical exertion or rest, in such a way that the acquired data undergoes processing in a dedicated algorithm, providing various medical parameters such as stress level, homeostasis, HRV parameters in the time and frequency domains, animal discomfort level, physical fitness, cardiovascular load, exertion level and exercise tolerance, but not limited to these. BACKGROUND OF THE INVENTION

[0002] The methods currently available for obtaining medical data from horses have serious shortcomings and technical problems with their applicability.

[0003] Note, for example, document NZ542682 “EQUINE FITNESS MONITORING”, which discloses an apparatus for monitoring the condition of a horse that includes a blanket (1) that has a first sensor (5, 6) including a detector and a processing module (3), where the first sensor (5, 6) is adapted to generate indicator data indicative of at least one health status indicator, and a second sensor (11) to generate position data indicative of the horse's position. During use, a processing system is adapted to determine, from the data of The system aims to measure the horse's position and movement rate, and determine its health status using the position and movement data. At least one battery is connected to the first and second sensors and to the first part of an inductive coupling. The inductive coupling is provided in a recess in the processing module where, in use, the battery is recharged by causing the first part of the inductive coupling to cooperate with a second part of the inductive coupling provided in a protrusion that is part of a suspension mechanism. The second part is coupled to an external power supply to allow the battery to be charged when the blanket is hung on the suspension mechanism.

[0004] However, this document describes a mechanism and method for assessing conditioning, based on the concept known as V200, as well as specifying a test protocol for measuring V200 itself. This concept is widely known in the literature, and the way the patent describes its measurement technique differs completely from the system now disclosed.

[0005] There is also document US8738117 “METHOD AND APPARATUS FOR PHYSIOLOGICAL ASSESSMENT IN MAMMALS” which discloses a method and tool for comparing mammals, such as horses, including the tool an electrode; an electronic instrument for measuring heartbeats; an analyzer to determine a result indicative of adaptability, reactivity or equanimity; and a plotting system to plot the temperament parameter or temperament quotient on a grid. The tool is used to provide a value for the sympathetic nervous system index (SNSI), the parasympathetic nervous system index (PNSI) or the standard deviation of the mean normal-to-normal intervals (SDMNN). This value is correlated with a selected mammalian trait.

[0006] Note that the aforementioned prior work describes a technique for analyzing data derived from certain HRV parameters, in order to correlate these parameters with the components of the autonomic nervous system, classifying responses according to specific patterns. However, the system now revealed, despite using HRV, presents a classification, method of use, and final results that differ from the aforementioned prior work.

[0007] Finally, document WO2010086753 “METHOD AND APPARATUS FOR PHYSIOLOGICAL ASSESSMENT IN MAMMALS” reveals a technique for analyzing data derived from certain HRV parameters, correlating these parameters with components of the autonomic nervous system and classifying responses according to specific patterns. Our technique also uses HRV statistical parameters. However, the classification, methods of use, and results have no relation to the technique described in the system revealed here.

[0008] Therefore, considering all the drawbacks of the systems and equipment currently used, described above in the state of the art, a gap is evident in the creation of a method for monitoring horses through the combined use of Heart Rate Variability (HRV) and Global Positioning System (GPS) data, acquired from horses (athletes or not) using portable Bluetooth-type electro-measurement sensors during physical exertion or rest, so that the data Acquired samples undergo processing using a dedicated algorithm, providing various medical parameters. SUMMARY AND OBJECTIVES OF THE INVENTION

[0009] The invention now revealed offers several advantages, including practical applications of various theories of equine exercise physiology. These advantages can be summarized as follows: quickly, simply, continuously, and at very low cost, it allows for determining a horse's stress, health, and discomfort levels; verifying gains or losses in physical fitness according to the training program applied to the horse; and scheduling training or rest periods for the animal based on previous responses and its physiological state estimated by the technique, thus avoiding a lack or excess of exercise during the training process.

[0010] The central objective of the proposed method is to analyze and identify horses in a fast, inexpensive, and non-invasive way, so that the trainer can work with the horse in the most favorable scenario possible, preventing injuries from exertion and repetitions in case of overtraining, preserving the animal's health and maximizing gains in physical conditioning.

[0011] Although veterinary monitoring is recommended, especially for athletic horses, this methodology provides a preliminary and daily overview of the health status of each monitored animal.

[0012] Therefore, with the aim of addressing the shortcomings of the current state of the art highlighted above, this patent application aims to propose a solution for methods of monitoring horses, in particular, through the combined use of Heart Rate Variability (HRV) information and Global Positioning System (GPS) data. BRIEF DESCRIPTION OF THE ATTACHED DRAWINGS

[0013] In order for the present invention to be fully understood and put into practice by any technician in this technological sector, it will be described in a clear, concise and sufficient manner, based on the attached drawings, which illustrate and support it, listed below:

[0014] Figure 1 represents the animal with the elastic band equipped with a sensor.

[0015] Figure 2 represents the RR interval histogram.

[0016] Figure 3 represents the RR interval histogram with baseline width.

[0017] Figure 4 represents the RR interval histogram with classification of the base width.

[0018] Figure 5 represents a graph of Frequencies vs. Magnitude.

[0019] Figure 6 represents a time-domain radar graph.

[0020] Figure 7 represents the graph of Homeostasis levels.

[0021] Figure 8 represents a Radar VFC graph in the Frequency Domain.

[0022] Figure 9 represents the stress level graph.

[0023] Figure 10 represents the SNA's balance sheet chart.

[0024] Figure 11 represents a graph of stress level over time.

[0025] Figure 12 represents a data analysis where the record is considered good in terms of stress level, heart health, and animal comfort.

[0026] Figure 13 represents an analysis of data whose record is considered normal, in terms of stress level, heart health and animal comfort.

[0027] Figure 14 represents an analysis of data whose record is considered normal with a slight tendency towards discomfort, in terms of stress level, cardiac health, and animal comfort.

[0028] Figure 15 represents a data analysis whose record shows that the horse is slightly tired and shows no signs of discomfort, in terms of stress level, cardiac health, and animal comfort.

[0029] Figure 16 represents a data analysis in which the overall stress level was classified as high, where the evidence suggests that the horse should have rested more in relation to the previous training session, due to signs of poor cardiac recovery.

[0030] Figure 17 represents a data analysis where the record is considered high.

[0031] Figure 18 represents a graph of heart rate zones and a proportion of zones worked during a workout.

[0032] Figure 19 represents the V200 curve.

[0033] Figure 20 represents the increase in the V200 curve.

[0034] Figure 21 represents a non-segmented example of V200.

[0035] Figure 22 represents a segmented example of V200.

[0036] Figure 23 represents a graph with the extrapolation lines of the V200 segmentation.

[0037] Figure 24 represents a graph of cumulative conditioning level.

[0038] Figure 25 represents a graph of effort versus tolerance.

[0039] Figure 26 represents a graphic record of a real training session considered "light," where the horse's heart rate worked exclusively in Zone 1.

[0040] Figure 27 represents a graphic record of a real training session considered "light," where the horse's heart rate worked in zones 1 and 2.

[0041] Figure 28 represents a record of a real training session considered "moderate," where the horse's heart rate worked in zones 1, 2, 3, and 4.

[0042] Figure 29 represents a record of a real training session considered "intense".

[0043] Figure 30 depicts the moment of veterinary medical intervention and its results.

[0044] Figure 31 represents the stress level, effort vs. tolerance, and conditioning level graphs of the horse from the previous example, as well as a representation of a data range over a one-year period.

[0045] Figure 32: Flowchart of the sensor data analysis methodology.

[0046] Figure 33: Flowchart of the methodology for analyzing data from GPS. DETAILED DESCRIPTION OF THE INVENTION

[0047] To acquire the basic data for using the system now disclosed, a Bluetooth sensor (1) is used, as it is a compact device which allows for the acquisition of heart rate data in a precise and reliable manner.

[0048] The device consists of an adjustable elastic band (2), which wraps around the animal's thorax (3), at the level of the heart, and an electronic module (4) that connects to the elastic band (2).

[0049] Module (4) contains a heart rate sensor (not shown), which detects heartbeats and transmits the data via Bluetooth to mobile devices (5), such as smartphones, tablets or smartwatches.

[0050] The sensor (1) is compatible with a wide range of applications, allowing it to monitor and analyze heart rate data in real time, as well as providing statistics, feedback and insights on the animal’s performance (3) during physical exercise, using long-lasting batteries, and is water resistant, allowing its use during activities involving water and / or sweat.

[0051] The monitoring system now revealed uses GPS data to collect important information about the animal’s performance (3) during physical exercise. This data includes information about speed, geographic position, distance traveled and training time.

[0052] This data allows the creation of graphs and reports that show the horse’s performance (3) during sports training. In addition, they can be used to identify more efficient training routes, avoiding unwanted physical overload of the animals.

[0053] Data acquisition is done through an application, and the data acquired at rest are the individual parameter base for each horse (3). It is from these that all training parameters are compared.

[0054] However, it is recommended, as a preferred method of execution, that a resting test be carried out between 6am and 9am, preferably every three months, for at least 20 min, with the animal (3) in a state of maximum relaxation, preferably performing its normal resting activities.

[0055] The objective of the resting test is to parameterize the animal (3) in a healthy and calm condition, without external interference or with signs of injury or illness. It is not always possible to meet all the requirements of a test at rest. Therefore, preference should be given to acquisition at times when the animal (3) is rested from previous training.

[0056] Data acquired outside of rest (during training) do not need to meet the criteria described above, as the idea is precisely to verify the animal's state (3) under exertion. In this case, it is preferable that training recordings last at least 15 minutes.

[0057] Note that, since this work is based on several statistical parameters, the larger the volume of data, the better the analysis will be.

[0058] Furthermore, the resting test can also be useful for providing a complete assessment of the animal (3) in terms of health, provided there is another resting data point serving as a basis for comparison.

[0059] In the first stage of analysis, a histogram is created with the acquired RR interval values. Only the RR interval values ​​are used in constructing this histogram.

[0060] To create the RR Interval histogram, the number of "Bins" (histogram bands or classes) is first calculated using the RR Interval data (time intervals between two consecutive QRS complexes on an electrocardiogram (ECG), representing the time elapsed between two consecutive heart contractions, used to assess heart rate regularity).

[0061] The number of "Bins" is defined as the square root of the number of elements present in RR intervals. Then, the mean and standard deviation of the values ​​in this column are calculated. It is A set of values ​​is generated for a variable X, ranging from 0 to 5000. These values ​​are used to calculate a probability density function (PDF) of the normal distribution based on the mean and standard deviation of the RR intervals column.

[0062] Calculating the probability density function of the normal distribution is used to find, within the set of values ​​of the variable X, the largest value that satisfies the condition that the PDF value at that point is greater than or equal to the constant 0.00000000098.

[0063] This value is called "X-value," and the "base width" is defined as the integer part of "X-value." The base width of the generated histogram is interpreted as a signal of effort level when the analysis focuses only on a single acquisition (without considering past acquisitions), or as a recovery level (when the analysis considers other analyses together).

[0064] The classification of the histogram's base width is based on a series of predefined limits. If the base width is less than 1527.5 ms, it is classified as "Very Short"; if it is between 1527.5 and 2902.25 ms, it is classified as "Short"; if it is between 2902.25 and 3207.75 ms, it is classified as "Compatible"; if it is between 3207.75 and 4582.5 ms, it is classified as "Wide"; and if it is greater than 4582.5 ms, it is classified as "Very Wide". The classification of the base width is called the "base width classification", see Table 1 below: Greater than 4582.5 Very Wide 6 Table 01 – Classification and Scoring of Base Width (ms).

[0065] Using the number of "Bins", the Heterogeneity value is calculated, which is defined as the number of "Bins" divided by 100.

[0066] The histogram heterogeneity classification uses the "heterogeneity" value, classifying it into one of three predefined categories. If the heterogeneity is less than 0.3, it is classified as "Low"; if it is between 0.3 and 0.5, it is classified as "Medium"; and if it is greater than or equal to 0.5, it is classified as "High". This classification is stored as the variable "heterogeneity classification".

[0067] Based on the heterogeneity classification, a score is assigned. If the heterogeneity is "low," the score is 0; if it is "medium," the score is 1; and if it is "high," the score is 2. The score is stored in the variable "heterogeneity score." Table 2 below summarizes the heterogeneity classification and scoring. Table 02 – Classification and scoring of heterogeneity.

[0068] In the generated histogram, the midpoint of each bin is calculated and stored. Then, a normal curve is fitted to the data distribution. Finally, the correlation coefficient between the... The histogram and the normal curve. The result of the correlation is stored as a variable called "histogram normality" with a value between 0 and 100.

[0069] The "histogram normality" value is classified into one of three predefined categories. If the normality is less than 70, the histogram is classified as "Not Compatible"; if it is between 70 and 90, it is classified as "Slightly Compatible"; and if it is greater than or equal to 90, it is classified as "Compatible".

[0070] If the histogram is "slightly compatible," the score is 1; if it is "compatible," the score is 2; and if it is "not compatible," the score is 0. This classification is called the "Normality Score." Table 03 summarizes the Histogram Normality Classification and Score. Table 03 – Classification and Scoring of Histogram Normality

[0071] The final evaluation of the RR Interval Histogram is based on the sum of the scores for base width, heterogeneity, and normality. Histogram scoring is an important step in the stress assessment process in horses. Based on this score, classification is made as follows: "very bad," "bad," "average," "good," and "very good." This classification is based on a defined scoring scale, as follows: Histogram score below 3: "very bad". Histogram score between 3 and 5: "bad". Histogram score between 5 and 6: "average". Histogram score between 6 and 8: "good". Histogram score above 8: "very good".

[0072] After classifying the score, it is possible to define the horse's stress level, with the proportion estimated by the Histogram: Score classification as "very bad": "VERY HIGH" stress; Score classification as "bad": "HIGH" stress; Score classification as "average": "NORMAL" stress; Score classification as "good": "LOW" stress; Score classification as "very good": "VERY LOW" stress.

[0073] Table 4 below summarizes the Histogram Score Classification (sum of the points for Base Width, Heterogeneity, and Normality of the Histogram), and the Stress Level Classification by Histogram: Table 04 – Histogram Score Classification (sum of points for Base Width, Heterogeneity, and Normality of the Histogram), and Stress Level Classification by Histogram.

[0074] Time-domain RR interval data analysis is a technique used to assess heart rate variability (HRV).

[0075] This technique involves measuring the RR interval and analyzing its variation over time. Analysis of RR intervals allows for the assessment of the autonomic balance of the nervous system and provides information about the sympathetic and parasympathetic activity of the heart, helping to detect cardiovascular dysfunctions, respiratory disorders, neurological problems, and other conditions.

[0076] Furthermore, the analysis of RR interval data can aid in decision-making when monitoring the effectiveness of treatments in patients with cardiovascular conditions.

[0077] That said, the RR Interval data is divided into 2 equal parts (Part One and Part Two) in order to facilitate understanding of the evolution of the analysis.

[0078] The data volume is divided in a way that respects multiples of 256; that is, each part of the divided data will have at least 256 points, with the possibility of overlapping data from the end of Part One. and the beginning of the second part, so that there is always a total of analyzable data that is a multiple of 256.

[0079] For the technique described in this system, a limit of 4096 maximum data points was set for each part, meaning that the total limit of possible data points to be analyzed at a time should be 8192 total RR interval data points.

[0080] In the analysis of RR interval data in the time domain, there are several parameters that can be evaluated, each providing different information about heart rate variability. The definitions for each of these parameters are described below:

[0081] RRMED (Average RR Interval): is the average of the RR intervals measured over a given period. It is a parameter that provides a general measure of heart rate variability (Eq. 01). Eq. 01 – Equation for the Average RR Interval, where RRi are the measured RR intervals and n is the total number of RR intervals measured in the period.

[0082] SDNN (Standard Deviation of RR Intervals): is the measure of the standard deviation of RR intervals over a given period. This parameter provides a measure of the total heart rate variability, representing the sum of short-term and long-term variations (Eq. 02). Eq. 02 – Equation for the Standard Deviation of RR Intervals, where N is the total number of RR intervals, RRi is the value of the i-th RR interval, and RR is the mean of the RR intervals.

[0083] SDANN (Standard Deviation of Moving Average RR Intervals): is the measure of the standard deviation of the moving averages of the RR intervals over a given period. This parameter provides a measure of the long-term variability of heart rate (Eq. 03). Eq. 03 – Equation for the Standard Deviation of RR Moving Average Intervals, where N is the total number of moving averages calculated, Mi is the value of the i-th moving average of the RR intervals, and M is the average of the moving averages of the RR intervals.

[0084] SDNNi (Standard Deviation of RR Interval Differences): is a measure of the standard deviation of the differences between adjacent RR intervals over a given period. This parameter provides a measure of short-term heart rate variability (Eq. 04). Eq. 04 – Equation for the Standard Deviation of RR Difference Intervals, where N is the total number of RR intervals, and RRi and RRi+1 are the values ​​of adjacent RR intervals.

[0085] RMSSD (Root Mean Squared Difference of RR Intervals): is the measure of the square root of the average of the squares of the differences between adjacent RR intervals over a given period. This parameter provides a measure of short-term heart rate variability, being especially sensitive to parasympathetic activity (Eq. 05).

[0086] Eq. 05 – Equation for the Square Root Mean of the Squares of the Differences of the RR Intervals, where N is the total number of RR intervals and RRi+1 are the values ​​of adjacent RR intervals.

[0087] pNN50 (Percentage of Differences of RR Intervals Greater than 50 ms): is the measure of the percentage of differences between adjacent RR intervals that are greater than 50 ms over a given period. This parameter provides a measure of short-term heart rate variability, being especially sensitive to parasympathetic activity (Eq. 06).

[0088] Eq. 06 – Equation for the Percentage Difference of RR Intervals Greater than 50 ms, where N is the total number of RRi intervals and RRi+1 are the values ​​of adjacent RR intervals.

[0089] Limits were defined for classifying Training and Test data using points for each of the six Time Domain parameters.

[0090] For the RRMED of training, the classification of this indicator is on a scale of 1 to 5, based on pre-established value ranges. If the RRMED value is below 444.55, it will be classified as 1. If it is between 444.55 and 844.55, it will be classified as 2. If it is between 844.55 and 933.45, it will be classified as 3. If it is between 933.45 and 1333.5, it will be classified as 4. If RRMED is equal to or greater than 1333.5, it will be classified as 5. Table 05, below, summarizes the classification of the RRMED of Training according to its value. Table 05 – RRMED Classification of Training.

[0091] For the RRMED test, the classification of this indicator is on a scale of 1 to 5, based on pre-established value ranges. If the RRMED test value is below 690.5, it will be classified as 1. If it is between 690.5 and 1311.95, it will be classified as 2. If it is between 1311.95 and 1450.05, it will be classified as 3. If it is between 1450.05 and 2071.5, it will be classified as 4. If RRMED is equal to or greater than 2071.5, it will be classified as 5. Table 06 summarizes the classification of the RRMED test based on its value. Table 06 – RRMED Classification of the Test.

[0092] For the SDNN indicator of training, the classification of this indicator is on a scale of 1 to 5, based on pre-established value ranges. If the SDNN value is below 255, it will be classified as 1. If it is between 255 and 484.5, it will be classified as 484.5. If it is classified as 2, it will be classified as 3. If it is between 484.5 and 535.5, it will be classified as 4. If it is equal to or greater than 765, it will be classified as 5. Table 07 summarizes the classification of the Training SDNN based on its value. Table 07 – SDNN Classification of the Training.

[0093] For the SDNN indicator of the test, the classification of this indicator is on a scale of 1 to 5, based on pre-established value ranges. If the SDNN value of the test is below 572.7, it will be classified as 1. If it is between 572.7 and 1088.13, it will be classified as 2. If it is between 1088.13 and 1202.67, it will be classified as 3. If it is between 1202.67 and 1718.1, it will be classified as 4. If the SDNN is equal to or greater than 1718.1, it will be classified as 5. Table 08 summarizes the classification of the SDNN of the Test according to its value. Table 08 – SDNN Classification of the Test.

[0094] For the SDANN training indicator, classification is done on a scale of 1 to 5, based on pre-established value ranges. If the SDANN value is below 150, it will be classified as 1. If it is between 150 and 285, it will be classified as 2. If it is between 285 and 315, it will be classified as 3. If it is between 315 and 450, it will be classified as 4. If SDANN is equal to or greater than 450, it will be classified as 5. Table 09 summarizes the classification of SDANN training based on its value. Table 09 – SDANN Classification of the Training.

[0095] For the SDANN indicator of the test, the classification is made on a scale of 1 to 5, based on pre-established value ranges. If the SDANN value of the test is below 621.5, it will be classified as 1. If it is between 621.5 and 1180.85, it will be classified as 2. If it is between 1180.85 and 1305.15, it will be classified as 3. If it is between 1305.15 and 1864.5, it will be classified as 4. If SDANN is equal to or greater than 1864.5, it will be classified as 5. Table 10 summarizes the classification of the SDANN of the Test according to its value. Table 10 – SDANN Test Classification.

[0096] For SDNNi, the classification of this indicator is on a scale of 1 to 5, based on pre-established value ranges. If the SDNNi value is below 229, it will be classified as 1. If it is between 229 and 435.1, it will be classified as 2. If it is between 435.1 and 480.9, it will be classified as 3. If it is between 480.9 and 687, it will be classified as 4. If SDNNi is equal to or greater than 687, it will be classified as 5. In the case of SDNNi, the training and test parameters are the same. Table 11 summarizes the classification of SDNNi, for both training and test data, based on its value. Table 11 – SDNNi Classification from Training and Testing.

[0097] For the RMSSD indicator of training, the rating is on a scale of 1 to 5, based on pre-established value ranges. If the RMSSD value is below 290, it will be rated as 1. If it is between 290 and 551, it will be rated as 2. If it is between If the value is between 551 and 609, it will be classified as 3. If it is between 609 and 870, it will be classified as 4. If the RMSSD is equal to or greater than 870, it will be classified as 5. Table 12 summarizes the RMSSD classification for training data based on its value. Table 12 – Training RMSSD Classification.

[0098] For the RMSSD indicator of the test, the classification is on a scale of 1 to 5, based on pre-established value ranges. If the RMSSD value of the test is below 767.05, it will be classified as 1. If it is between 767.05 and 1457.395, it will be classified as 2. If it is between 1457.395 and 1610.805, it will be classified as 3. If it is between 1610.805 and 2301.15, it will be classified as 4. If the RMSSD is equal to or greater than 2301.15, it will be classified as 5. Table 13 summarizes the RMSSD classification for the Test data based on its value. Table 13 – RMSSD Test Classification

[0099] For the pNN50 of the training data, the value must be multiplied by 1500, and only then will the classification of this indicator be on a scale of 1 to 5, based on pre-established value ranges. If the pNN50 value is below 314.5, it will be classified as 1. If it is between 314.5 and 597.55, it will be classified as 2. If it is between 597.55 and 660.45, it will be classified as 3. If it is between 660.45 and 943.5, it will be classified as 4. If pNN50 is equal to or greater than 943.5, it will be classified as 5. Table 14 summarizes the pNN50 classification for the training data based on its value. Table 14 – pNN50 Classification of the Training.

[0100] For the pNN50 of the test, the value must be multiplied by 1500, and only then will the classification of this indicator be on a scale of 1 to 5, based on pre-established value ranges. If the pNN50 value of the test is below 589.5, it will be classified as 1. If it is between 589.5 and 1120.05, it will be classified as 2. If it is between 1120.05 and 1237.95, it will be classified as 3. If it is between 1237.95 and 1768.5, it will be classified as 4. If pNN50 is equal to or greater than 1768.5, it will be classified as 5. Table 15 summarizes the pNN50 classification for the Test data as a function of its value. Table 15 – pNN50 Classification of the Test.

[0101] Figure 5 shows a radar chart of HRV in the Time Domain. This is an example of visualizing Training Time Indicator data compared to Test Time Indicators, which becomes a standardized reference for horse monitoring.

[0102] The final Time Domain indicator will be the Sum of Time Indicators, which is the sum of all Time Domain indicators from the training data, classified on a scale of 1 to 5, based on pre-established value ranges. If the value of the sum of the time domain indicators is less than 1550, it will be classified as 1. If it is between 1550 and 2945, it will be classified as 2. If it is between 2945 and 3255, it will be classified as 3. If it is between 3255 and 4650, it will be classified as 4. If the sum of the time domain indicators is equal to or greater than 4650, it will be classified as 5. Table 16 summarizes the classification of the Sum of Time Indicators for the Training data based on its value. Table 16 – Classification of the Sum of Training Time Indicators.

[0103] For the Sum of Test Time Indicators, which is the sum of all indicators in the Test Time Domain, classified on a scale of 1 to 5, based on pre-established value ranges. If the value of the sum of the time domain indicators is less than 3760.2, it will be classified as 1. If it is between 3760.2 and 7144.38, it will be classified as 2. If it is between 7144.38 and 7896.42, it will be classified as 3. If it is between 7896.42 and 11280.6, it will be classified as 4. If the sum of the time domain indicators is equal to or greater than 11280.6, it will be classified as 5. Table 17 summarizes the classification of the Sum of Time Indicators for the Test data based on its value. Table 17 – Classification of the Sum of Test Time Indicators.

[0104] Finally, the Time Domain Indicator Score is calculated for both training and test time indicators, which can range from 0 to 10, using the following equation (Eq.07): Eq. 07 – Time Domain Indicator Scoring Equation, which is applied to both Training and Test data.

[0105] Applying Eq. 07 to the test data generates a value that becomes a comparative reference in relation to the evaluation of the training values.

[0106] To perform RR interval analyses in the Frequency Domain, it is necessary to extract information from the prepared database using the same division of data into First Part and Second Part used for Time Domain analysis. Each part of the data preserves the Time Count of Each Record (ms) and RR Interval (ms) information. This same logic must be followed with the data relating to the resting test, in order to create reference parameters in the Frequency Domain.

[0107] In the cited data, the Fast Fourier Transform (FFT) calculation is performed, which is applied to a wave signal in order to decompose the signal into its frequency components, allowing for spectral analysis of the signal. The transform is calculated Using a Fast Fourier Transform (FFT) algorithm, this function returns a vector of complex numbers representing the frequencies and amplitudes of the signal's frequency components. Then, the magnitude of the complex values ​​is calculated using the modulus of these complex numbers, resulting in a vector representing the amplitudes of the signal's frequency components.

[0108] The magnitude values ​​are added to the data in the form of a new column called "FFT". It is worth noting that the FFT calculation is performed to allow for a spectral analysis of the wave signal, identifying the distribution of frequency components in the signal and providing relevant information for frequency domain analysis.

[0109] Using the data mentioned above, Frequency vs. Magnitude (ms² / Hz) graphs are generated for different parts of the dataset, with the frequency zones. The x-axis represents the frequency in Hertz (Hz) and the y-axis represents the magnitude in ms² / Hz.

[0110] Each graph has three distinct areas, representing the three frequency zones of the RR Interval data. The first zone is the Very Low Frequency zone, ranging from 0 to 0.04 Hz. The second zone is the Low Frequency zone, ranging from 0.04 to 0.15 Hz. The third zone is the High Frequency zone.

[0111] The frequency zones ranging from 0 to 0.04 Hz (Very Low Frequency - VLF) represent signals generated by the sympathetic and parasympathetic nervous systems, and primarily reflect the modulation of vagal tone in the heart (E. Von Borell et al, 2007). This zone has been associated with conditions of emotional and physical stress, as well as cardiovascular diseases.

[0112] The frequency zone ranging from 0.04 to 0.15 Hz (Low Frequency - LF) is associated with the neuro-hormonal regulation of heart rate. This zone has been used as an indicator of sympathetic nervous system activity.

[0113] The frequency zone ranging from 0.15 to 0.4 Hz (High Frequency - HF) is associated with the modulation of vagal tone and, to a lesser extent, sympathetic regulation. Activity in this area has been associated with cardiovascular regulation as well as respiratory control.

[0114] The graphs have a logarithmic scale on the y-axis for better data visualization. The x-axis is defined to range from 0 to 0.4 Hz.

[0115] It is defined that three variables: VLF, LF, and HF, represent the amount of energy of the electrocardiographic signal in each of the specified frequency zones. The amount of energy is measured as the sum of the magnitudes of each RR interval (in ms² / Hz).

[0116] For each of the three variables, a 100% normalization of the frequency component relative to the total energy is performed, creating normalized variables: VLFn, LFn, and HFn, which indicate the percentage of each component relative to the total sum of the three.

[0117] These three variables are considered the basis for interpreting Heart Rate Variability (HRV), helping to interpret the responses of the Autonomic Nervous System in horses.

[0118] Each of the three normalized variables has its values ​​classified according to its difference from... Standard reference values. The reference values ​​are: 44 for HFn, 23 for LFn, and 35 for VLFn.

[0119] Using the respective values ​​of each variable, the difference between this value and the normalized variable is calculated, and the modulus (or absolute value) of this result is classified as follows: HFn For values ​​of |44 - HFn| less than or equal to 2, the classification of HFn (Class_HFn) is “Normal”; For values ​​of |44 - HFn| greater than 2 and less than or equal to 4, the classification of HFn is “Smaller”; For values ​​of |HFn - 44| greater than 2 and less than or equal to 4, the classification of HFn is “Larger”; For values ​​of |44 - HFn| greater than 4, the classification of HFn is “Much Smaller”; For values ​​of |HFn - 44| greater than 4, the classification of HFn is “Much Larger”. Table 18 – Classification of Normalized High Frequency (HFn). LFn For values ​​of |23 - LFn| less than or equal to 2, the LFn classification (Class_LFn) is “Normal”; For values ​​of |23 - LFn| greater than 2 and less than or equal to 4, the LFn classification is “Smaller”; For values ​​of |LFn - 23| greater than 2 and less than or equal to 4, the LFn classification is “Larger”; For values ​​of |23 - LFn| greater than 4, the LFn classification is “Much Smaller”; For values ​​of |LFn - 23| greater than 4, the LFn classification is “Much Larger”. Table 19 – Normalized Low Frequency (LFn) Classification. VLFn For values ​​of |35 - VLFn| less than or equal to 2, the VLFn classification (Class_VLFn) is “Normal”; For values ​​of |35 - VLFn| greater than 2 and less than or equal to 4, the VLFn classification is “Lower”; For values ​​of |VLFn - 35| greater than 2 and less than or equal to 4, the VLFn classification is “Higher”; For values ​​of |35 - VLFn| greater than 4, the VLFn classification is “Much Lower”; For values ​​of |VLFn - 35| greater than 4, the VLFn classification is "Much Larger". Table 20 – Normalized Very Low Frequency (VLFn) Classification

[0120] The HFn, LFn, and VLFn classifications generate scores that are stored as the variables Pont_HFn, Pont_LFn, and Pont_VLFn, where “Much Lower” equals 1, “Lower” equals 2, “Normal” equals 3, “Lower” equals 4, and “Much Higher” equals 5. The classifications in text form are stored to compose the summary text of the analysis, while the scores are used in the stress assessment metrics. Table 21 summarizes the scores for the HFn, LFn, and VLFn classifications. Table 21 – Scores for the HFn, LFn and VLFn Classifications.

[0121] Using the HRV variables from both parts of the training and the resting test, a three-axis (radar-type) pie chart is created, spaced 120° apart (Fig. 10). Each axis represents a frequency range (HF, LF, and VLF) and has its limits varying from 0 to 60%. The values ​​of the three variables mentioned above form triangles, where we have the two parts of the training compared to the standard HRV reference from the Resting Test. The graph generated with the description above is the Heart Rate Variability in the Frequency Domain, as can be seen in Figure 7, which presents a radar chart with the Heart Rate Variability variables in the Frequency Domain, both from the Resting Test and from the two parts of the Training.

[0122] To calculate the "Homeostasis" variable, it is necessary to first define other variables, which are: "training_1", "training_2", and "test".

[0123] The variables “training_1” and “training_2” are derived from a data matrix formed by the values ​​of VLFn, LFn, and HFn from the first and second parts of the training records. Using this data, the Euclidean distance between the aforementioned values ​​and their corresponding values ​​in the matrix formed by the values ​​of VLFn_test, LFn_test, and HFn_test is calculated.

[0124] The results of the Euclidean distance calculations for parts 1 and 2 of the training are stored as the variables “euclidean_dist_1” and “euclidean_dist_2”. The product of these variables by 100 generates the variables “Homeostasis_1” and “Homestasis_2”, referring to the two parts of the training.

[0125] The homeostasis variables are classified textually (Class_homeostase_1 and Class_homeostase_2), according to their values, as follows: If “Homeostasis” is less than or equal to 3, the “Class_homeostase” is “Very Good”. If “Homeostasis” is greater than 3 and less than or equal to 5, the “Class_homeostase” is “Good”. If “Homeostasis” is greater than 5 and less than or equal to 10, the “Class_homeostase” is “Average”. If “Homeostasis” is greater than 10 and less than or equal to 20, the “Class_homeostase” is “Poor”. If “Homeostasis” is greater than 20, the “Class_homeostase” is “Very Poor”.

[0126] Homeostasis classifications receive scores according to the class, stored in the form of the variables “Pont_homeostase_1” and “Pont_homeostase_2”. The scoring rules are as follows: If the classification is “Very Good”, the score is 5. If the classification is “Good”, the score is 4. If the classification is “Average”, the score is 3. If the classification is “Poor”, the score is 2. If the classification is “Very Poor”, the score is 1.

[0127] The classification and scoring of Homeostasis can be summarized as shown in Table 22 below: Table 22 – Classifications and Scores for Homeostasis.

[0128] Based on the values ​​of the Homeostasis_1 and Homeostasis_2 variables, a graph is generated divided into homeostasis levels, following the criteria described previously for the Class_homeostasis_1 and Class_homeostasis_2 variables, where the scores of the Pont_homeostasis_1 and Pont_homeostasis_2 variables position markers on a graph that has an axis ranging from 0 to 100, as can be seen in Figure 8, whose graph presents the scale of homeostasis levels for the two parts of a workout.

[0129] Stress level measurements can be represented by the following equation, using the previously provided scoring variables: Eq. 08 – Stress Level Equations for the First and Second Parts of a Workout, based on the scores of the Time and Frequency Domain components of each part of the workout, and the Histogram score.

[0130] Using the Stress Level 1 and 2 equations above, an average stress level (Stress_Level) is calculated between the two variables. Stress_Level is then classified to transform its values ​​into text, following this logic: If Stress_Level is less than 2, then the classification (Class_Stress_Level) is “Very High”. If Stress_Level is greater than or equal to 2 and less than 4, then the classification (Class_Stress_Level) is “High”. If Stress_Level is greater than or equal to 4 and less than 6, then the classification (Class_Stress_Level) is “Medium”. If Stress_Level is greater than or equal to 6 and less than 8, then the classification (Class_Stress_Level) is “Low”. If Stress_Level is greater than or equal to 8 and less than 10, then the classification (Class_Stress_Level) is “Very Low”. Table 23 – Stress Level Classification.

[0131] Based on the values ​​of the variables Stress_Level_1 and Stress_Level_2, a graph is generated divided into stress zones, following the criteria described previously for the variables Class_Stress_Level, where the scores of the variables Stress_Level_1 and Stress_Level_2 position markers on a graph that has an axis ranging from 0 to 10, as can be seen in Figure 9, whose graph shows the stress levels for the two parts of a workout.

[0132] The Autonomic Nervous System (ANS) Balance can be assessed through the ratio between the variables LFn_1 / HFn_1 and LFn_2 / HFn_2, creating the variables Balanco_SNA_1 and Balanco_SNA_2, respectively. Using the values ​​of these variables, a classification is made that generates the variables (Class_balanco_SNA_1 and 2). This classification defines whether the training segments are in the “Parasympathetic” state, for values ​​greater than or equal to 0.515; “Central” state, for values ​​between 0.515 and 0.485; and “Sympathetic” state, for values ​​less than or equal to 0.485. Table 24 represents the classification of the Autonomic Nervous System Balance. Table 24 – Classification of the condition of the Autonomic Nervous System (ANS), obtained from the value of the ratios between LFn / HFn of the two parts of a training session.

[0133] Based on the values ​​of the variables Balanco_SNA_1 and Balanco_SNA_2, a graph is generated divided into two zones of the ANS, sympathetic and parasympathetic, following the criteria described previously for the variables Class_balanco_SNA_1 and Class_balanco_SNA_2, where the scores of the variables Balanco_SNA_1 and Balanco_SNA_2 position markers on a graph that has an axis ranging from 0 to 1, as observed in Figure 10, whose graph presents the classification of the balance of the autonomic nervous system (ANS), showing the "Sympathetic" and "Parasympathetic" positions for the two parts of a workout.

[0134] Every time the complete code is run, the analysis data is stored, so that this data is archived in only one repository. The data consists of the values ​​of the many variables that were created throughout the description above. In summary, the variables are as follows: 'Data', 'Cavalo', 'Tipo de Treino', 'Largura_Base', 'RRMED_1', 'RRMED_2', 'SDNN_1', 'SDNN_2', 'SDANN_1', 'SDANN_2', 'SDNNi_1', 'SDNNi_2', 'RMSSD_1', 'RMSSD_2', 'pNN50_1', 'pNN50_2', 'pNN50_1', 'pNN50_2', 'Soma_Ind_Tempo_1', 'Soma_Ind_Tempo_2', 'VLFn_1', 'VLFn_2', 'LFn_1', 'LFn_2', 'HFn_1', 'HFn_2', 'Homeostase_1', 'Homeostase_2', 'Nivel_estresse_1', 'Stress_Level_2'.

[0135] The function of cumulatively archiving parameters that characterize stress in horses is extremely useful, both in the daily analysis of the animals' conditions and in analysis over time, allowing one to verify the type of evolution or regression of stress. animal situation. An example of this type of view over time is shown in Figure 11, whose graph presents the stress level of a horse, showing the variations in stress recorded over a period of 1 year.

[0136] The entire description of the system presented here depends on an analysis of the variables involved in this work as a whole. Therefore, these analyses are exemplified below, using a series of records from real training sessions, with their responses and respective interpretations, revealing preferred execution models, but the present system is not limited to this.

[0137] Figure 12 represents a record considered good in terms of stress level, heart health, and animal comfort.

[0138] In this case, the balance of the autonomic nervous system (ANS) remained in the parasympathetic position during the first part of the training, but shifted to sympathetic, reflecting active functions. It can be interpreted that there was an increase in activity throughout the training, as if there were an increase in fatigue or stress. The first part of the training showed an excellent level of homeostasis, shifting to a medium level in the final part of the training.

[0139] The normal distribution of heart rate variability has a shape consistent with expectations, showing high heterogeneity of RR intervals. The base of the histogram has a width of 5000ms, which is considered a very wide base compared to the standard reference value for good cardiac health and a good stress level, which is 3055ms. The combined evaluation of the histogram components was considered very good.

[0140] The overall stress level was classified as can be seen in Figure 13, which reveals the record considered normal in terms of stress level, heart health, and animal comfort.

[0141] In the Example 02 training session, the Autonomic Nervous System (ANS) balance remained in the Central position throughout the training. This is the ideal position, reflecting a balanced autonomic nervous system.

[0142] The first part of the training showed a good level of homeostasis, which decreased to a medium level in the final part of the training. The HRV responses in the Frequency Domain were very close to what was expected.

[0143] The normal distribution of heart rate variability has a shape that is slightly consistent with expectations, showing high heterogeneity of RR intervals. The base of the histogram has a width of 3323ms, which is considered a wide base compared to the standard reference value for good cardiac health and a good level of stress, which is 3055ms. The combined assessment of the histogram components was considered good.

[0144] The overall stress level was classified as medium, as shown in Figure 14, which reveals a normal record with a slight tendency towards discomfort, in terms of stress level, heart health, and animal comfort.

[0145] In the training session of Example 03, the Autonomic Nervous System (ANS) balance remained in the Central position during the first part of the training, shifting to Sympathetic in the second part. This change in ANS balance may indicate that the horse was in good pre-training condition, becoming more alert accordingly. As the exercise progresses, having the ANS (Autonomic Nervous System) balance slightly off from the central position can be interpreted as a sign that the workout was performed in a healthy manner.

[0146] Throughout the training, there were average signs of homeostasis, without major changes. However, in the Frequency Domain, during the first part of the training, there were indications of increased activity related to the Very Low Frequency (VLF) range, reflecting situations related to hormonal activity in the ANS and / or thermoregulation (sweating).

[0147] The normal distribution of heart rate variability has a shape consistent with expectations, showing high heterogeneity of RR intervals. The base of the histogram has a width of 3308ms, which is considered a wide base compared to the standard reference value for good cardiac health and a good stress level, which is 3055ms. The combined evaluation of the histogram components was not considered very good only due to a slight deviation from the Normalized Trend around 2000ms. This response, together with the trend towards VLF, is a slight indicator of discomfort during training, especially in the initial part.

[0148] The overall stress level was classified as medium, as shown in Figure 15, which presents a record of a slightly tired horse with no signs of discomfort in terms of stress level, cardiac health, and animal comfort.

[0149] In the Example 04 training session, the Autonomic Nervous System (ANS) balance remained in the Central position throughout the training. This is the ideal position, reflecting a balanced autonomic nervous system.

[0150] The first part of the training showed a poor level of homeostasis, improving in the final part of the training. Heart rate variability (HRV) in the frequency domain showed a strong tendency towards the free-flow velocity (FVT) range. This type of tendency is usually interpreted as a strong hormonal influence (cortisol) on the autonomic nervous system, and / or situations of high animal thermoregulation.

[0151] The normal distribution of heart rate variability has a shape that is slightly consistent with expectations, showing average heterogeneity of RR intervals. The base of the histogram has a width of 2152ms, which is considered a short base compared to the standard reference value for good cardiac health and a good stress level, which is 3055ms. The combined assessment of the histogram components was considered poor, even without deviations from the Normalized Trend.

[0152] As can be seen in Figure 16, the overall stress level was classified as high, where the evidence suggests that the horse should have rested more in relation to the previous training session, due to signs of poor cardiac recovery.

[0153] In Example 05, the Autonomic Nervous System (ANS) balance remained in a balanced position during the first part of the training, but shifted to a sympathetic position in the final part, reflecting an increase in stress levels.

[0154] The first part of the training showed a poor level of homeostasis, improving to a medium level in the final part of the training. Heart rate variability (HRV) in the frequency domain showed the dominant trend. This type of trend tends to be interpreted as a sign of stress, with hormonal action (cortisol) in the autonomic nervous system, and / or thermoregulation issues in the horse.

[0155] The normal distribution of heart rate variability has a shape that is slightly consistent with expectations, but with a loss of correlation around 2000 ms. It showed high heterogeneity in RR intervals, and the histogram base has a width of 2797 ms, which is considered a short base compared to the standard reference value for good cardiac health and a good stress level, which is 3055 ms. The combined evaluation of the histogram components was considered average, where the combination of indicators demonstrates signs of fatigue and mild animal discomfort.

[0156] Continuous and prolonged responses, such as in Example 04, may suggest some type of infection or injury, which the animal may be tolerating in the short term, but which require veterinary investigation.

[0157] As can be seen in Figure 17, the overall stress level was classified as high.

[0158] In Example 06, the Autonomic Nervous System (ANS) balance remained in the Sympathetic position throughout the training. This position reflects the activation of bodily mechanisms for active situations (fight, flight, attention, etc.). It can be seen as a sign of stress.

[0159] Homeostasis levels were very poor, demonstrating that something is wrong with the horse's autonomic nervous system. The triangle representing heart rate variability (HRV) in the frequency domain showed a strong tendency to deviate from the standard resting response towards very low frequencies (VLF), indicating that hormonal activity and thermoregulation occurred throughout the training.

[0160] The normal distribution of heart rate variability has a shape that is not compatible with what is expected, showing high heterogeneity of the RR intervals. The base of the histogram has a width of 1986ms, which is considered a short base compared to the standard reference value for good cardiac health and a good level of stress, which is 3055ms. The combined evaluation of the histogram components was considered poor.

[0161] When all indicators are observed together, there is a clear sign of non-compliance with a healthy resting pattern. This type of response raises an alert that something needs to be investigated with the help of a veterinary professional. It is not recommended to continue training under these conditions. The overall stress level was classified as high.

[0162] The examples presented earlier were placed in a gradual manner of stress level and health conditions, starting in a fairly comfortable state and ending with a set of warning signs. The interpretation of the results needs to take into account the efforts made by the horse before, during, and after training, in order to plan the best way to handle the horse.

[0163] In order to understand the horse's exertion level, tolerance to exertion, and to plan new training, rest, or veterinary care, it is necessary to move on to the next topic, which involves entering GPS data.

[0164] GPS data includes information on training time, distance traveled, average speed, and coordinates (latitude and longitude). This data is available through apps from Polar or other brands, such as Garmin, for example.

[0165] Initially, it is necessary to define the horse's maximum heart rate, either through previously acquired data or from an estimated average value from other horses (220 bpm is an acceptable value). It is with this maximum heart rate data that the five Heart Rate Zones (Z1, Z2, Z3, Z4 and Z5) are defined (BITSCHNAU et al (2013)), where: Z1 is the Recovery Zone, and includes heart rates below 60% of maximum heart rate; Z2 is the Moderate Zone, and includes heart rates between 60% and 70% of maximum heart rate; Z3 is the Aerobic Zone, and includes heart rates between 70% and 80% of maximum heart rate; Z4 is the Borderline Zone, and includes heart rates between 80% and 90% of maximum heart rate; Z5 is the Maximum Zone, and includes heart rates above 90% of maximum heart rate. Table 25 – Classification of Heart Rate Zones, based on the percentage of heart rate relative to the maximum heart rate already recorded by the horse.

[0166] After this definition, the GPS data is grouped into a database containing: Time (date and time of each record per second); Latitude; Longitude; Altitude (m); Heart Rate (bpm); Speed ​​(km / h); Distance (km).

[0167] With the data mentioned above, it is possible to obtain the three main pieces of information from GPS data, which are: Cardiovascular Load, Percentage of Training in Each Heart Rate Zone, and Fitness Level.

[0168] In this system, it is necessary to estimate the level of cardiac effort to which the horse is subjected during exercise. To this end, a GPS data processing sequence was developed to quantify the effort or Cardiovascular Load generated throughout training, taking into account the sum of all records generated for a given heart rate, heart rate zone, terrain characteristics (topography), speed, and distance traveled by the horse.

[0169] The following paragraphs contain the sequential details for the point-by-point quantification of a Point Cardiovascular Load value, which will be summed at the end of the horse's training record, resulting in the Cardiovascular Load. Properly speaking, this sequence involves manipulating training data in order to generate products based on the originally acquired data.

[0170] For the Altitude and Speed ​​data, it is necessary to create two data groups, using the values ​​of their moving averages with a value window of 5. The two new groups are called Altitude_med_mov and Speed_med_mov.

[0171] Two new columns are inserted into the database: the first new column is called 'Mag_Alt' (Altitude Magnitude). This column is populated with the result of dividing the 'Altitude_med_mov' column by the value of the same column, shifted one row upwards. This is done to calculate the "Magnitude" of the altitude relative to the previous row.

[0172] The second new column is called 'Mag_Vel' (Magnitude of Velocity). It is filled with the result of dividing the 'Velocidade_med_mov' column by the value in the same column, shifted one line up. This is done to calculate the "Magnitude" of the velocity relative to the previous line.

[0173] In these two new columns, the value 1 is assigned to cases where the previous row is equal to 0 (zero). This is done in the newly created columns 'Mag_Alt' and 'Mag_Vel', where the values ​​are null, that is, where you don't have a number, usually indicating that division by zero occurred. These nulls are replaced by 1, in order to avoid divisions by zero.

[0174] By dividing Mag_Vel by Mag_Alt, the Movement Magnitude data column, called “Mag_Mov”, is created. Whenever the result of Mag_Mov is zero, this value is replaced by 1 to prevent future divisions by zero.

[0175] A new column called 'Mag_FC' or Heart Rate Magnitude is created in the database. This column is populated with values ​​referring to a score assigned to the Heart Rate Zone type in which the Heart Rate value is located, as per Table 26. Table 26 – Heart Rate Magnitude Score, assigned to the Heart Rate value according to the respective Heart Rate Zone.

[0176] The Point Cardiovascular Load value for each data point is calculated as the product of Heart Rate, Movement Magnitude, and Heart Rate Magnitude (Eq. 09). It is necessary to ensure that the values ​​in the 'Mag_mov' and 'Mag_FC' columns are non-integer values. Eq. 09 – Point Cardiovascular Load Equation.

[0177] Finally, the Cardiovascular Load is calculated, which can be expressed as the integer part of the sum of the Point Cardiovascular Loads, plus the difference between the maximum and minimum altitude, divided by the total training distance (Eq.10). Eq.10 – Cardiovascular Training Load Equation.

[0178] During training, a horse's heart rate varies primarily as a function of the level of exertion and the animal's cardiorespiratory fitness, in addition to... Cardiac conditions prior to training. An animal that is rested or has recovered from previous exertion will have different heart rate variability characteristics compared to an animal that is still recovering from a past workout.

[0179] Each heart rate zone has a benefit, provided the time requirements for maintaining the heart rate in those zones are met. The time requirements for maintaining a given heart rate zone and its benefits can be summarized in Table 27 below: Neuromuscular system of the equine observed. Table 27 – Requirements (working time in a given FC Zone) and their respective benefits generated for a healthy horse.

[0180] Graphically representing a horse's training based on the Heart Rate Zones worked, and the proportion of work in each zone, is a very useful tool for guiding training, whether for conditioning gains or for the horse's physical recovery. Figure 18 illustrates a graph of HR zones and a proportion of zones worked during a training session.

[0181] A horse's fitness level can be estimated in several ways, some considered simpler and cheaper but less precise, while others are more complex, invasive, and costly but yield more quantitative and reliable results than the simpler methods. Below are the three main methods used to measure fitness in horses: 1 - Cardiac Stress Tests: These tests measure the horse's heart rate in response to specific exercises. Tests can be performed on a treadmill or in the field, assessing how the heart rate increases with exercise and how it returns to normal after exertion. 2 - Blood Lactate Analysis: This measures the concentration of lactate in the horse's blood after exercise. Elevated lactate values ​​indicate that the horse has reached its anaerobic threshold, which helps determine the intensity of exercise the horse can withstand before becoming fatigued. 3 - Maximum Aerobic Capacity (VO2 max) Tests: These tests assess the horse's maximum oxygen consumption during intense exercise. This is usually done on a treadmill equipped to measure respiratory gas exchange.

[0182] Of the three methods for estimating physical fitness presented above, cardiac stress tests are the simplest, cheapest, and most non-invasive way to obtain a baseline that can be used as a fitness level. However, most physical stress tests are considered complex for continuous replication, as they involve specific and standardized training environments where test conditions require systematic control.

[0183] PERSSON (1983) defined that a measurement of exercise speed at which the heart rate reaches 200 bpm (V200) provides information that can be useful for estimating whether the capacity aerobic work in horses, with its increase during training suggesting an increase in aerobic capacity.

[0184] Treadmill exercise is considered a safe procedure, and has not been shown to increase the risk of musculoskeletal injuries in horses (FRANKLIN et al., 2010). However, PERSSON (1983) and KING et al. (1995) reported greater safety in these tests using submaximal to maximal tests, due to the difficulty of acclimation to maximal tests. Submaximal parameters such as V150 and V180 are considered inadequate for evaluating animal performance, given the variability of heart rate (HR) during low-intensity exercise. Therefore, there is a tendency to use submaximal to maximal parameters, such as V200 and HRmax, which increase the accuracy of the test (ROSE & CHRISTLEY, 1995).

[0185] V200 is obtained through linear extrapolation of pairs of Cartesian coordinates (x, y), where x is the speed (km / h) and y is the heart rate (bpm), obtained in a training session with a progressive and continuous increase in speed, as shown in Figures 19 and 20. These figures show, respectively, two Cartesian pairs containing the speed and heart rate measurements in a training session with a progressive increase in speed. The extrapolation represents the line generated between the two points of the V200 example (adapted from EVANS, 2000), and the evolution of conditioning from the horizontal displacement at the V200 station. This displacement reflects that the horse needs to reach higher speeds to achieve 200 bpm (adapted from EVANS, 2000).

[0186] A test to obtain data that will be used in an extrapolation to V200 needs to follow the series of protocols below: - Warm-up: Begins with a warm-up period to prepare the horse, usually consisting of trotting and light galloping. - Progressive Test: After the warm-up, the horse undergoes a progressive test on a track or treadmill. The speed is gradually increased at regular intervals until the horse's heart rate reaches 200 bpm. - Measurement: A heart rate monitor, usually a chest strap connected to a watch or other recording device, is used to continuously measure the horse's heart rate during the test. - Determination of V200: The speed corresponding to the moment or the extrapolation of the values ​​obtained up to a heart rate of 200 bpm is recorded as V200.

[0187] Due to the fact that it is not always possible to follow these protocols in a workout, this work proposes the innovative use of V200, obtained from the extrapolation of maximum speed and heart rate data from 100 segments of any workout, always using a speed of 0 (zero km / h) and a heart rate of 20 bpm as the origin coordinate, to draw the line to be extrapolated to the heart rate of 200 bpm in each segment. Figure 21 illustrates a workout with heart rate and speed information. Figure 22 illustrates the same workout, but divided into segments, where each segment... to be the source of maximum speed and heart rate information for creating the extrapolation lines of the V200.

[0188] For each segment, identifying the maximum heart rate and maximum speed is crucial. Based on the maximum heart rate and speed of each segment, the speed the horse would have (or has) if its heart rate were 200 beats per minute is calculated.

[0189] Figure 23 shows a graph illustrating the relationship between speed and heart rate for each segment, extrapolated up to a heart rate of 200 bpm. For the calculated V200 values, only data within the range of 20 to 50 km / h are used. The average of these values ​​is calculated to obtain a summary measure of the horse's aerobic conditioning. This value is defined as the Horse's Conditioning Level.

[0190] Data on a horse's conditioning level should not be analyzed individually, i.e., after each training session. Ideally, this data should be analyzed cumulatively, based on the median of the last 15 days of data. This reduces the dispersion of recorded conditioning level values, generating data that forms a trend curve useful for monitoring a horse's physical conditioning results. Figure 24 illustrates a graph of a horse's conditioning level over a year, for each training session, with a high level of dispersion. The red curve shows the trend of gain or loss of physical conditioning, based on the median calculated every 15 days of records.

[0191] This work also involves understanding and respecting a horse's training limits, as part of the policy. anti-stress and promotes healthy training. A horse that is not used to heavy training should not be subjected to something like this, as the likelihood of this type of exposure causing injury or serious physical and mental discomfort to the animal is high.

[0192] This type of awareness contradicts what is proposed as the Tolerance Level, which is defined here as the moving average value of the Cardiovascular Load over the last 28 days of a horse's training records.

[0193] The Tolerance Level (or simply Tolerance) is a way to keep a horse's training within the animal's own effort reference, generating conditioning goals based on the effort limits already achieved by the horse in its past training sessions.

[0194] In this study, the definition of the Effort Level (or simply Effort) is the moving average value of the Cardiovascular Load over the last 7 days of training records for a horse.

[0195] With the Effort and Tolerance data, it is possible to establish some relationships that can serve as guides for responsible training, where the trainer can program their activities in a way that guarantees continuous and responsible conditioning gains, since there is knowledge about the limits that should be considered to achieve a goal in a way that respects animal welfare, with lower risks of injuries, physical and mental stress, and other harm.

[0196] The ratio between Effort and Tolerance is calculated to understand different training and recovery states, each associated with different impacts on the horse's health and athletic ability. These conditions are divided into five categories, each one... representing a distinct level of training intensity and its suitability, as shown in Table 28 below: Table 28 – Classification categories of the ratio between Effort and Tolerance and the description of the result condition of a horse's training status.

[0197] Figure 25 presents two Effort vs. Tolerance graphs for two horses, representing training intervals over a year.

[0198] Figure 26 shows a record of a real training session considered "light," where the horse's heart rate worked exclusively in Zone 1, which is the Recovery Zone. In this training session, there was little variation in topography and low speeds of movement for the horse, resulting in manageable heart rates.

[0199] Training under these conditions benefits the horse in terms of maintaining physical conditioning and cardiovascular recovery, especially compared to previous, more strenuous training sessions.

[0200] Figure 27 shows a record of a real training session considered "light," where the horse's heart rate was in zones 1 and 2, which are the Recovery and Moderate Zones, respectively. In this training session, there was little variation in topography and low movement speeds for the horse, similar to the previous example.

[0201] However, the type of movement, duration, and type of training generated higher heart rates than in the previous example, increasing the final cardiovascular load. Training under these conditions benefits the horse in terms of maintaining physical conditioning and cardiovascular recovery, as well as improving metabolism and strengthening the horse's musculoskeletal system.

[0202] Figure 28 shows a record of a real training session considered "moderate," where the horse's heart rate worked in zones 1, 2, 3, and 4, which are the Recovery, Moderate, Aerobic, and Borderline zones, respectively.

[0203] In this training session, despite little variation in topography, the variations in the horse's movement speed were more pronounced than in the previous training sessions exemplified, generating greater effort in the horse. Conducting training under these conditions brings benefits in terms of increased endurance in the horse relative to its aerobic capacity, increasing the animal's physical resistance, in addition to strengthening the musculoskeletal system.

[0204] Figure 29 shows a record of a real training session considered "intense." In this training session, all heart rate zones were reached due to the high level of effort the horse was subjected to during the training.

[0205] Although this type of training provides extra benefits for the horse's conditioning, such as improving the horse's response to anaerobic activity (bursts of energy), it should not be practiced frequently or without the animal's physical fitness being properly prepared, as all this effort demands a lot from the horse's body.

[0206] The methods for analyzing the health and physical and mental conditions of a horse proposed in this work need to be carried out jointly, so that interpretations take into account the responses of each technique presented over time. An analysis of a real condition of a horse that demonstrated a situation of non-compliance with health, the moment of veterinary medical intervention and its results can be seen in a simplified way in Figure 30.

[0207] The example above shows a horse that began exhibiting results interpreted as typical of discomfort or non-compliance with health conditions. This occurred even with rest periods between training sessions and without any physical signs being found that could be interpreted as such. Typical of injury, responses considered negative persisted for more than 2 consecutive months (responses highlighted in red).

[0208] At the request of a veterinarian, a blood test was performed, which detected that the horse had babesiosis, a disease caused by a microscopic parasite transmitted by ticks, infecting the red blood cells of its hosts. It is similar to malaria and can lead to symptoms such as fever, fatigue, and anemia. Treatment for babesiosis was initiated at that time, altering the test results to conditions interpreted as more favorable and ideal for a healthy horse.

[0209] Figure 31 shows graphs of stress level, effort vs. tolerance, and conditioning level of the horse from the previous example, as well as a representation of a data range over a period of one year.

[0210] The red zone refers to the period where consecutive poor health assessment responses were observed, with stress levels higher than average.

[0211] The blue band refers to the period when the horse is undergoing treatment for babesiosis, showing improvement in stress responses.

[0212] The level of physical fitness continued to improve, due to the training sessions being productive and focused on maintenance, with a short period of overload, thus reducing the possibility of a drop in fitness.

[0213] Note that the assessment techniques for Stress Level, Health, Physical Conditioning, and Orientation, which are the subject of this study, are extremely valuable for the daily evaluation and training of horses, whether athletes or not. This is because it is a non-invasive and low-cost system. Cost-effective, and by leveraging equipment already available on the market, this new technology represents a significant advantage for trainers and veterinarians.

[0214] With the system available, it is possible to make informed decisions and provide accurate guidance to improve the conditioning of horses, always focusing on preserving animal welfare.

Claims

1 / 13 CLAIMS:

1. EQUINE MONITORING METHOD comprising a Bluetooth sensor (1), adjustable elastic band (2) that wraps around the animal's thorax (3) at heart level and electronic module (4) connected to the elastic band (2), the module (4) being equipped with a heart rate sensor that transmits data, preferably via Bluetooth to mobile devices (5), characterized by being composed of resting test and stress test stages and creating the RR Interval histogram, through the calculation of the square root of the number of elements present in the RR Intervals, using the RR Interval data, from which the mean and standard deviation are calculated, generating a set of values ​​for a variable X, which varies between 0 and 5000, generating a PDF function such that the highest value obtained satisfies the condition that the value of the PDF function at that point is greater than or equal to the constant 0.00000000098,This value is defined as "Value X" and its integer part represents the width of the histogram base, resulting in a signal of the effort level for isolated acquisitions or the recovery level for joint acquisitions.

2. EQUINE MONITORING METHOD, according to claim 1, characterized by the classification of the histogram base width being performed in such a way that if the base width is less than 1527.5ms, it is classified as "Very Short"; if it is between 1527.5 and 2902.25ms, it is classified as "Short"; if it is between 2902.25 and 3207.75ms, it is classified as "Compatible"; if it is between 3207.75 and 4582.5ms, it is classified as "Wide"; and if it is greater than 4582.5ms, it is classified as, 2 / 13 "Very Wide"; the base width classification being the variable called "base width classification".

3. EQUINE MONITORING METHOD, according to claim 1, characterized by the variable "heterogeneity" being classified into three predefined categories, such that if the heterogeneity is less than 0.3, it is classified as "Low"; if it is between 0.3 and 0.5, it is classified as "Medium"; and if it is greater than or equal to 0.5, it is classified as "High", the heterogeneity classification being the variable called "heterogeneity classification".

4. EQUINE MONITORING METHOD, according to claim 3, characterized by scores being assigned to the variable "heterogeneity", which is classified as "low", with a score equal to 0; "medium", with a score equal to 1; and "high", with a score equal to 2, with the "heterogeneity" score being the variable called "heterogeneity score". 5.A method for monitoring equines, according to claim 1, characterized in that the variable "normality of the histogram" is classified as "Not Compatible" if the normality is less than 70; "Slightly Compatible" if it is between 70 and 90; and "Compatible" if it is greater than or equal to 90.

6. A method for monitoring equines, according to claim 5, characterized in that scores are assigned to the variable "normality of the histogram," which is classified as 1 if it is "slightly compatible"; 2 if it is "compatible"; and 0 (zero) if it is "not compatible." 3 / 13 7. EQUINE MONITORING METHOD, according to claim 1, characterized by the sum of the scores for base width, heterogeneity, and normality being used to classify the animal's stress level, with a sum below 3 classified as "very bad"; between 3 and 5 "bad"; between 5 and 6 "average"; between 6 and 8 "good"; and above 8 "very good".

8. EQUINE MONITORING METHOD, according to claim 7, characterized by the condition "very bad" being classified as a "VERY HIGH" stress level; "bad" as a "HIGH" stress level; "average" as a "NORMAL" stress level; "good" as a "LOW" stress level; and "very good" as a "VERY HIGH" stress level. 9.A method for monitoring horses, according to claim 1, characterized by the analysis of RR interval data in the time domain performed to evaluate heart rate variability (HRV) by measuring the RR interval and analyzing its variation over time. A method for monitoring horses, according to claim 9, characterized by the RR interval data being divided into First and Second Parts, with the data volume divided into multiples of 256 points, with the possibility of overlapping data from the end of the First Part and the beginning of the Second Part, always having a total of analyzable data in multiples of 256 and a maximum limit of 4096 data points for each part, with a total limit of 8192 data points to be analyzed at a time in the RR intervals. 4 / 13 11. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Average RR Interval (RRMED) is the average of the RR intervals measured over a given period.

12. EQUINE MONITORING METHOD, according to claim 11, characterized in that the RRMED of the training is classified on a scale of 1 to 5, such that if it is below 444.55, it will be classified as 1; if it is between 444.55 and 844.55, it will be classified as 2; if it is between 844.55 and 933.45, it will be classified as 3; if it is between 933.45 and 1333.5, it will be classified as 4; and if it is equal to or greater than 1333.5, it will be classified as 5. 13.A method for monitoring equines, according to claim 11, characterized in that the RRMED of the test is classified on a scale of 1 to 5, such that if it is below 690.5, it will be classified as 1; if it is between 690.5 and 1311.95, it will be classified as 2; if it is between 1311.95 and 1450.05, it will be classified as 3; if it is between 1450.05 and 2071.5, it will be classified as 4; and if it is equal to or greater than 2071.5, it will be classified as 5. A method for monitoring equines, according to claim 1, characterized in that the Standard Deviation of RR Intervals (SDNN) is the measure of the standard deviation of RR intervals over a given period.

15. A METHOD FOR MONITORING EQUINES, according to claim 14, characterized by the SDNN of the training being classified on a scale of 1 to 5, such that if it is below 255, it will be classified as 1; if it is between 255 and 5, it will be classified as 255. 5 / 13 484.5, will be classified as 2; if it is between 484.5 and 535.5, it will be classified as 3; if it is between 535.5 and 765, it will be classified as 4; and if it is equal to or greater than 765, it will be classified as 5.

16. EQUINE MONITORING METHOD, according to claim 14, characterized by the SDNN of the test being classified on a scale of 1 to 5, such that if it is below 572.7, it will be classified as 1; if it is between 572.7 and 1088.13, it will be classified as 2; if it is between 1088.13 and 1202.67, it will be classified as 3; if it is between 1202.67 and 1718.1, it will be classified as 4; and if it is equal to or greater than 1718.1, it will be classified as 5.

17. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Standard Deviation of Moving Average RR Intervals (SDANN) is the measure of the standard deviation of the moving averages of the RR intervals over a given period. 18.A method for monitoring equines, according to claim 17, characterized by the training SDANN being classified on a scale of 1 to 5, such that if it is below 150, it will be classified as 1; if it is between 150 and 285, it will be classified as 2; if it is between 285 and 315, it will be classified as 3; if it is between 315 and 450, it will be classified as 4; and if it is equal to or greater than 450, it will be classified as 5. A method for monitoring equines, according to claim 14, characterized by the test SDANN being classified on a scale of 1 to 5, such that if it is below 621.5, it will be classified as 1; if it is between... 6 / 13 621.5 and 1180.85, it will be classified as 2; if it is between 1180.85 and 1305.15, it will be classified as 3; if it is between 1305.15 and 1864.5, it will be classified as 4; and if it is equal to or greater than 1864.5, it will be classified as 5.

20. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Standard Deviation of RR Difference Intervals (SDNNi) is the measure of the standard deviation of the differences between adjacent RR intervals over a given period.

21. EQUINE MONITORING METHOD, according to claim 20, characterized in that the SDNNi of training and testing are classified on a scale of 1 to 5, such that if it is below 229, it will be classified as 1; If it is between 229 and 435.1, it will be classified as 2; if it is between 435.1 and 480.9, it will be classified as 3; if it is between 480.9 and 687, it will be classified as 4; and if it is equal to or greater than 687, it will be classified as 5. 22.A method for monitoring equines, according to claim 1, characterized in that the Root Mean Square of the Differences of RR Intervals (RMSSD) is the measure of the square root of the mean square of the differences between adjacent RR intervals over a given period.

23. A method for monitoring equines, according to claim 22, characterized in that the RMSSD of the training is classified on a scale of 1 to 5, such that if it is below 290, it is classified as 1; if it is between 290 and 551, it is classified as 2; if it is between 551 and 609, it is... 7 / 13 classified as 3; if it is between 609 and 870, it will be classified as 4; and if it is equal to or greater than 870, it will be classified as 5.

24. EQUINE MONITORING METHOD, according to claim 22, characterized by the test RMSSD being classified on a scale of 1 to 5, such that if it is below 767.05, it will be classified as 1; if it is between 767.05 and 1457.395, it will be classified as 2; if it is between 1457.395 and 1610.805, it will be classified as 3; if it is between 1610.805 and 2301.15, it will be classified as 4; and if it is equal to or greater than 2301.15, it will be classified as 5.

25. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Percentage of Differences of RR Intervals Greater than 50ms (pNN50) is the measure of the percentage of differences between adjacent RR intervals that are greater than 50 ms over a given period. 26.A METHOD FOR MONITORING EQUINES, according to claim 25, characterized by the training pNN50 being multiplied by 1500 to be classified on a scale of 1 to 5, such that if it is below 314.5, it will be classified as 1; if it is between 314.5 and 597.55, it will be classified as 2; if it is between 597.55 and 660.45, it will be classified as 3; if it is between 660.45 and 943.5, it will be classified as 4; and if it is equal to or greater than 943.5, it will be classified as 5.

27. EQUINE MONITORING METHOD, according to claim 25, characterized by the training pNN50 being multiplied by 1500 to be classified on a scale of 1 to 5, so that if it is below 589.5, it will be classified as 1; if it is between 589.5 and 1120.05, it will be classified as . 8 / 13 2; if it is between 1120.05 and 1237.95, it will be classified as 3; if it is between 1237.95 and 1768.5, it will be classified as 4; and if it is equal to or greater than 1768.5, it will be classified as 5.

28. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Percentage of Differences of RR Intervals Greater than 50ms is the measure of the percentage of differences between adjacent RR intervals that are greater than 50 ms over a given period.

29. EQUINE MONITORING METHOD, according to claim 1, characterized in that the Sum of Time Indicators is the sum of all indicators in the Time Domain. 30.A method for monitoring horses, according to claim 29, characterized by the sum of training time indicators being classified on a scale of 1 to 5, such that if it is less than 1550, it will be classified as 1; if it is between 1550 and 2945, it will be classified as 2; if it is between 2945 and 3255, it will be classified as 3; if it is between 3255 and 4650, it will be classified as 4; and if it is equal to or greater than 4650, it will be classified as 5. A method for monitoring horses, according to claim 29, characterized by the sum of test time indicators being classified on a scale of 1 to 5, such that if it is less than 3760.2, it will be classified as 1; If it is between 3760.2 and 7144.38, it will be classified as 2; if it is between 7144.38 and 7896.42, it will be classified as 3; if it is between 7896.42 and 11280.6, it will be classified as 4; and if it is equal to or greater than 11280.6, it will be classified as 5. 9 / 13 32. EQUINE MONITORING METHOD, according to claim 1, characterized by varying the Time Domain Indicator Score variable from 0 (zero) to 10, applied to both training and test data, wherein when applied to test data a comparative reference value is generated in relation to the training values.

33. EQUINE MONITORING METHOD, according to claim 1, characterized by performing the Fast Fourier Transform (FFT) calculation applied to a wave signal in order to decompose the signal into its frequency components, allowing spectral analysis of the signal.

34. A METHOD FOR MONITORING EQUINES, according to claim 33, characterized by generating Frequency x Magnitude (ms² / Hz) graphs, where the x-axis represents the frequency in Hertz (Hz) and the y-axis represents the magnitude in ms² / Hz. 35.A method for monitoring equines, according to claim 34, characterized in that each generated graph has three distinct areas, representing the three frequency zones of the RR interval data, the first zone being the Very Low Frequency zone, ranging from 0 to 0.04 Hz, the second zone being the Low Frequency zone, ranging from 0.04 to 0.15 Hz, and the third zone being the High Frequency zone, ranging from 0.15 to 0.4 Hz.

36. A method for monitoring equines, according to claim 34, characterized in that three variables are defined: VLF, LF, and HF, representing the amount of energy of the electrocardiographic signal in each of the specified frequency zones; the amount of energy being measured as a. 10 / 13 sum of the magnitudes of each RR interval (in ms² / Hz); wherein for each of the three variables, a 100% normalization of the frequency component is performed in relation to the total energy, creating normalized variables VLFn, LFn and HFn, which indicate the percentage of each component in relation to the total sum of the three.

37. EQUINE MONITORING METHOD, according to claim 1, characterized by the interpretation of Heart Rate Variability (HRV) performed through the normalized variables VLFn, LFn and HFn.

38. EQUINE MONITORING METHOD, according to claim 37, characterized by the reference values ​​of the normalized variables VLFn, LFn and HFn being 44, 23 and 35, respectively. 39.A method for monitoring equines, according to claim 36, characterized in that the classification of |44 - HFn|, when less than or equal to 2, is given by (Class_HFn) “Normal”; of |44 - HFn| greater than 2 and less than or equal to 4, is given by (Class_HFn) “Less”; of |HFn - 44| greater than 2 and less than or equal to 4, is given by (Class_HFn) “Greater”; of |44 - HFn| greater than 4, is given by (Class_HFn) “Much Less”; of |HFn - 44| greater than 4, is given by (Class_HFn) “Much Greater”.

40. A method for monitoring equines, according to claim 36, characterized in that the classification of |23 - LFn|, when less than or equal to 2, is given by (Class_LFn) “Normal”; of |23 - LFn| greater than 2 and less than or equal to 4, given by (Class_LFn) “Less than”; of | LFn - 23| greater than 2 and less than or equal to 4, given by (Class_LFn) “Greater than”; of |23 - LFn| greater than 4, given. 11 / 13 by (Class_LFn) “Much Smaller”; of |LFn - 23| greater than 4, given by (Class_LFn) “Much Larger”.

41. EQUINE MONITORING METHOD, according to claim 36, characterized by the classification of |35 - VLFn|, when less than or equal to 2, given by (Class_VLFn) “Normal”; of |35 - LFn| greater than 2 and less than or equal to 4, given by (Class_VLFn) “Smaller”; of |VLFn - 35| greater than 2 and less than or equal to 4, given by (Class_VLFn) “Larger”; of |35 - VLFn| greater than 4, given by (Class_VLFn) “Much Smaller”; of |VLFn - 35| greater than 4, given by (Class_VLFn) “Much Larger”.

42. A METHOD FOR MONITORING EQUINES, according to claim 36, characterized in that the HFn, LFn, and VLFn classifications are scored and stored in variables Pont_HFn, Pont_LFn, Pont_VLFn, where “Much Smaller” equals 1, “Smaller” equals 2, “Normal” equals 3, “Larger” equals 4, and “Much Larger” equals 5. 43.A METHOD FOR MONITORING EQUINES, according to claim 1, characterized by the variables 'Data', 'Horse', 'Training Type', 'Base_Width', 'RRMED_1', 'RRMED_2', 'SDNN_1', 'SDNN_2', 'SDANN_1', 'SDANN_2', 'SDNNi_1', 'SDNNi_2', 'RMSSD_1', 'RMSSD_2', 'pNN50_1', 'pNN50_2', 'pNN50_1', 'pNN50_2', 'Sum_Index_Time_1', 'Sum_Index_Time_2', 'VLFn_1', 'VLFn_2', 'LFn_1', 'LFn_2', 'HFn_1', 'HFn_2', 'Homeostasis_1', 'Homeostasis_2', 'Stress_Level_1', 'Stress_Level_2' are stored in a single repository.

44. EQUINE MONITORING METHOD, according to claim 1, characterized by FCmáx being the frequency. 12 / 13 maximum heart rate of the animal and Z1 being the Recovery Zone, comprising heart rates below 60% of HRmax; Z2 the Moderate Zone comprising heart rates between 60% and 70% of HRmax; Z3 the Aerobic Zone comprising heart rates between 70% and 80% of HRmax; Z4 the Borderline Zone comprising heart rates between 80% and 90% of HRmax; and Z5 the Maximum Zone comprising heart rates above 90% of HRmax.

45. EQUINE MONITORING METHOD, according to claim 1, characterized by the resting test phase being carried out between 6:00 am and 9:00 am, every three months, for at least 20 min, with the animal (3) in a state of maximum relaxation.

46. ​​EQUINE MONITORING METHOD, according to claims 1 and 2, characterized by the resting test stage being responsible for promoting the parameterization of the animal (3) in a healthy and calm condition, without external interference, signs of injury or disease. 47.A method for monitoring equines, according to claim 1, characterized by the stress test stage being performed with recordings of at least 15 minutes in duration.

48. A method for monitoring equines, according to claim 1, characterized by the GPS data being grouped in the form of a database containing Time (date and time of each recording per second); Latitude; Longitude; Altitude (m); Heart Rate (bpm); Speed ​​(km / h); and Distance (km). 13 / 13 49. EQUINE MONITORING METHOD, according to claim 1, characterized by the measurement of the animals' physical conditioning being carried out through the methods of Cardiac Stress Tests, Blood Lactate Analysis or Maximum Aerobic Capacity Tests (VO2 max).

50. EQUINE MONITORING METHOD, according to claim 1, characterized by the ratio between Effort and Tolerance being divided into the categories of Detraining, when less than or equal to 0.7, Maintenance, when greater than 0.7 and less than or equal to 1, Productive, when greater than 1 and less than or equal to 1.2, Overload, when greater than 1.2 and less than or equal to 1.4, and Risk of Injury, when greater than 1.4.

Images