Equine tail dock mounted biometric monitoring system and method

Abstract

A biometric monitoring system for an equine includes a sensor module secured to a tail dock such that a sensor contacts an underside of the tail dock. The module includes a photoplethysmography sensor for heart rate and a tail movement sensor for detecting motion and orientation. The system obtains biometric and movement data, determines baseline data for the equine, and correlates current readings with baseline data to identify deviations. Based on the deviations, the system determines physiological or behavioural parameters. The arrangement enables non-invasive, continuous monitoring using combined optical and motion sensing at a stable anatomical location.

Description

Equine Tail Dock Mounted Biometric Monitoring System and Method Field of the Invention

[0001] This disclosure relates generally to biometric monitoring systems. More particularly, the disclosure relates to a system and method for monitoring physiological and movement-related parameters in equines using a sensor module mounted at a tail dock.Background of the Invention

[0002] Field monitoring of equine health and behaviour is of significant importance across veterinary care, performance management, breeding, and general animal husbandry. Early identification of changes in physiological and behavioural parameters may assist caretakers in responding to developing conditions in a timely manner. Traditionally, such monitoring has relied on periodic manual observation, including visual inspection of behaviour, palpation, or intermittent measurement of vital signs. These approaches may be labour-intensive and may not capture transient or gradual changes that occur between observations.

[0003] In recent years, various wearable and sensor-based monitoring systems have been proposed for animals, including equines, to enable more continuous data acquisition. Such systems may utilise sensors to measure parameters such as heart rate, temperature, motion, or location, and may transmit data to remote devices for analysis. However, many existing systems are limited by sensor placement, signal reliability, or the ability to meaningfully correlate multiple physiological and behavioural indicators.

[0004] For example, WO 2017 / 164807 (Andersson) published on 28 September 2017 discloses an animal vital signs monitoring system comprising a sensing band configured to be positioned around the underside of the tail dock and a separate housing attached to tail hair, wherein physiological parameters may be detected. While such arrangements provide a mechanism for acquiring multiple signals, the use of a separate sensing band and housing may introduce complexity in mounting and may be susceptible to movement artefacts or inconsistent sensor contact.

[0005] US 4,630,613 (Dennis) published on 23 December 1986 discloses a pulse detector for animals in which a sensing device is positioned at the root of the tail and secured by bandaging, with signals transmitted remotely. The detector comprises a closed hollow structure having a flexible surface placed against the tail, the structure being substantially filled with fluid such that pulsations in the tail cause movement of the fluid, which is then detected to generate a pulse signal. Such systems may enable pulse detection but rely on indirect fluid displacement sensing and may be limited in terms of multi-parameter monitoring and integration with additional behavioural data.

[0006] US 2015 / 0257664 (Esposito) published on 17 September 2015 discloses an equine heart rate monitoring device configured to be positioned beneath the tail near the dock, wherein the device includes a sensing assembly comprising a pressure transducer disposed within a sealed cavity formed by a flexible bulb structure. The bulb extends over a depressed portion of a printed circuit board to define the sealed cavity, such that pulsations in a caudal vein or artery cause deformation of the bulb and corresponding pressure changes within the cavity that are detected by the pressure transducer. Such arrangements rely on pressure-based sensing within a deformable cavity and are generally directed to single-parameter heart rate detection, without integration of multi-parameter sensing or correlated analysis of movement and physiological data.

[0007] US 6,436,038 (Engstrom) published on 20 August 2002 discloses an animal vital signs monitoring system comprising a probe including a rectal thermometer for measuring core temperature and a cuff adapted to be wrapped around an extremity of the animal, typically the tail, for measuring pulse and blood pressure. The cuff operates by applying pressure to the tail to occlude blood flow and detecting pressure variations associated with arterial pulsations during controlled inflation and deflation. Such arrangements may provide clinically accurate measurements but rely on invasive or restrictive components, including rectal probes and occlusive cuffs, which may limit suitability for continuous or long-term monitoring in free-moving animals.

[0008] WO 2013 / 186232 (Austin) published on 19 December 2013 discloses a birthing sensor configured to be mounted on the tail of a pregnant animal, the sensorcomprising a sealed casing containing a movement sensor in the form of a multi-axis accelerometer, a microcontroller, memory, and a wireless communication module. The accelerometer is arranged to capture tail movement data across multiple axes, and the system processes the movement data by averaging, filtering, and calculating a magnitude of motion over time, which is compared against predefined threshold patterns representative of impending birth. When the processed movement data exceeds the threshold, an alert is transmitted to a remote user. Such systems are directed to detecting a specific labour-related event based on thresholded movement magnitude and do not provide broader physiological monitoring or multi-parameter correlation of biometric and behavioural data.

[0009] US 2010 / 0036277 (Austin) published on 11 February 2010 discloses an animal temperature monitoring device configured to be mounted on the tail of an animal, wherein the device includes an internal temperature sensor arranged to sense a temperature within a gap formed between the buttocks when the tail is in a normal position. The device may further include control and wireless transmission components, and may utilise additional sensing to determine whether the tail is positioned appropriately for obtaining a valid temperature reading. Such arrangements are primarily directed to temperature measurement within a specific anatomical gap and validation of that measurement, and do not incorporate integrated analysis of multiple physiological and behavioural parameters.

[0010] WO 2015 / 107521 (Menkes et al.) published on 23 July 2015 discloses a pet monitoring device in the form of a collar configured to measure parameters including temperature, pulse rate, respiration rate and movement patterns. The device comprises multiple sensors distributed around the collar and a processing unit configured to analyse sensor data and transmit information to a remote system. While such systems provide multi-parameter monitoring, the collar-based configuration is adapted for companion animals and relies on neck-mounted sensing, which may be less suitable for equine applications and may be subject to displacement, interference, or reduced measurement reliability in certain use environments.

[0011] Accordingly, there remains a need for an improved monitoring arrangement capable of obtaining reliable physiological and behavioural data from an equine in a manner that supports continuous monitoring and enables more robust interpretation of combined sensor inputs.

[0012] It is to be understood that, if any prior art information is referred to herein, such reference does not constitute an admission that the information forms part of the common general knowledge in the art, in Australia or any other country.Summary of the Disclosure

[0013] A biometric monitoring system for an equine is provided, the system comprising a biometric sensor module including a housing and a harness, wherein the housing defines an inner contact face configured for placement against an underside of a tail dock of the equine such that at least one biometric sensor is operative through the inner contact face in direct contact with the tail dock. The harness is configured to secure the housing to the tail dock to retain the inner contact face in aligned contact with the underside of the tail dock during movement of the equine. The biometric sensor comprises a photoplethysmography sensor configured to obtain heart rate data, and the biometric sensor module further comprises a tail movement sensor configured to detect movement and / or orientation of the tail dock. The system is configured to obtain biometric sensor readings and tail movement data, determine baseline biometric sensor data and baseline tail movement data for the equine based on previously acquired data, and correlate the biometric sensor readings with the tail movement data by comparing the readings against the respective baseline data to identify a deviation and determine at least one physiological or behavioural parameter of the equine.

[0014] The use of a housing defining an inner contact face for direct placement against the underside of the tail dock provides a stable sensing interface at a location that is relatively free of hair and less susceptible to external disturbance, thereby improving signal quality and consistency of measurement. The integration of the biometric sensor within the housing, rather than relying on separate sensingelements, may reduce mechanical complexity and improve robustness during equine movement.

[0015] The securement of the housing to the tail dock using the harness to maintain aligned contact may reduce motion artefacts and improve reliability of both optical and motion-based sensing. Maintaining consistent alignment of the inner contact face with the tail dock may enable more accurate acquisition of physiological signals over extended periods, including during locomotion or environmental interaction.

[0016] The use of a photoplethysmography sensor at the tail dock location enables non-invasive acquisition of heart rate data without requiring intrusive probes or electrode-based arrangements. The tail movement sensor provides additional data relating to both dynamic motion and orientation of the tail dock, enabling capture of behavioural information that is not available from physiological sensing alone.

[0017] The correlation of biometric sensor readings with tail movement data, in combination with baseline modelling for the individual equine, enables multiparameter analysis that may improve robustness of detection compared to singleparameter systems. By comparing current readings against baseline data, the system may account for individual variability between equines and adapt to changes over time, thereby reducing false detections and improving sensitivity to meaningful deviations.

[0018] The combination of these features provides a monitoring system capable of continuous operation in a range of environments, including paddock, stable, transport, and veterinary contexts, while maintaining a non-invasive and unobtrusive form factor suitable for prolonged use.

[0019] In an embodiment, the system may be configured to determine a respiratory rate by identifying periodic modulation in a photoplethysmography signal and correlating the modulation with cyclic tail movement detected by the tail movement sensor. Preferably, the processor may synchronise peaks in the photoplethysmography signal with oscillatory tail movement to identify respiration cycles. By utilising both optical signal modulation and mechanically detected tail motion, the system may provide a more robust estimation of respiratory rate,particularly in conditions where either signal alone may be affected by noise or motion artefacts.

[0020] In an embodiment, the system may be configured to establish a baseline tail movement pattern for the equine based on historical tail movement data. The baseline may represent characteristic values of movement frequency, amplitude, and orientation for the individual equine. The system may further be configured to detect a deviation from the baseline tail movement pattern by comparing current movement data against the baseline. By referencing individual baseline patterns rather than fixed thresholds, the system may accommodate inter-animal variability and improve sensitivity to meaningful behavioural changes.

[0021] Preferably, the system may quantify tail swishing by calculating a frequency and amplitude of lateral tail movement over a time interval. In an embodiment, the system may detect tail clamping by identifying a sustained reduction in movement amplitude combined with a maintained downward orientation, and may detect sustained tail elevation by identifying an angular orientation maintained above a predefined angular threshold for a minimum duration. These parameterised definitions of movement patterns may enable consistent and repeatable identification of specific tail behaviours based on measurable signal characteristics.

[0022] In an embodiment, the system may be configured to correlate a detected tail movement deviation with a heart rate exceeding a predefined threshold. Additionally or alternatively, the system may correlate a detected tail movement deviation with a temperature measurement exceeding a predefined threshold, wherein the temperature is obtained using a temperature sensor integrated with the biometric sensor module. By correlating movement-derived and physiological parameters, the system may improve the reliability of detecting combined events compared to evaluating each parameter independently.

[0023] In an embodiment, the system may be configured to detect a combined event by identifying concurrent occurrence of a tail movement deviation and a heart rate above a predefined threshold. Preferably, the system may require that the combined event persists for a minimum time duration before generating an output. Theapplication of temporal persistence criteria may reduce sensitivity to transient fluctuations and improve robustness against false positives arising from short-lived signal anomalies.

[0024] In an embodiment, the system may further comprise a position sensing module configured to determine an ambulation characteristic based on position data over time. The ambulation characteristic may include speed, distance travelled, or movement intensity. The system may be configured to correlate the ambulation characteristic with biometric sensor readings to identify a condition defined by a heart rate above a threshold and an ambulation characteristic below a threshold. This combination of reduced movement and elevated physiological response may provide a technically grounded mechanism for identifying abnormal states.

[0025] Preferably, the system may be configured to detect a gait irregularity based on variation in periodic tail movement corresponding to locomotion cycles. By analysing periodic components of tail movement associated with gait, the system may infer irregularities in locomotion without requiring sensors mounted on the limbs, thereby simplifying deployment while still enabling detection of movement anomalies.

[0026] In an embodiment, the system may be configured to detect a repetitive tail movement pattern by identifying a periodic signal within a defined frequency band over a time window. The system may further compare the periodic signal against a baseline frequency distribution to identify an increase in repetition rate. The use of frequency-domain analysis may enable precise characterisation of repetitive motion and differentiation from irregular or non-periodic movement.

[0027] In an embodiment, the system may be configured to detect a posture event based on a change in tail orientation exceeding a predefined angular threshold and maintained for a predefined duration. By combining angular thresholds with temporal persistence, the system may reliably identify sustained postural changes while reducing sensitivity to transient orientation fluctuations.

[0028] In an embodiment, the biometric sensor module may further comprise a temperature sensor, and the system may be configured to determine a physiological readiness state based on a skin-contact temperature exceeding a baselinetemperature by a predefined margin. Preferably, the baseline temperature may be determined from historical temperature data for the equine. By comparing current temperature to an individual baseline, the system may provide a technically grounded indication of temperature-related changes while accounting for environmental and individual variability.

[0029] In an embodiment, the system may be configured to detect a post-foaling behavioural condition based on proximity between the equine and a foal. The system may include or communicate with an identifier device associated with the foal, the identifier device being configured to transmit an identifier signal detectable by the biometric sensor module or an associated receiver. The processor may be configured to determine a proximity metric between the equine and the foal based on the identifier signal, for example using received signal strength, time-of-flight, or other signal propagation characteristics. The processor may analyse the proximity metric over time to identify a deviation from an expected proximity pattern and / or to identify a feeding interaction characterised by sustained proximity within a predefined distance range for a minimum duration. By analysing proximity-derived interaction patterns, the system may provide an indication of post-foaling behaviour and feeding regularity in a non-invasive manner.

[0030] The combination of baseline modelling, multi-parameter correlation, and threshold-based detection across these embodiments may enable a flexible and scalable monitoring framework that can be adapted to different use cases while maintaining consistent and repeatable signal interpretation.

[0031] Other aspects of the invention are also disclosed.Brief Description of the Drawings

[0032] Notwithstanding any other forms which may fall within the scope of the present invention, preferred embodiments of the disclosure will now be described, by way of example only, with reference to the accompanying drawings in which:

[0033] Figure 1 shows a system schematic of a biometric monitoring system for equines, including a biometric sensor module in communication with a processing system and memory.

[0034] Figure 2 shows the biometric sensor module applied to the underside of the tail dock of an equine.

[0035] Figure 3 shows an inner contact face of the housing of the biometric sensor module, including sensor placement and harness configuration.

[0036] Figure 4 shows an exterior view of the housing of the biometric sensor module, including a user interface.

[0037] Figure 5 shows an exemplary method of monitoring an equine, including acquisition of biometric and tail movement data, baseline determination, correlation, and output generation.Description of Embodiments

[0038] Referring to Figures 1 to 4, a biometric monitoring system 100 is provided for monitoring physiological and behavioural parameters of an equine. The system 100 comprises a biometric sensor module 101 configured to be secured to a tail dock 103 of the equine. The biometric sensor module 101 includes a housing 109 and a harness 102, wherein the harness 102 is configured to attach the housing 109 to the tail dock 103 such that an inner contact face of the housing 109 is maintained in aligned contact with an underside of the tail dock 103.

[0039] The housing 109 defines the inner contact face through which at least one biometric sensor 105 is operative to obtain physiological measurements directly from the tail dock 103. Positioning the biometric sensor 105 at the underside of the tail dock 103 may provide a region of relatively low hair density and consistent skin contact, thereby improving signal quality and reducing optical or electrical interference that may otherwise arise from fur coverage or inconsistent contact at other anatomical locations. Additionally, this placement may reduce exposure of the biometric sensor module 101 to external disturbance, as the underside of the tail dock 103 is generally less accessible to the animal and less prone to impact or dislodgement during normal movement.

[0040] The harness 102 is configured to secure the housing 109 to the tail dock 103 so as to retain the inner contact face in aligned contact with the underside of the tail dock 103 during movement of the equine. As shown in Figure 3, the harness 102 maycomprise a strap 110 extending around the tail dock 103, the strap 110 being adjustable to accommodate different tail dock sizes while maintaining a secure but non-restrictive fit. In an embodiment, the strap 110 may attach to opposed long sides of the housing 109, and may diverge into limb portions along the sides of the housing 109, thereby stabilising the orientation of the housing 109 and reducing rotational or lateral displacement during locomotion. This configuration may assist in maintaining consistent sensor contact and reducing motion artefacts in acquired sensor data.

[0041] The biometric sensor 105 comprises a photoplethysmography sensor configured to obtain heart rate data. The photoplethysmography sensor may include one or more light emitters and one or more photodetectors arranged to emit light into the tissue of the tail dock 103 and detect variations in reflected or transmitted light associated with pulsatile blood flow. The housing 109 may be configured such that the photoplethysmography sensor is exposed through the inner contact face, enabling direct optical coupling with the skin of the tail dock 103. In this manner, the system 100 may obtain continuous heart rate measurements without requiring invasive probes or electrode-based sensing.

[0042] The biometric sensor module 101 further comprises a tail movement sensor, which may be provided as part of an auxiliary sensor 106, configured to detect movement and / or orientation of the tail dock 103. The tail movement sensor may include motion sensing components such as an accelerometer and / or a gyroscope to detect linear acceleration, angular velocity, and orientation changes of the tail dock 103. By monitoring both dynamic motion and static orientation, the tail movement sensor may provide data representative of tail movement patterns, including oscillatory movement, positional changes, and sustained postures.

[0043] The system 100 may comprise a processor configured to perform data acquisition, processing, and analysis operations. In an embodiment, the biometric sensor module 101 comprises the processor in the form of a microprocessor 104, as shown in Figure 1. The microprocessor 104 may be in operable communication with memory 112 and is configured to fetch, decode, and execute computer program code instructions and associated data stored within the memory 112. The microprocessor104 is configured to obtain biometric sensor readings from the photoplethysmography sensor 105 and tail movement data from the auxiliary sensor 106. In some embodiments, the processor functionality may be distributed, such that the microprocessor 104 performs at least part of the processing locally and cooperates with remote processing resources, such as the analytics server 108, via the wireless transceiver 107.

[0044] The processor is further configured to determine baseline biometric sensor data and baseline tail movement data for the equine based on previously acquired data. Such baseline data may be established over a defined observation period and may represent typical physiological and behavioural patterns for the individual equine. The baseline may be stored in memory 112 and may be updated over time to reflect changes in the equine’s normal condition.

[0045] The processor is configured to correlate the biometric sensor readings with the tail movement data by comparing the acquired readings against the respective baseline biometric sensor data and baseline tail movement data to identify deviations. This correlation may involve analysing temporal alignment, frequency characteristics, amplitude variations, and orientation changes in the respective data streams. By integrating biometric and motion-derived information, the system 100 may improve robustness in detecting meaningful deviations while reducing false positives associated with transient or isolated signal anomalies.

[0046] Based on the identified deviations, the processor may determine at least one physiological or behavioural parameter of the equine. The determination may be based on predefined thresholds, pattern recognition algorithms, or other analytical techniques applied to the correlated data. In some implementations, the system 100 may generate an output indicative of the determined parameter, such as an alert or notification transmitted to a user interface, enabling monitoring of the equine’s condition in real time or over extended periods.

[0047] The housing 109 may further be configured to protect internal electronic components while maintaining a lightweight and ergonomic form factor suitable for prolonged use. In some embodiments, the housing 109 may be waterproof, forexample to an IP67 rating, and may be coated in a medical-grade silicone material to enhance durability, comfort, and resistance to environmental exposure. As shown in Figure 4, the housing 109 may further comprise a user interface 111 located on an exterior surface thereof. The user interface 111 may comprise one or more visual indicators, such as light-emitting indicators for operational status, wireless connectivity, battery condition, charging state, or alert indication, and may optionally include a user-actuatable control for activation, pairing, or mode selection. The smooth and compliant surface of the housing 109 may reduce irritation at the contact interface with the tail dock 103 while facilitating cleaning and maintenance.

[0048] In use, the biometric sensor module 101 may be applied to the tail dock 103 as illustrated in Figure 2, with the inner contact face positioned against the underside of the tail dock 103 and the harness 102 securing the module in place. Once positioned, the system 100 may continuously acquire biometric and motion data, establish and update baseline patterns, and perform correlation analysis to provide ongoing monitoring of the equine.

[0049] In an embodiment, the system 100 is configured to determine a respiratory rate by analysing interactions between biometric sensor readings obtained from the photoplethysmography sensor 105 and tail movement data obtained from the auxiliary sensor 106. The processor may process the photoplethysmography signal to identify periodic modulation superimposed on the pulsatile waveform associated with cardiac activity. Such modulation may arise due to respiratory-induced variations in venous return, intrathoracic pressure, and peripheral blood volume, which influence the amplitude and baseline of the photoplethysmography signal.

[0050] Concurrently, the processor may analyse tail movement data to identify cyclic or oscillatory movement patterns of the tail dock 103. These movement patterns may include subtle periodic displacements or oscillations that occur as a result of thoracic expansion and contraction during respiration. In some implementations, the tail movement sensor within the auxiliary sensor 106 may detect low-amplitude rhythmic motion corresponding to breathing cycles, even in the absence of overt tail swishing.

[0051] The processor may correlate the periodic modulation in the photoplethysmography signal with the cyclic tail movement patterns to determine respiratory cycles. This correlation may involve aligning temporal features of the respective signals, such as peaks, troughs, zero-crossings, or envelope variations. In an embodiment, the processor may synchronise peaks in the photoplethysmography signal with oscillatory tail movement to identify individual respiration events. For example, periodic increases or decreases in the amplitude of the photoplethysmography waveform may be matched with corresponding directional changes or oscillations in the tail movement signal.

[0052] Signal processing techniques may be employed to enhance detection accuracy. For instance, the processor may apply filtering operations to isolate frequency components within a range corresponding to expected respiratory rates, thereby reducing noise and motion artefacts unrelated to respiration. Crosscorrelation, phase alignment, or frequency-domain analysis may be used to determine the degree of synchronisation between the photoplethysmography signal modulation and the tail movement signal. By identifying consistent periodic relationships between these signals, the processor may derive a respiratory rate over a defined time interval.

[0053] In some embodiments, the processor may compute the respiratory rate by counting the number of correlated cycles detected within a time window and converting this count into breaths per minute. The system 100 may update the respiratory rate continuously or at defined intervals, enabling real-time monitoring. The use of both photoplethysmography-derived modulation and tail movement data may improve robustness compared to relying on a single signal source, particularly in conditions where motion artefacts or environmental disturbances are present.

[0054] The determination of respiratory rate using this combined approach may provide a non-invasive and continuous means of monitoring respiratory activity from the tail dock 103 location, without requiring sensors positioned on the thorax or head of the equine. This may improve ease of deployment and reduce interference with normal behaviour while still enabling reliable physiological monitoring.

[0055] In an embodiment, the processor is configured to establish a baseline tail movement pattern for the equine based on historical tail movement data obtained from the auxiliary sensor 106. The baseline tail movement pattern may represent a characteristic set of motion parameters for the individual equine under normal conditions, including typical movement frequency, amplitude, and orientation distributions over time. Such baseline data may be generated by recording tail movement data over an initial observation period and may be stored in memory 112 for subsequent comparison.

[0056] The baseline tail movement pattern may be determined using statistical or signal processing techniques. For example, the processor may compute average values, standard deviations, and frequency distributions of tail movement parameters, or may construct time-series profiles representative of typical behaviour. In some embodiments, the baseline may be segmented according to contextual factors such as time of day, activity level, or environmental conditions, thereby enabling more accurate comparisons under varying operating conditions.

[0057] The processor may be configured to detect a deviation from the baseline tail movement pattern by comparing current tail movement data against the established baseline. This comparison may involve analysing at least one of movement frequency, movement amplitude, or tail orientation. For example, the processor may determine whether the frequency of oscillatory tail motion exceeds or falls below a baseline range, whether the amplitude of movement deviates from expected levels, or whether the orientation of the tail dock 103 differs from a typical posture.

[0058] In an embodiment, the processor may quantify tail swishing by calculating a frequency and amplitude of lateral tail movement over a defined time interval. The processor may identify lateral oscillations in the tail movement signal and determine a swishing frequency based on the number of oscillation cycles per unit time. The amplitude of each oscillation may be derived from the magnitude of displacement or acceleration associated with the movement. These values may be compared against baseline swishing characteristics to identify increases in frequency or intensity.

[0059] In an embodiment, the processor may detect tail clamping by identifying a sustained reduction in movement amplitude combined with a maintained downward orientation of the tail dock 103. The processor may analyse orientation data obtained from the tail movement sensor to determine whether the tail is held in a lowered position relative to a baseline orientation. Simultaneously, the processor may assess whether dynamic movement is reduced below a predefined amplitude threshold for a specified duration. The combination of these parameters may be used to identify a clamped tail condition.

[0060] In an embodiment, the processor may detect sustained tail elevation by identifying an angular orientation of the tail dock 103 that is maintained above a predefined angular threshold for a minimum duration. Orientation data derived from the tail movement sensor may be used to determine the angular position of the tail relative to a reference axis. If the tail is held in an elevated position exceeding the threshold angle for a continuous period, the processor may classify this as a sustained elevation event.

[0061] In some implementations, the detection of these movement patterns may involve applying filtering and segmentation techniques to isolate relevant signal components. For example, low-pass or band-pass filters may be used to remove high-frequency noise or unrelated motion artefacts, while windowing techniques may be applied to analyse movement over discrete time intervals. By comparing processed movement data against baseline patterns, the system 100 may reliably identify deviations indicative of changes in the equine’s behavioural state.

[0062] The use of baseline-referenced analysis enables the system 100 to adapt to individual variability between equines, thereby improving the sensitivity and specificity of movement-based assessments. Rather than relying on fixed thresholds applicable to all animals, the system may tailor detection criteria to the normal behaviour of a particular equine, thereby reducing false detections and improving overall monitoring performance.

[0063] In an embodiment, the processor is configured to correlate detected deviations in tail movement with biometric sensor readings to identify combined events basedon multiple physiological inputs. The processor may compare tail movement data against baseline tail movement data to identify a deviation, and may concurrently compare biometric sensor readings, such as heart rate obtained from the photoplethysmography sensor 105, against predefined thresholds or baseline-derived thresholds.

[0064] In an embodiment, the processor may correlate a detected tail movement deviation with a heart rate exceeding a predefined threshold. The predefined threshold may be determined based on typical heart rate ranges for the equine or may be derived from baseline biometric sensor data established for the individual equine. For example, the processor may determine whether the current heart rate exceeds a baseline heart rate by a specified margin or falls outside a statistically defined normal range. By correlating elevated heart rate with deviations in tail movement patterns, such as increased swishing frequency or reduced movement amplitude, the system 100 may provide a more robust assessment of physiological changes than would be possible using either parameter independently.

[0065] In an embodiment, the biometric sensor module 101 may further comprise a temperature sensor configured to obtain temperature measurements from the tail dock 103. The processor may correlate a detected tail movement deviation with a temperature measurement exceeding a predefined threshold. The temperature threshold may be defined relative to an absolute value or relative to a baseline temperature established for the equine. For example, an increase in temperature beyond a baseline range, when occurring concurrently with a deviation in tail movement, may be used to identify combined physiological changes.

[0066] In an embodiment, the processor may be configured to detect a combined event by identifying concurrent occurrence of a tail movement deviation and a biometric parameter exceeding a predefined threshold, such as heart rate or temperature. The processor may evaluate temporal overlap between these conditions, ensuring that the deviation in tail movement and the threshold exceedance in the biometric parameter occur within a defined time window.

[0067] In some embodiments, the processor may require that the combined event persists for a minimum time duration before generating an output. This persistence requirement may be implemented to reduce sensitivity to transient fluctuations or noise in the sensor data. For example, the processor may determine that both the tail movement deviation and the elevated biometric parameter must be maintained continuously, or substantially continuously, for a predetermined duration before the event is classified as significant.

[0068] The processor may implement temporal filtering or hysteresis mechanisms to enforce the persistence condition. For instance, the processor may use a sliding time window to evaluate whether the combined condition remains present over the duration of the window. Alternatively, the processor may require that the condition be detected in a minimum number of consecutive measurement intervals. These approaches may reduce false positives arising from short-lived artefacts or isolated signal spikes.

[0069] In some implementations, the processor may assign weighting factors to different parameters when evaluating combined events. For example, greater weighting may be assigned to sustained heart rate elevation compared to transient movement changes, or vice versa, depending on the monitoring context. The processor may compute a composite score based on the correlated parameters and compare the score against a threshold to determine whether the combined event should be flagged.

[0070] By correlating tail movement deviations with biometric thresholds and enforcing persistence criteria, the system 100 may improve the reliability of event detection. The use of multiple independent data sources, combined with temporal validation, may reduce the likelihood of false detections while enabling earlier identification of meaningful physiological or behavioural changes.

[0071] In an embodiment, the system 100 further comprises a position sensing module, which may be implemented as part of the auxiliary sensor 106. The position sensing module may include a global positioning system (GPS) receiver, an inertial measurement unit (IMU), or a combination of positioning and inertial sensing components configured to determine position and movement of the equine over time.The processor may utilise position data to determine an ambulation characteristic, such as speed, distance travelled, or movement intensity, based on displacement and / or motion-derived measurements.

[0072] The processor may determine the ambulation characteristic by analysing position data over time. For example, speed may be calculated by determining displacement between successive position measurements and dividing by elapsed time. Distance may be determined by accumulating incremental displacements over a defined period. In some embodiments, the processor may combine GPS data with inertial data to improve accuracy, particularly in environments where satellite signal quality is reduced, such as enclosed stables or covered arenas.

[0073] In an embodiment, the processor is configured to correlate the ambulation characteristic with biometric sensor readings to identify a condition defined by a heart rate above a threshold and an ambulation characteristic below a threshold. The heart rate threshold may be derived from baseline biometric sensor data or predefined physiological ranges, while the ambulation threshold may represent a minimum expected level of movement for the equine under normal conditions. By identifying a combination of elevated heart rate and reduced movement, the processor may detect a pattern indicative of abnormal physiological behaviour, while remaining grounded in measurable signal relationships.

[0074] In an embodiment, the processor may further analyse temporal trends in ambulation characteristics in conjunction with biometric sensor readings. For example, the processor may detect sustained reductions in movement over a defined period while monitoring corresponding changes in heart rate or other biometric parameters. Such temporal correlation may enhance robustness by distinguishing between transient inactivity and prolonged deviations from normal movement patterns.

[0075] In an embodiment, the processor is configured to detect a gait irregularity based on variation in periodic tail movement corresponding to locomotion cycles. The tail movement sensor may capture rhythmic movement of the tail dock 103 associated with locomotion, such as oscillations that occur during walking, trotting, or galloping.The processor may analyse this periodic movement to identify deviations from a baseline locomotion pattern, including irregular timing, asymmetry, or variation in amplitude between successive cycles.

[0076] The processor may extract locomotion-related features from the tail movement data by identifying periodic components within a frequency range corresponding to expected gait cycles. Signal processing techniques such as frequency-domain analysis, peak detection, or autocorrelation may be used to characterise the periodicity and consistency of the movement. Deviations from baseline periodic characteristics, such as irregular cycle timing or inconsistent amplitude, may be used to identify irregular gait patterns.

[0077] In some embodiments, the processor may correlate detected gait irregularities with other parameters, such as ambulation characteristics or biometric sensor readings, to provide additional context for the detected variation. For example, a reduction in speed combined with irregular periodic tail movement may indicate a change in locomotion behaviour. By deriving gait-related information from tail movement data obtained at the tail dock 103, the system 100 may provide insight into locomotion without requiring sensors mounted on the limbs of the equine.

[0078] The integration of ambulation characteristics with biometric and tail movement data enables the system 100 to evaluate multiple aspects of equine behaviour simultaneously. This multi-parameter approach may improve the robustness of detection by ensuring that identified patterns are supported by consistent trends across independent data sources.

[0079] In an embodiment, the processor is configured to detect a repetitive tail movement pattern by identifying a periodic signal within a defined frequency band over a time window. The tail movement data obtained from the auxiliary sensor 106 may be processed to isolate periodic components associated with repeated tail motion. For example, the processor may apply band-pass filtering to the tail movement signal to extract frequency components within a range corresponding to expected tail oscillation frequencies. This may allow the processor to distinguish repetitive motion from irregular or transient movements.

[0080] The processor may analyse the filtered signal using techniques such as spectral analysis, autocorrelation, or peak detection to identify periodicity within the defined frequency band. For instance, the processor may compute a power spectral density of the tail movement signal and identify dominant frequency components, or may detect recurring peaks in the time-domain signal at regular intervals. By identifying consistent periodic patterns over a defined time window, the processor may characterise repetitive tail movement.

[0081] In an embodiment, the processor is configured to compare the identified periodic signal against a baseline frequency distribution to identify an increase in repetition rate. The baseline frequency distribution may be derived from previously acquired tail movement data and may represent typical frequencies and amplitudes of tail motion for the individual equine. The processor may determine whether the detected periodic signal exhibits a frequency, amplitude, or energy level that deviates from the baseline distribution. For example, an increase in repetition rate may be identified when the dominant frequency of the tail movement signal exceeds a baseline frequency range or when the energy associated with that frequency increases beyond a threshold.

[0082] The processor may perform this comparison using statistical measures, such as comparing the current frequency to a mean baseline frequency with an associated tolerance, or by evaluating whether the current signal falls outside a predefined confidence interval derived from baseline data. In some implementations, the processor may track changes in repetition rate over time to identify trends, rather than relying solely on instantaneous deviations.

[0083] In an embodiment, the processor is configured to detect a posture event based on a change in tail orientation exceeding a predefined angular threshold and maintained for a predefined duration. Orientation data obtained from the tail movement sensor may be used to determine the angular position of the tail dock 103 relative to a reference orientation. The processor may identify a posture event when the tail orientation changes beyond a specified angular threshold and remains in that orientation for a continuous period exceeding the predefined duration.

[0084] In an embodiment, the processor may be configured to detect tail orientation and movement patterns associated with foaling behaviour. For example, the processor may identify a sequence of sustained tail elevation events, repeated changes in tail orientation, or intermittent clamping and release patterns occurring over a defined time period. The processor may analyse temporal patterns of such orientation and movement data, including frequency, duration, and repetition of posture events, and compare these patterns against baseline tail behaviour for the equine. In some implementations, the processor may identify a foaling-related event when a combination of sustained tail elevation and repeated orientation changes persists over a predefined duration or exceeds a threshold pattern. This approach enables detection of foaling-related behavioural changes using quantified tail movement and orientation data without reliance on intrusive sensing.

[0085] The detection of posture events may involve analysing both instantaneous orientation and temporal persistence. For example, the processor may first identify a candidate orientation change event when the measured angle exceeds the threshold, and may then verify that the orientation remains within a defined range for the required duration. This two-stage approach may reduce sensitivity to transient movements and improve robustness in identifying sustained postural changes.

[0086] In some embodiments, the processor may apply smoothing or filtering to the orientation data to reduce noise and ensure stable detection of posture events. For example, a moving average filter may be applied to the orientation signal to mitigate short-term fluctuations. Additionally, hysteresis thresholds may be used to prevent rapid switching between states when the orientation is near the threshold boundary.

[0087] In an embodiment, the processor may be further configured to detect postfoaling behavioural patterns based on proximity and interaction between a mare and a foal. The system may utilise position data obtained from the position sensing module to determine a relative proximity between the monitored equine and a second equine, such as a foal, wherein the second equine may be associated with an identifier device. The identifier device may comprise a wireless tag, for example a Bluetooth Low Energy (BLE) beacon, ultra-wideband (UWB) tag, radio-frequencyidentification (RFID) tag, or other short-range communication device configured to periodically transmit an identifier signal. The biometric sensor module 101 or an associated receiver may be configured to detect the identifier signal and determine a proximity metric based on at least one of received signal strength indicator (RSSI), time-of-flight measurement, phase difference, or other signal propagation characteristics. In some embodiments, proximity may be estimated by converting RSSI values into an approximate distance using a calibrated path-loss model, optionally adjusted based on environmental factors. In other embodiments, higher-precision ranging may be achieved using UWB time-of-flight or time-difference-of-arrival techniques to determine relative distance between the mare-mounted module and the foal-associated tag. The processor may analyse proximity data over time to determine whether the mare and foal remain within a defined distance range for expected durations following a foaling event. The processor may further apply temporal filtering, such as moving averages or hysteresis thresholds, to mitigate transient signal fluctuations and reduce false proximity transitions. Deviations from expected proximity patterns, such as prolonged separation, absence of detected identifier signals, or irregular proximity intervals, may be identified as indicative of atypical post-foaling behaviour.

[0088] In an embodiment, the processor may be configured to detect feeding regularity associated with mare-foal interaction by analysing temporal patterns of proximity events in combination with movement and orientation data. The processor may identify candidate feeding events based on sustained proximity between the mare-mounted biometric sensor module 101 and the foal-associated identifier device within a predefined distance threshold for a minimum duration. In some implementations, the processor may further refine detection of feeding events by analysing orientation data of the tail dock 103 and / or low-amplitude motion signatures indicative of stationary or near-stationary posture during nursing. For example, the processor may detect a combination of reduced tail movement amplitude, stable orientation, and close-proximity persistence as a feeding state. The processor may segment the proximity signal into discrete interaction intervals and computeparameters including frequency of feeding events, duration of each event, and intervals between successive events over a defined monitoring period. These parameters may be compared against baseline or expected feeding profiles for the individual equine or for a population model, using statistical measures such as mean interval duration, variance, or deviation thresholds. In some embodiments, the processor may apply clustering or pattern recognition techniques to distinguish feeding-related proximity events from incidental proximity, such as brief encounters or environmental co-location. The processor may further correlate feeding-related proximity patterns with other sensed parameters, including heart rate variability, temperature, or posture events, to increase confidence in detection and to identify irregular feeding behaviour, such as reduced feeding frequency, inconsistent intervals, shortened feeding duration, or absence of expected feeding events over a defined time window.

[0089] By identifying repetitive movement patterns and sustained posture events using frequency-based and orientation-based analysis, the system 100 may characterise tail behaviour in a structured and quantifiable manner. The use of baseline frequency distributions and threshold-based orientation detection enables consistent identification of deviations while accommodating variability in individual equine behaviour.

[0090] In an embodiment, the biometric sensor module 101 further comprises a temperature sensor configured to obtain a skin-contact temperature measurement from the tail dock 103. The temperature sensor may be integrated within the inner contact face of the housing 109 such that it maintains direct thermal contact with the skin of the tail dock 103. This configuration may enable continuous measurement of surface temperature while minimising the influence of ambient air temperature fluctuations.

[0091] The processor is configured to determine a physiological readiness state based on a skin-contact temperature exceeding a baseline temperature by a predefined margin. The baseline temperature may represent a typical resting or nonexertion temperature for the individual equine and may be established from previouslyacquired temperature data. The predefined margin may be selected to represent a meaningful increase above baseline, indicative of a change in physiological state.

[0092] In an embodiment, the processor may compare the current temperature measurement against the baseline temperature and determine whether the difference exceeds the predefined margin. The margin may be defined as an absolute temperature difference or as a relative percentage increase from the baseline. For example, the processor may determine that a readiness state is reached when the temperature exceeds the baseline by a specified number of degrees or when the temperature trend demonstrates a sustained upward deviation over a defined period.

[0093] The processor may incorporate temporal filtering when determining the readiness state to ensure that transient temperature fluctuations do not trigger false detections. For instance, the processor may require that the temperature remains above the baseline plus the predefined margin for a minimum duration before confirming the readiness state. Alternatively, the processor may evaluate a rate of change of temperature over time to identify gradual warming trends.

[0094] In an embodiment, the processor is configured to determine the baseline temperature from historical temperature data for the equine. The historical data may be stored in memory 112 and may be used to compute a representative baseline value using statistical measures such as averaging over selected time intervals. In some implementations, the baseline temperature may be periodically updated to reflect long-term changes in the equine’s physiological state or environmental conditions.

[0095] The determination of physiological readiness based on temperature deviation may provide a non-invasive indication of changes in the equine’s condition associated with activity or environmental factors. By utilising baseline-referenced comparison and persistence criteria, the system 100 may provide a reliable indication of temperature-related changes while reducing sensitivity to short-term variations.

[0096] In some embodiments, the readiness determination may be integrated with other parameters, such as heart rate or tail movement data, to provide additional context. For example, the processor may evaluate whether an increase intemperature is accompanied by corresponding changes in heart rate or movement patterns. However, the determination of the readiness state based on temperature deviation may be performed independently, thereby enabling flexibility in how the system 100 is configured and deployed.Exemplary system architecture

[0097] Referring again to Figure 1, one exemplary implementation of the system 100 may be realised as a compact tail-dock-mounted telemetry node forming the biometric sensor module 101 and a remote analytics platform 108 in data communication therewith. In one practical arrangement, the biometric sensor module 101 may be built around a Nordic nRF52840 system-on-chip serving as the local controller and short-range communications device. The nRF52840 provides a 64 MHz Arm Cortex-M4 processor and supports Bluetooth Low Energy and other 2.4 GHz protocols, making it suitable for local acquisition, packetisation and wireless forwarding of sensor data from the module 101.

[0098] In this example, the biometric sensor 105 may be implemented using an optical acquisition front end such as an Analog Devices MAX86141. The MAX86141 is a highly integrated optical data acquisition device suited to photoplethysmography, and may be coupled to one or more light emitters and one or more photodetectors positioned so as to operate through the inner contact face of the housing 109 against the underside of the tail dock 103. The light emitters may, for example, comprise green and / or infrared emitters selected to improve performance across different coat, skin and ambient-light conditions. The controller may drive the optical front end in a sampled mode, such as 50 Hz to 200 Hz, and may adapt LED current, pulse width and sampling cadence to maintain signal quality while conserving battery energy.

[0099] The auxiliary sensor 106 may, in one arrangement, comprise an inertial measurement unit such as a Bosch BMI270, which combines triaxial accelerometer and triaxial gyroscope sensing in a compact wearable-oriented package. The inertial measurement unit may be mounted on the printed circuit board within the housing 109 with its sensing axes aligned relative to the longitudinal axis of the housing 109 so that lateral swishing, sustained downward clamping and raised-tail angular posturecan be resolved consistently. In use, accelerometer and gyroscope data may be sampled at different rates depending on operating mode, for example a lower-rate background monitoring mode and a higher-rate event capture mode when motion exceeding a trigger threshold is detected.

[0100] Where location or ambulation monitoring is required, the auxiliary sensor 106 may further include a GNSS receiver, such as a u-blox MAX-M10 or UBX-M10 series module. Such modules are designed for low-power positioning and may be used to derive speed, displacement and geofence-related events while preserving battery autonomy. In some configurations, GNSS acquisition may be duty-cycled, for example activated only every few seconds or minutes during paddock or transport monitoring, and omitted altogether during stable monitoring where position change is minimal. The local controller may further combine GNSS-derived position updates with inertial data to smooth movement estimation when satellite reception is degraded.

[0101] The wireless transceiver 107 may be implemented in more than one form according to deployment context. In a short-range arrangement, the nRF52840 may transmit data directly by Bluetooth Low Energy to a nearby gateway device such as a mobile telephone, tablet, stable hub or trailer-mounted receiver. In a longer-range arrangement, the module 101 may additionally or alternatively include a Semtech SX1262 sub-GHz transceiver to provide LoRa-based LPWAN communication across paddocks, rural properties or large equestrian facilities. In such an arrangement, the controller may select between Bluetooth and LoRa transmission modes according to link availability, battery state, or urgency of the detected event. For example, routine telemetry may be buffered and uploaded opportunistically over Bluetooth, whereas threshold events may be transmitted immediately over the available long-range link.

[0102] At firmware level, the controller may execute a layered processing pipeline. A low-level acquisition layer may read the optical and inertial sensors over an I2C and / or SPI bus. A signal-conditioning layer may apply ambient-light compensation, baseline wander removal, motion-artifact suppression, and low-pass or band-pass filtering to the raw photoplethysmography signal. By way of example, the PPG channel may be processed using a finite impulse response or infinite impulseresponse band-pass filter spanning an expected equine pulse band, while inertial channels may be filtered to isolate posture drift, low-frequency respiration-associated motion, and higher-frequency swishing events. A feature-extraction layer may then derive heart-rate peaks, inter-beat intervals, movement amplitude envelopes, angular orientation traces, oscillation frequency, and temporal persistence metrics. In some embodiments, baseline modelling may be performed using rolling windows, exponentially weighted moving averages, percentile bands, or horse-specific z-score normalisation so that current readings can be compared against previously acquired data for the same animal. These techniques may be implemented locally within the controller or remotely at the analytics platform 108 depending on available power and communication bandwidth.

[0103] In one practical communications architecture, the module 101 may publish timestamped telemetry packets using MQTT. Where a cloud service is employed, the packets may be ingested through AWS loT Core, which provides a managed MQTT broker and device connectivity infrastructure. The payload may include, for example, device identifier, horse identifier, PPG-derived heart-rate features, filtered motion features, posture state, temperature, battery state and optionally location. The analytics server 108 may then consume the stream and write the incoming data into a time-series database such as InfluxDB, enabling rapid storage and querying of historical telemetry and baseline model generation over extended periods.

[0104] The analytics platform 108 may execute condition-analysis logic using a combination of deterministic rules and trained statistical models. By way of example, respiratory rate may be computed by identifying amplitude modulation in the PPG waveform and synchronising that modulation with low-frequency oscillation in the inertial signal. Tail clamping may be identified where a downward angular posture persists beyond a duration threshold while motion amplitude remains below a floor value. Repetitive swishing may be identified by spectral energy within a defined frequency band exceeding a horse-specific baseline distribution. A combined event may be raised where elevated heart rate coincides with abnormal posture or reduced ambulation for longer than a persistence window. In some implementations, theanalytics platform may maintain multiple contextual baselines for the same equine, such as paddock baseline, transport baseline and overnight stable baseline, thereby allowing the same raw signals to be interpreted differently according to operating context.

[0105] Power management may also be implemented as part of the exemplary architecture. The housing 109 may contain a rechargeable lithium-polymer cell with a charge-management integrated circuit and a regulated power rail for the optical sensor, inertial sensor and communications circuitry. The controller may switch between an active monitoring mode, an event-confirmation mode and a low-power standby mode. For example, during quiescent periods the optical sensor may be sampled intermittently and the inertial sensor may operate in a wake-on-motion mode; upon detection of an orientation change, elevated PPG variability, or a threshold event, the controller may temporarily increase sample rate and transmission cadence. Such duty-cycled operation may extend service life while preserving the ability to capture short-duration physiological and behavioural changes.

[0106] A mobile or web-based user interface associated with the analytics platform 108 may display live and historical traces of heart rate, temperature, posture state, swishing rate, ambulation metrics and alert history. In one arrangement, the platform may provide configurable alert rules, such as a combined-event rule requiring elevated heart rate together with reduced ambulation for a minimum duration, or a posture-event rule requiring sustained tail elevation beyond an angular threshold. The same platform may also support veterinarian and trainer views, with the former emphasising long-term physiological trends and the latter emphasising readiness, transport stress and pre-exercise monitoring.Exemplary method of monitoring

[0107] Referring to Figure 5, an exemplary method 200 of monitoring an equine using the system 100 will now be described in greater technical detail.

[0108] At step 201, a biometric sensor module 101 is applied to the tail dock 103 of the equine. The housing 109 is positioned such that the inner contact face is placed against the underside of the tail dock 103, which may present a region of reducedhair density and improved vascular access. The harness 102, including strap 110, is secured around the tail dock 103 and adjusted to maintain consistent contact pressure between the inner contact face and the skin surface. In some implementations, the strap 110 may be tensioned to achieve a target contact force range sufficient to maintain optical coupling for the photoplethysmography sensor 105 while avoiding excessive compression that may occlude blood flow or cause discomfort.

[0109] At step 202, biometric sensor readings are obtained using the photoplethysmography sensor 105. The photoplethysmography sensor 105 may be operated in a sampled mode in which one or more light emitters are driven with a defined pulse sequence, for example at a sampling rate between approximately 50 Hz and 200 Hz. The emitted light may penetrate the tissue of the tail dock 103 and be partially absorbed and reflected in accordance with local blood volume. A photodetector may capture the returned optical signal, which may include a pulsatile component corresponding to cardiac cycles and a slowly varying component corresponding to baseline tissue properties. The processor may digitise the optical signal and apply signal conditioning, such as ambient light subtraction, DC offset removal, and band-pass filtering to isolate the pulsatile waveform associated with heart activity.

[0110] At step 203, tail movement data is obtained from the auxiliary sensor 106, which may include an accelerometer and / or gyroscope. The accelerometer may provide linear acceleration data along multiple axes, while the gyroscope may provide angular velocity data indicative of rotational movement of the tail dock 103. In some embodiments, the processor may derive orientation data from the inertial signals using sensor fusion techniques, such as complementary filtering or Kalman filtering, to estimate angular position relative to a reference frame. The tail movement data may therefore represent both dynamic motion (e.g. oscillations associated with swishing) and static or quasi-static orientation (e.g. raised or lowered tail posture).

[0111] At step 204, the acquired biometric sensor readings and tail movement data are time-stamped and either stored in memory 112 or transmitted via the wirelesstransceiver 107 to the analytics server 108. In some implementations, the processor may buffer data locally and transmit it in packets at defined intervals, while in other implementations, data may be streamed in near real-time. Time synchronisation between biometric and motion data streams may be maintained to enable accurate correlation in subsequent processing.

[0112] At step 205, baseline biometric sensor data and baseline tail movement data are determined for the equine based on previously acquired data. The processor may compute baseline values using statistical techniques, such as calculating mean, median, variance, or percentile ranges over a defined historical window. For example, baseline heart rate may be established by averaging photoplethysmography-derived heart rate measurements over periods of rest, while baseline tail movement may be characterised by frequency distributions of oscillatory motion and typical orientation angles during normal activity. In some embodiments, the baseline may be adaptive, with the processor updating baseline parameters using a rolling window or exponentially weighted averaging to account for gradual changes in the equine’s condition.

[0113] At step 206, the biometric sensor readings are processed to identify periodic modulation within the photoplethysmography signal. In addition to detecting primary cardiac peaks, the processor may analyse amplitude modulation or envelope variations in the signal that occur at a lower frequency corresponding to respiratory activity. For example, the processor may compute an envelope of the photoplethysmography waveform and apply frequency-domain analysis, such as a fast Fourier transform, to identify dominant low-frequency components.

[0114] At step 207, the tail movement data is processed to identify cyclic or periodic movement patterns and orientation characteristics. The processor may apply filtering to isolate motion within a frequency band corresponding to expected respiratory-induced tail movement, for example low-frequency oscillations associated with thoracic expansion. Additionally, the processor may detect higher-frequency oscillations corresponding to tail swishing by identifying repeated lateral movementswithin a defined frequency range. Orientation data may be analysed to determine angular position relative to a baseline posture.

[0115] At step 208, the biometric sensor readings are correlated with the tail movement data by comparing both data streams against their respective baseline datasets. This correlation may involve aligning temporal features, such as matching peaks in the photoplethysmography envelope with oscillations in the tail movement signal. The processor may compute cross-correlation functions, phase relationships, or synchronisation metrics to determine the degree of alignment between the signals. For example, periodic modulation in the photoplethysmography signal may be matched with cyclic tail motion to identify respiratory cycles.

[0116] At step 209, a deviation from the baseline is identified based on the correlation. The processor may determine whether one or more parameters fall outside baseline ranges, such as a heart rate exceeding a baseline by a defined margin, an increase in tail swishing frequency beyond a baseline distribution, or a sustained change in tail orientation exceeding a threshold angle. In some implementations, the processor may evaluate combined deviations, such as concurrent occurrence of elevated heart rate and reduced tail movement amplitude, or repetitive oscillatory motion within a defined frequency band exceeding baseline levels.

[0117] At step 210, at least one physiological or behavioural parameter of the equine is determined based on the identified deviation. For example, the processor may determine a respiratory rate by counting correlated cycles between photoplethysmography modulation and tail movement oscillations over a time interval. In another example, the processor may identify a posture event by detecting a sustained tail orientation exceeding a threshold for a predefined duration. In further examples, the processor may identify a combined event where elevated heart rate and reduced ambulation, as determined from position sensing data, occur simultaneously over a persistence window.

[0118] At step 211, an output is generated indicative of the determined parameter. The output may comprise a data record, alert, or notification transmitted to a user interface associated with the system 100. In some embodiments, the output mayinclude time-stamped event information, parameter values, and confidence metrics. For example, an alert may be generated when a combined event persists for a minimum duration, such as elevated heart rate above a threshold together with reduced movement below a threshold. The output may be communicated to a mobile device, web interface, or monitoring dashboard for review by a user.

[0119] The method 200 may be executed continuously or in a duty-cycled manner, depending on power management strategies implemented within the biometric sensor module 101. For example, the system may operate in a low-power monitoring mode during periods of inactivity and transition to a higher sampling rate mode upon detection of significant movement or biometric variation. By integrating optical sensing, inertial sensing, baseline modelling, and multi-parameter correlation, the method 200 provides a structured approach for deriving physiological and behavioural parameters from measurements obtained at the tail dock 103.

[0120] The foregoing description, for purposes of explanation, used specific nomenclature to provide a thorough understanding of the invention. However, it will be apparent to one skilled in the art that specific details are not required in order to practise the invention. Thus, the foregoing descriptions of specific embodiments of the invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed as obviously many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, thereby enabling others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the following claims and their equivalents define the scope of the invention.

Claims

Claims1. A biometric monitoring system for an equine, the system comprising:a biometric sensor module including a housing and a harness,wherein the housing defines an inner contact face configured for placement against an underside of a tail dock of the equine such that at least one biometric sensor is operative through the inner contact face in direct contact with the tail dock;wherein the harness is configured to secure the housing to the tail dock to retain the inner contact face in aligned contact with the underside of the tail dock during movement of the equine;wherein the biometric sensor comprises a photoplethysmography sensor configured to obtain heart rate data;wherein the biometric sensor module further comprises a tail movement sensor configured to detect movement and / or orientation of the tail dock; anda processor configured to:obtain biometric sensor readings using the photoplethysmography sensor; obtain tail movement data from the tail movement sensor;determine baseline biometric sensor data and baseline tail movement data for the equine based on previously acquired data; andcorrelate the biometric sensor readings with the tail movement data by comparing the readings against the respective baseline biometric sensor data and baseline tail movement data to identify a deviation and determine at least one physiological or behavioural parameter of the equine.

2. The system of claim 1, wherein the processor is configured to determine a respiratory rate by identifying periodic modulation in a photoplethysmography signal and correlating the modulation with cyclic tail movement detected by the tail movement sensor.

3. The system of claim 2, wherein the processor is configured to synchronise peaks in the photoplethysmography signal with oscillatory tail movement to identify respiration cycles.

4. The system of claim 1, wherein the processor is configured to establish a baseline tail movement pattern for the equine based on historical tail movement data.

5. The system of claim 4, wherein the processor is configured to detect a deviation from the baseline tail movement pattern by comparing at least one of frequency, amplitude, or orientation of tail movement against the baseline.

6. The system of claim 5, wherein the processor is configured to quantify tail swishing by calculating a frequency and amplitude of lateral tail movement over a time interval.

7. The system of claim 5, wherein the processor is configured to detect tail clamping by identifying a sustained reduction in movement amplitude combined with a maintained downward orientation.

8. The system of claim 5, wherein the processor is configured to detect sustained tail elevation by identifying an angular orientation maintained above a predefined angular threshold for a minimum duration.

9. The system of claim 1, wherein the processor is configured to correlate a detected tail movement deviation with a heart rate exceeding a predefined threshold.

10. The system of claim 1, wherein the biometric sensor module further comprises a temperature sensor, and wherein the processor is configured to correlate a detected tail movement deviation with a temperature measurement exceeding a predefined threshold.

11. The system of claim 1 , wherein the processor is configured to detect a combined event by identifying concurrent occurrence of:(a) a tail movement deviation; and(b) a heart rate above a predefined threshold.

12. The system of claim 11, wherein the processor is further configured to require that the combined event persists for a minimum time duration before generating an output.

13. The system of claim 1, further comprising a position sensing module, wherein the processor is configured to determine an ambulation characteristic based on position data over time.

14. The system of claim 13, wherein the processor is configured to correlate the ambulation characteristic with biometric sensor readings to identify a condition defined by a heart rate above a threshold and an ambulation characteristic below a threshold.

15. The system of claim 13, wherein the processor is configured to detect a gait irregularity based on variation in periodic tail movement corresponding to locomotion cycles.

16. The system of claim 1, wherein the processor is configured to detect a repetitive tail movement pattern by identifying a periodic signal within a defined frequency band over a time window.

17. The system of claim 1, wherein the processor is configured to detect a postfoaling behavioural condition based on proximity between the equine and a foal, the processor being configured to:receive a signal from an identifier device associated with the foal;determine a proximity metric between the equine and the foal based on the signal; andanalyse the proximity metric over time to identify at least one of:a deviation from an expected proximity pattern; ora feeding interaction characterised by sustained proximity within a predefined distance range for a minimum duration.

18. The system of claim 1, wherein the processor is configured to detect a posture event based on a change in tail orientation exceeding a predefined angular threshold and maintained for a predefined duration.

19. The system of claim 1, wherein the biometric sensor module further comprises a temperature sensor, and wherein the processor is configured to determine a physiological readiness state based on a skin-contact temperature exceeding a baseline temperature by a predefined margin.

20. The system of claim 19, wherein the processor is configured to determine the baseline temperature from historical temperature data for the equine.

Images