WO2015030611A1 - Procédé et appareil de détermination des caractéristiques de respiration d'un animal - Google Patents

Procédé et appareil de détermination des caractéristiques de respiration d'un animal Download PDF

Info

Publication number
WO2015030611A1
WO2015030611A1 PCT/NZ2014/000185 NZ2014000185W WO2015030611A1 WO 2015030611 A1 WO2015030611 A1 WO 2015030611A1 NZ 2014000185 W NZ2014000185 W NZ 2014000185W WO 2015030611 A1 WO2015030611 A1 WO 2015030611A1
Authority
WO
WIPO (PCT)
Prior art keywords
animal
respiration
determining
indication
image
Prior art date
Application number
PCT/NZ2014/000185
Other languages
English (en)
Inventor
Allan SCHAEFER
Mairi Stewart
Mark Wilson
Original Assignee
Interag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Interag filed Critical Interag
Publication of WO2015030611A1 publication Critical patent/WO2015030611A1/fr

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
    • A61B5/083—Measuring rate of metabolism by using breath test, e.g. measuring rate of oxygen consumption
    • A61B5/0836—Measuring rate of CO2 production
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
    • A61B5/0075—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2503/00—Evaluating a particular growth phase or type of persons or animals
    • A61B2503/40—Animals

Definitions

  • Embodiments of the present invention relates to a method and apparatus for determining at least one respiratory related characteristic of an animal.
  • embodiments of the present invention relate to determining respiratory characteristics of an animal using infrared images.
  • respiration rate is the act of breathing, or more specifically, the acts of taking in oxygen, using it in the body tissues, and giving off carbon dioxide.
  • the respiratory system of mammals is frequently subjected to primary and secondary diseases which affect respiratory rate. Negative affective experiences relating to respiration such as breathlessness have been identified as being highly significant to animal welfare in "Introducing breathlessness as a significant animal welfare issue", Beausoleil and Mellor (2014). Further, respiratory rate has been identified as an indicator of stress and/or pain - for example heat stress and stress severity. Automatic monitoring of respiration rate could potentially enable quick detection of the symptoms of heat stress and provides a means for assessing the severity of a stress condition. This can reduce performance losses and even prevent the death of animals.
  • a respiration rate monitor was suggested in "Development of a New Respiration Rate Monitor for Cattle", Eigenberg et al (2000).
  • the monitor requires securing a belt to the animal, with the belt holding the sensor in place.
  • Such a system is impractical in practice, with many animals (cows in particular) being highly sensitive to changes in routine - especially devices being attached to them. This has the potential to cause changes in normal behavioural patterns and increase stress levels in the animal - affecting both animal welfare and accuracy of the monitor.
  • There are other drawbacks to the system including being susceptible to damage or displacement. Further, costs do not scale well for larger operations - with each animal to be monitored requiring a belt necessitating additional labour to fit, and costs associated with ongoing maintenance.
  • the feeding monitoring approach uses the measured feed consumed for a measured amount of growth to calculate which animals are comparatively more efficient.
  • a method of determining at least one characteristic of respiration of a non-human animal including the steps of: obtaining a plurality of thermal infrared images of the animal using at least one thermographic camera, the area of the images capturing a region of interest relating to respiration of the animal; analysing each image to obtain data relating to respiration of the animal; and determining at least one characteristic of respiration of the animal based at least in part on the data obtained from the images.
  • an apparatus for determining respiratory rate of a non-human animal including: at least one thermographic camera configured to obtain a plurality of thermal infrared images of the animal, the area of the images capturing a region of interest relating to respiration of the animal; and at least one processor configured to: analyse each image to obtain data relating to respiration of the animal; and determine at least one characteristic of respiration of the animal based at least in part on the data obtained from the images.
  • thermographic camera should be understood to mean an imaging device forming images from infrared radiation.
  • thermographic cameras include those commercially available from FLIR Systems Inc. of USA, the general configuration and calibration of which is described, for example, in US Patent No. 7679046.
  • thermographic camera may be configured to detect absorption of infrared radiation in a selected spectral bandwidth - for example that corresponding to an infrared absorption band of one or more gasses.
  • a selected spectral bandwidth for example that corresponding to an infrared absorption band of one or more gasses.
  • An example of such a configuration is described in US Patent No. 8653461 .
  • thermographic camera may be configured to apply selectable filters in order to capture images of desired targets - for example as described in US Patent No. 8223206. It should be appreciated that this is not intended to be limiting, and that in exemplary embodiments multiple thermographic cameras may be provided, each having a desired filter.
  • Reference to a characteristic of respiration should be understood to mean any feature or quality associated with the action of breathing which may be desirable to be identified or determined.
  • the characteristic may be rate, depth, pattern, flow rate, volume, or gas content (such as carbon dioxide), or variations in these measurements.
  • a region of interest should be understood to mean a selected subset of information captured by the images. More particularly, the region of interest should be understood to mean an area within the image capturing a feature from which a characteristic of respiration may be determined.
  • the region of interest may include at least a portion of a nostril of the animal.
  • the region of interest may include the space into which an animal exhales.
  • the plurality of images may be obtained by sampling a thermal infrared video at a desired frequency, or discretely capturing each image using the camera. In either case, each image may be time stamped in order to determine the relative timing of data contained therein in comparison with that of another image.
  • thermographic camera may be positioned so as to focus the viewing area of the camera, and therefore images obtained, on a region of interest of the animal. However, it is envisaged that increased accuracy and processing efficiencies may be gained by the removal of excess information from the images.
  • analysing each image to obtain the data relating to respiration of the animal includes: determining the region of interest in the image; cropping the image to the region of interest; and analysing the cropped image to obtain the data relating to respiration of the animal.
  • Determination of the region of interest may be performed using any suitable means known to those skilled in the art of image processing.
  • feature detection may be used to automatically detect the location of the at least one nostril of the animal in the image, which are unique in shape and may be readily differentiated from other features of the animal.
  • the surrounding region of interest may be readily determined.
  • greater sensitivity may be achieved by more closely defining the region of interest and reducing the amount of noise.
  • thermographic camera may be captured in a controlled environment with known dimensions - for example in a milking station or feed station - it may be desirable to account for situations in which the distance may vary beyond a desired range, particularly where a pixel count is used to obtain temperature data.
  • distance may be determined using time of flight or other ranged imaging techniques known in the art.
  • the at least one characteristic of respiration to be determined may be respiration rate, determination of which may be based at least in part on changes in temperature between the images.
  • a method of determining respiratory rate of a animal including the steps of: obtaining a plurality of thermal infrared images of the animal using at least one thermographic camera, the area of the images capturing at least a portion of one nostril of the animal; analysing each image to obtain temperature data; determining respiration rate of the animal based at least in part on changes in temperature between the images.
  • an apparatus for determining respiratory rate of a animal including: at least one thermographic camera configured to obtain a plurality of thermal infrared images of the animal, the area of the images capturing at least one nostril of the animal; and at least one processor configured to: analyse each image to obtain temperature data; and determine respiration rate of the animal based at least in part on changes in temperature between the images.
  • analysing each image to obtain the temperature data includes applying a binary filter to the image such that each pixel is filtered relative to a temperature threshold.
  • the pixels above the threshold in the resulting image may be counted to provide a relative indication of temperature.
  • inspiration brings in external air that cools the nostrils while expiration brings warm air from the body core to the nostrils.
  • Mean nostril temperature is generally proportional to air flow, and as such the coldest temperature may generally be identified as the peak of inspiration, and the warmest temperature as the nadir of expiration.
  • the temperature threshold for the binary filter may be established by sampling the temperature between the nostrils of the animal at a peak of inspiration - i.e. when the animal has fully inhaled and the mean nostril temperature is generally at its coldest. The position between the nostrils is likely to be at or below this temperature, and may also be relatively easy to locate given the visually distinctive nature of the nostrils. It should be appreciated that this is not intended to be limiting, and that the threshold may be adjusted after an initial calibration or set by other means. It should be appreciated that in an exemplary embodiment the thermographic camera may be configured to only capture image data above this temperature threshold. However, it is envisaged that capturing broader temperature spectrum information may allow for greater flexibility in terms of adjusting the threshold to suit environmental conditions and baseline animal characteristics.
  • respiration rate may be determined by counting the number of turning points of the temperature data over a time period. It should be appreciated that in exemplary embodiments the turning points may be local maxima or minima, or both.
  • the turning points may be determined using the pixel count of the binary image described above.
  • smoothing of the pixel counts may be performed - for example by applying a rolling average - in order to assist with turning point identification.
  • the frame number for the image containing the turning point may be identified in order to determine a time at which the image was captured.
  • the frame may be identified, for example, by applying a turning point algorithm to the temperature data.
  • an indication of the depth of respiration may be determined at least in part by the differential between adjacent turning points of the temperature data. It should be appreciated that other means for determining depth of respiration may be used, for example based at least in part by the rate of change in temperature between adjacent turning points of the temperature data.
  • the depth of respiration is related to the volume of air displaced during respiration - as nostril temperature is generally proportional to airflow, the temperature differential or rate of change in that temperature between adjacent turning points may be used as an indication of depth of respiration.
  • a respiratory pattern may be determined based at least in part on the temperature data.
  • the respiratory pattern may relate to the regularity of respiration, or the proportion of depth or rate between the exhalation and inhalation phases.
  • temperature data being obtained using a binary filtering technique is not intended to be limiting, and that any suitable technique known to a person skilled in the art of image processing for obtaining this data (for example machine learning) may be used.
  • any suitable technique known to a person skilled in the art of image processing for obtaining this data for example machine learning
  • the principles used in densitometry may be used to differentiate pixel data from the images.
  • At least one thermographic camera may be configured to filter infrared wavelength ranges suitable for capturing images of a target gas of interest.
  • Such cameras, or filters for same, are known in the art of thermographic imaging.
  • C0 2 carbon dioxide
  • the target gas may be methane, production of which by animals is known to be indicative of metabolic function.
  • C0 2 content may be influenced by volume of gas exhaled by the animal, or concentration of the C0 2 within that volume of exhaled gas.
  • the region of interest of the image may include the expiratory plume of the animal. It should be appreciated that reference to an expiratory plume may include a plurality of plumes depending on the physiology of the animal and angle from which the images are captured.
  • the region of interest may be horizontally centred about the nostrils of the animal.
  • an exemplary region of interest may extend to either side of, and below, the nostrils of the animal in order to capture the substantive portion of the expiratory plume.
  • determining an indication of C0 2 may include comparison of data obtained from the images with previously established calibration or standard curves for the desired characteristic of the C0 2 content. Such calibration curves may be established by recording images of plumes of gas of known concentration and flow rate, and analysing the images to identify characteristics correlating with these factors.
  • determining an indication of C0 2 content may include: determining an area of the expiratory plume; and determining an indication C0 2 concentration based at least in part on intensity of pixels within the expiratory plume.
  • determination of C0 2 content may include applying a binary filter to the image such that each pixel is filtered relative to a predetermined threshold. The pixels above the threshold in the resulting image may be counted to provide a relative indication of C0 2 content.
  • determination of an indication of C0 2 content may be used to determine other respiratory characteristics of the animal such as respiration rate.
  • respiration rate For example, the time between occurrences of peak exhalation, or beginning of exhalation, may be used to determine respiration rate.
  • a condition of the animal may be determined based at least in part on comparison of the determined respiratory characteristic of the animal with previously established thresholds.
  • Reference to a condition may include, for example, a disease, infection, injury, thermal stress, psychological state (for example stress), metabolic operation, or any other factor which may influence animal health or production.
  • Exemplary conditions relating directly to the respiratory system of the animal which may be determined in exemplary embodiments described herein include: the diagnosis of respiratory acidosis and respiratory alkalosis in animals having ingested inappropriate feed quality or quantity, respiratory parasite infections, asymmetrical nasal turbinate or lung lobe infections, or other mechanical aberrations in the respiratory tract such as fractures of ribs. It should be appreciated that these are provided by way of example, and are not intended to be limiting.
  • an indication of metabolic rate of the animal may be determined based at least in part on the determined indication of gas content - in particular C0 2 content.
  • a method of determining an indication of metabolic rate of an animal including the steps of: obtaining a plurality of thermal infrared images of the animal using at least one thermographic camera, the area of the images capturing a region of interest relating to respiration of the animal; analysing each image to obtain data relating to carbon dioxide production of the animal; and determining the indication of metabolic rate of the animal based at least in part on the carbon dioxide production.
  • Carbon dioxide levels exhaled by an animal are indicative of the consumption of oxygen, and therefore metabolic heat production, as known in the art.
  • Metabolic rate i.e. the amount of energy used per unit of time impacts on the efficiency of feed utilization by the animal.
  • determination of the indication of metabolic rate may be based at least in part on methane production of the animal.
  • an indication of methane production may be determined from the images as referenced previously.
  • an indication of methane production may be obtained from a dedicated methane sensing device, as known in the art.
  • a method of determining the impact of feeding regime on an animal including the steps of: feeding the animal in accordance with a predetermined feeding regime; obtaining a plurality of thermal infrared images of the animal using at least one thermographic camera, the area of the images capturing a region of interest relating to respiration of the animal; analysing each image to obtain data relating to carbon dioxide production of the animal; and determining the impact of the predetermined feeding regime on the animal based at least in part on the carbon dioxide production.
  • references to the impact of an animal feeding regime should be understood to mean the influence of the feeding regime on performance metrics of the animal, for example production in terms of weight gain or milk production/quality, metabolic rate, or health. It should be appreciated that reference to a feeding regime may include the composition of the animal feed in addition to controls in terms of timing and quantity.
  • an indication of the change in metabolic rate of the animal may be determined.
  • a more efficient animal feed or regime should result in a comparative decrease in production of carbon dioxide from the animal.
  • metabolic rate may be indicative of health problems in an animal.
  • Milk fever Hypocalcaemiain, or parturient paresis
  • milk fever is a metabolic disease caused by a low blood calcium level (hypocalcaemia), early detection of which is highly important in order to intervene to prevent death. It is known to supplement diet of cows with magnesium for two to three weeks pre-calving in order to reduce the risk of milk fever. However, this regime does not build up a store of magnesium and it can be difficult to balance the optimal levels - potentially resulting in onset of milk fever. If detected early enough - e.g. through determination of changes in metabolic rate in accordance with embodiments of the present invention - cows can be treated, for example with a combined solution of calcium borogluconate, magnesium, phosphorus and dextrose.
  • Ketosis is an extension of a normal metabolic process that occurs in high producing dairy cows, and is also known as pregnancy toxaemia in sheep, resulting from a deficiency of glucose in the blood and body tissues. With early detection through determined variation in metabolic rate, the main treatment of a quick-acting glucose supplement may be used to counteract the negative effects.
  • rumen acidosis is a metabolic disease of cattle caused by insufficient fibre in the diet and occurs when the pH of the rumen falls to less than 5.5. Symptoms include reduced appetitie, lethargy and high respiration rate. In addition to respiratory characteristics such as rate and C0 2 production (and therefore metabolic rate), monitoring of an animal's feed intake could be used as a factor in determining a likelihood of rumen acidosis.
  • an indication of stress of the animal may be determined, based at least in part, on the determined characteristic of respiration.
  • the indication of stress may be determined based on comparison with determined respiration rate with a baseline respiration rate.
  • Terms such as fear, stress, anxiety, pain or a combination thereof may be used to describe the responses of animals to various husbandry practices (e.g. transport, painful procedures, weaning) and environmental conditions (e.g. thermal stress, housing conditions) that they are routinely exposed to as part of normal management on a commercial operation.
  • Stress has been defined as "a state that occurs when an animal is required to make abnormal or extreme adjustments in its physiology or behaviour in order to cope with adverse aspects of its environment and management" (Fraser et al., 1975). This may include, for example, any environmental factor that causes an increase in the endocrine steroid Cortisol.
  • Pain has been defined as "an aversive sensory and emotional experience representing an awareness by the animal of damage or threat to the integrity of its tissues; it changes the animal's physiology and behaviour to reduce or avoid damage, to reduce the likelihood of recurrence and to promote recovery” (Molony and Kent, 1997).
  • stress thresholds may be established by measuring normal Cortisol levels for an animal (for example in saliva over 3 to 5 days) to establish a normal range. Characteristics of respiration may be monitored during the same period. Subsequent variations in Cortisol level, with associated changes in respiration characteristics, outside that range may be considered to be indicative of undesirable degrees of stress.
  • determination of an indication of stress in an animal may be based at least in part on at least one contextual factor.
  • the contextual factor may be the location of the thermographic camera or the activity associated with that location - such as milking in a milking stall, a holding pen into which the animal has been drafted, or a birthing pen.
  • the contextual factor may also include scheduled animal husbandry activities.
  • the contextual factor may be a determined condition of the animal - whether currently determined or historical. For example, where the animal is known or determined to be suffering from an ailment, respiratory characteristics may be less indicative of stress from another source.
  • an indication of parturition of the animal may be determined, based at least in part, on the determined characteristic of respiration of the animal. It should be appreciated that the indication of parturition may relate to the eminent onset of parturition, in addition to parturition having begun.
  • the difficulty of birthing influences a number of factors impacting on the likelihood of survival and health of the offspring - for example, the risk for infectious disease, difficulties in maintaining body temperature, and decreases in absorption of antibodies. It may also affect the dam - including the potential for diminished productivity and conception rate.
  • Identification or prediction of parturition enables observation of periparturient dairy cattle and may assist in reducing dystocia related injuries leading to animal pain and suffering.
  • determination of an indication of parturition may include reference to a historical record of the animal, including a projected date for the onset of parturition.
  • the likelihood of a change in respiration rate indicating parturition may be influenced by proximity to the projected date.
  • exemplary embodiments may have particular application to determining the respiratory characteristics of production animals - for example milking animals or meat production animals.
  • milking animals or meat production animals When being milked or fed, such animals are generally confined to a stall facing a particular direction.
  • Other useful locations may include standalone drinking or feeding stations (for example calf feeders), or birthing pens for animals nearing parturition. This provides an opportunity to mount the thermographic camera(s) in a position which is likely to capture the desired region of interest.
  • animals cows in particular
  • thermographic camera is intended to capture images of the animal's expiratory plumes, it is envisaged that the animal may be positioned in a sheltered area to reduce variation introduced by environmental airflow.
  • the recording area may be such that the region of interest can exclude other animals or other sources of thermal variation in order to improve accuracy of image processing.
  • the positioning of the apparatus is coupled (directly or indirectly) with an identification (ID) reader so that data may be recorded against the animal's ID.
  • ID identification
  • RFID radio frequency identification
  • RFID radio frequency identification
  • Such an arrangement may be economically useful, as a single installation may be used to monitor a series of animals utilising the feeding or milking station.
  • exemplary embodiments may have particular application to automated milking robots, where animals may not be directly observed by farm personnel as regularly as in for labour intensive milking operations and otherwise previously undetectable conditions may be missed.
  • the apparatus forms part of a wider system receiving inputs from other sensors obtaining physiological, behavioural, and production indicators.
  • the system may include sensors for measuring milk yield, fat content, protein content, feed efficiency, somatic cell count, lactose, conductivity, animal weight, feeding patterns, and so on.
  • operation of at least one animal husbandry device may be controlled based at least in part on the determined condition of the animal.
  • the animal husbandry device may include at least one of: a milking device, a feeding device, a drafting device, and a treatment device.
  • determined respiratory characteristics may be used to adjust milking system characteristics - for example pulsation rate, vacuum level, or cup disconnection.
  • Animal management systems including automated drafting are well known in the art. These may be used to release an animal exhibiting excessively high stress levels - whether to pasture or a holding area. Further, such drafting systems may be used to direct an animal to a holding area for manual examination and/or treatment if symptoms indicate that this is required.
  • reference to exemplary embodiments of the invention being used with milking animals is not intended to be limiting, and may be applied to other production animals - for example beef, pigs, cattle, poultry, and sheep.
  • respiration rate may be used to monitor stress levels of a beef animal at a feeding or weighing station, and move the animal to a pen for monitoring should stress levels become too high.
  • exemplary embodiments may be used in relation to other non-human animals - for example companion animals, high performance animals such as horses, exotic animals, or humans in some instances.
  • non-human animals for example companion animals, high performance animals such as horses, exotic animals, or humans in some instances.
  • performance horses particularly "hot-blooded" breeds - are known for susceptibility to stress and associated health complications which may not be observed by personnel until irreparable damage has occurred.
  • determination of respiration characteristics may be used during the transportation of animals. It is envisaged that a thermographic camera could be positioned within the transportation vessel - for example a towed trailer, or stall in an airplane or ship - and analysis of determined respiration characteristics used to respond to potentially harmful conditions of the animal, whether that be intervention by personnel, or automated provision of feed/water/medication.
  • Respiratory characteristics may be influenced by many factors, including body size, age, exercise, excitement, environmental temperature, atmospheric conditions, pregnancy, and fullness of the digestive tract.
  • determining a condition of the animal based at least in part on a determined respiratory characteristic may include accessing historical data associated with the animal, or group of animals to which it belongs.
  • the processor may be configured to update a historical record associated with the identification of the animal with temperature and respiratory characteristic data.
  • An individual history may allow for more accurate modelling of the animal's respiratory characteristics. As with any organic entity, variations between individuals will lead to inaccuracies in models based on collective data.
  • the individual histories may store a wide range of data - not restricted to temperature alone.
  • the historical records may include treatment or diagnosis history, or details of pregnancy.
  • determining respiratory characteristics may involve comparing data points or trends across a group of animals - such as a herd, or at least animals processed within a particular timeframe.
  • thermography Bovine Viral Disease
  • steps of a method, process, or algorithm described in connection with the present invention may be embodied directly in hardware, in a software module executed by the processor, or in a combination of the two.
  • the various steps or acts in a method or process may be performed in the order shown, or may be performed in another order. Additionally, one or more process or method steps may be omitted or one or more process or method steps may be added to the methods and processes. An additional step, block, or action may be added in the beginning, end, or intervening existing elements of the processes.
  • the processor may include one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices or controllers (PLDs, PLCs), field programmable gate arrays (FPGAs), computers, lap tops, controllers, micro-controllers, microprocessors, electronic devices, other electronic units (whether analogue or digital) designed to perform the functions described herein, or a combination thereof.
  • ASICs application specific integrated circuits
  • DSPs digital signal processors
  • DSPDs digital signal processing devices
  • PLDs programmable logic devices or controllers
  • FPGAs field programmable gate arrays
  • computers lap tops, controllers, micro-controllers, microprocessors, electronic devices, other electronic units (whether analogue or digital) designed to perform the functions described herein, or a combination thereof.
  • FIG. 1 is a block diagram illustrating an exemplary system in which exemplary embodiments of the invention may be implemented;
  • FIG. 2 is a flow diagram of an exemplary method of determining at least one respiration characteristic of an animal;
  • FIG. 3 is an exemplary thermographic image of an animal
  • FIG. 4A-C are cropped views of exemplary regions of interest in exemplary thermographic images, before and after application of a binary filter
  • FIG. 5 is a line graph plotting pixel count against frame number
  • FIG. 6 is a flow diagram of an exemplary method of determining at least one respiration characteristic of an animal - in particular target gas content within gas expired by the animal;
  • FIG. 7 is another exemplary thermographic image of an animal;
  • FIG. 8A & B are cropped views of exemplary regions of interest in exemplary thermographic images, before and after application of a binary filter
  • FIG. 9 is a flow diagram of an exemplary method of determining the impact of feeding regime on an animal.
  • FIG. 10 is a flow diagram of an exemplary method of determining an indication of parturition of an animal, and
  • FIG. 11 is a bar chart illustrating change in respiration rate captured in response to a startle reflex.
  • FIG. 1 illustrates an exemplary milking system (generally indicated by arrow 100) in which embodiments of the present invention operate.
  • the milking system 100 includes a stall 102, in which a cow 104 is held during milking.
  • the stall 102 includes a milking apparatus 106 for extracting milk from the cow.
  • the milking apparatus 106 may be, for example, a robotic arm of an automated milking robot, as known in the art. However, it should be appreciated that this is not intended to be limiting, and the stall 102 may use more conventional milking setups requiring manual application.
  • Productivity sensors 108 associated with the milking apparatus 106 to capture productivity indicators associated with the cow 104 may include, for example, a conductivity sensor, yield sensor, fat sensor, protein sensor, lactose sensor, conductivity sensor, or somatic cell count sensor.
  • the stall 102 includes a weighing device - for example load cell 110 - to measure the cow's weight.
  • a radio frequency identification (RFID) reader 1 12 is configured to read an identification tag 114 associated with the cow 104. It should be appreciated that while the tag 114 is illustrated in the form of a leg band, the tag may take any other suitable form known in the art - for example an ear tag.
  • the system 100 also includes a network 1 16 over which various communication links between devices within the system 100 may be established. While the network 116 is illustrated as a single component, it should be appreciated that it may be composed of a number of sub-networks, potentially operating using distinct technologies - e.g. wired or wireless, fibre optic or radio.
  • the system 100 includes a central processor 118.
  • the central processor may be, for example, a dedicated device for the system 100 or for a collection of milking stalls in a milking installation, or a server remotely located from the milking installation.
  • the processor 118 is configured to receive the cow's ID, and weight and productivity indicators from the load cell 110 and sensors 108. This data, along with historical records associated with the cow 104 (and herd to which it belongs), may be stored in memory 120.
  • the central processor may communicate over the network 1 16 with user devices such as hand held communication device 122 and user workstation 124.
  • At least one thermographic camera 126 is mounted at an end of the stall 102, directed to the face of the cow 104. Thermal infrared images captured by the camera 126 are transmitted to the central processor 1 18 for processing, as will be described further below.
  • the camera 126 may be configured to capture a series of images at predetermined intervals, but it is envisaged that the camera 126 will supply a video to the processor 1 18 for sampling at the desired rate.
  • FIG. 2 describes an exemplary method 200 of determining at least one respiration characteristic of an animal - for example cow 104 - with reference to FIG. 1 , and FIG. 3 to 5. It should be appreciated that while the method is described with reference to specific techniques for extracting data from the images, this is not intended to be limiting.
  • step 202 video is received from the thermographic camera 126.
  • step 204 the region of interest within the image area is established and the images cropped to the region of interest.
  • the region of interest includes at least a portion of at least one of the cow's nostrils.
  • feature detection may be used to identify one or more of the nostrils of the cow, and an area surrounding the nostrils designated as the region of interest for further image processing.
  • FIG. 3 shows the full field of view of a thermographic image 300 captured using a FLIR infrared thermal camera (FLIR Systems Inc., USA). It may be seen that the nostrils 302a and 302b of the animal are clearly distinguishable within the image, and it would be appreciated by a person skilled in the art of image processing that feature detection (for example edge-based shape detection) may be performed to identify the nostrils 302a and 302b and therefore designate a region of interest 304.
  • a binary filter is applied to each image such that each pixel is filtered relative to a temperature threshold.
  • the temperature threshold established by sampling the temperature between the nostrils of the animal at a peak of inspiration - i.e. when the animal has fully inhaled and the mean nostril temperature is generally at its coldest.
  • unfiltered image 400 of FIG. 4A in which the sampling region 402 is generally indicated.
  • the light pixels indicate pixels in which temperature was above the threshold.
  • inspiration brings in external air that cools the nostrils while expiration brings warm air from the body core to the nostrils.
  • Mean nostril temperature is generally proportional to air flow, and as such the coldest temperature may generally be identified as the peak of inspiration, and the warmest temperature as the nadir of expiration.
  • step 206 the pixels above the threshold in each binary image are counted to provide a relative indication of temperature.
  • FIG. 4B shows an unfiltered image 406 and binary image 408 at a mid-point between the peak of inspiration (as illustrated in FIG. 4A), and the nadir of expiration.
  • FIG. 4C shows corresponding unfiltered image 410 and binary image 412 at the nadir of expiration. It may be observed visually that there is a distinct progression in terms of area of the images in which the pixels are above the threshold.
  • a smoothing function such as a rolling average, may be applied to the pixel count in order to assist in subsequent analysis.
  • FIG. 5 is a line graph plotting pixel count against the frame number of the respective images. It may be observed that there are a series of turning points in the form of maxima and minima, which correspond with nadirs of expiration and peaks of inspiration respectively.
  • respiration characteristics may be determined, for example respiration rate (by counting turning points within a time period), depth of breath (by determining the pixel count differential between adjacent local maxima and minima), and breathing patterns (both in terms of depth and regularity).
  • FIG. 6 describes another exemplary method 600 of determining at least one respiration characteristic of an animal - in particular target gas content within gas expired by the animal - with reference to FIG. 1 , and FIG. 7.
  • FIG. 8A and FIG. 8B It should be appreciated that while the method is described with reference to specific techniques for extracting data from the images, this is not intended to be limiting.
  • video is received from the thermographic camera 126.
  • the camera is tuned to the infrared absorption band of a target gas produced by the animal.
  • image 700 of FIG. 7 was captured using the FLIR SC8303 camera (FLIR Systems Inc., USA) fitted with a filter tuned to the infrared absorption band of carbon dioxide (C0 2 ).
  • the region of interest 702 includes at least the expiratory plumes 704 of the cow.
  • the region of interest 702 may extend in the order of 0.5 m to 1.0 m to either side of, and below, the nostrils of the animal in order to capture the substantive portion of the expiratory plumes 704.
  • the region of interest 702 may extend above the nostrils of the animal to capture the face of the animal to assist in further image processing.
  • step 606 the gas content of the exhaust plumes 704 is determined.
  • machine learning algorithms may analyse the pixels within the expiratory plumes to derive volume and/or concentration based on pixel count and intensity. Information regarding the gas content may be derived through comparison of this data with standard curves developed for the target gas using known standards
  • FIG. 8A illustrates an image 800 of a portion of a region of interest capturing an expiratory plume.
  • Image 800 is taken from a frame in which the animal has not begun exhalation (i.e. there is no expiratory plume).
  • a binary filter is applied to image 800 to produce binary image 802.
  • image 804 shows the portion of the region of interest during expiration. Comparing the binary image 806 of FIG. 8B with binary image 802 of FIG. 8A, it may be observed that the C0 2 within the plume has been distinguished from the remainder of the image in a manner which may be quantified (i.e. through a count of the bright pixels).
  • method 600 may also be used to determine respiration rate of the animal based on a count of occurrences of peak exhalation or beginning of exhalation.
  • numerous algorithms may be implemented to determine implications for conditions of the animal. This may include comparison of the measurements against predetermined thresholds, where particular combinations or variances are indicative of a particular condition. In doing so, the processor 118 of FIG. 1 may access historical data associated with the cow 104. The current measurements may be compared with historical data, for example a rolling average of that particular measurement, to determine the extent of deviation for that animal.
  • the thresholds utilised by the diagnostic algorithms may therefore not be a hard level, but relative to that particular animal.
  • the processor 118 may access historical data associated with the herd to which the cow 104 belongs to assess whether any changes in the measurements are consistent across the herd - potentially indicating changes in environmental conditions.
  • the processor 118 may control the milking apparatus 106 according to the determined condition of the cow 104 - for example the length or intensity of one or more of the stages of the pulsation cycle.
  • the system 100 includes a display 128 associated with the stall 102, and an alarm in the form of a siren/strobe 130. These may be used to notify farm workers in the vicinity of the status of the cow 130, particularly where manual intervention or action is required. Further, this data or alarm condition may be transmitted to the hand-held device 122 or workstation 124.
  • the processor 118 may also control operation of an animal husbandry device 132 in response to a determined condition, for example:
  • a feeding device to release feedstuff to the cow 104, for example in response to determination of an elevated level of stress
  • a temperature control device for example a sprinkler system for cooling an animal determined to be suffering from heat stress in order to decrease its body temperature
  • a treatment device to deliver treatment to the cow 104 • a treatment device to deliver treatment to the cow 104.
  • elements of the system 100 such as the camera, RFID equipment, load cell and other non-milking related equipment may be located at a feed station, or crush, or birthing pen.
  • FIG. 9 illustrates a method 900 of determining the impact of feeding regime on an animal.
  • the animal is fed in accordance with a predetermined feeding regime - whether that be in terms of the feedstuff itself, and/or controls around timing and/or quantity.
  • method 700 is performed to determine an indication of C0 2 production by the animal.
  • the metabolic rate of the animal may be determined from the indication of indication of C0 2 production.
  • this metabolic rate may be compared with that previously established for the animal on another feeding regime in order to assess the impact of the new feeding regime.
  • FIG. 10 illustrates a method 1000 of determining an indication of parturition of an animal.
  • method 200 is performed to determine respiratory characteristics - for example respiration rate - of the animal.
  • the determined respiration characteristics may be compared with baseline values for the animal.
  • onset of partition may be determined. Determination of parturition may include reference to the animal's gestation time - both current and historically in relation to previous pregnancies.
  • an alert is issued to notify farm personnel of the onset of parturition.
  • a FLIR T450SC camera FLIR Systems Inc., USA was used to monitor the facial areas of Cow A and Cow B. Working indoors out of sunlight, images were captured at a distance of approximately 1 metre. Atmospheric temperature was measured at 20.5°C, with relative humidity at 66.5%. A scan of nine animals on the day demonstrated an average eye temperature of between 32 to 36°C - a normal range.
  • Cow A had not been observed as experiencing noticeable health or metabolic problems, while Cow B had experienced metabolic problems with regard to reduced milk production over the previous two weeks.
  • Table 1 outlines the temperature data obtained from the images captured. All data is given in degrees Celsius (°C).
  • radiated temperatures for Cow B were generally cooler than for Cow A.
  • the nose temperatures for Cow B were up to 5°C cooler on average.
  • the respiration rate of Cow B was also determined to be approximately half that of Cow A, and the depth of breathing was also significantly reduced. This indicates the viability of using thermographic images to differentiate a metabolically or respiratory challenged animal from other animals.
  • EXAMPLE TWO Relationship between respiration rate and induced stress.
  • a FLIR S60 camera FLIR Systems Inc., USA was used to monitor the facial areas of Cow 1 and Cow 2. Baseline data was recorded while the cows were held in a race, prior to being moved into a crush and restrained in a head bale.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Medical Informatics (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Pulmonology (AREA)
  • Emergency Medicine (AREA)
  • Obesity (AREA)
  • Physiology (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

La présente invention concerne un procédé et un appareil destinés à déterminer au moins une caractéristique de respiration d'un animal non humain. Une pluralité d'images infrarouges thermiques de l'animal sont obtenues à l'aide d'au moins une caméra thermographique. La zone des images capturent une région à étudier se rapportant à la respiration de l'animal. Chaque image est analysée afin d'obtenir des données se rapportant à la respiration de l'animal. Au moins une caractéristique de respiration de l'animal est déterminée sur la base, au moins en partie, des données obtenues à partir des images.
PCT/NZ2014/000185 2013-09-02 2014-09-02 Procédé et appareil de détermination des caractéristiques de respiration d'un animal WO2015030611A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
NZ61497413 2013-09-02
NZ614974 2013-09-02

Publications (1)

Publication Number Publication Date
WO2015030611A1 true WO2015030611A1 (fr) 2015-03-05

Family

ID=52587020

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/NZ2014/000185 WO2015030611A1 (fr) 2013-09-02 2014-09-02 Procédé et appareil de détermination des caractéristiques de respiration d'un animal

Country Status (1)

Country Link
WO (1) WO2015030611A1 (fr)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017062125A (ja) * 2015-09-24 2017-03-30 株式会社Ｃｓソリューション 対象動物の体温を非接触で測定するための体温測定システム
WO2020025320A1 (fr) 2018-07-31 2020-02-06 Signify Holding B.V. Dispositif de commande permettant de détecter des animaux avec des états physiologiques
CN111387957A (zh) * 2020-03-13 2020-07-10 智方达(天津)科技有限公司 一种非接触式的体温与呼吸率联合检测方法
CN111707811A (zh) * 2020-06-29 2020-09-25 重庆市畜牧科学院 基于无菌环境下的羔羊呼吸气体浓度检测装置及方法
WO2020231250A1 (fr) * 2019-05-10 2020-11-19 Lely Patent N.V. Système de surveillance d'animaux ruminants
CN113576451A (zh) * 2021-07-30 2021-11-02 深圳市商汤科技有限公司 呼吸率检测方法、装置、存储介质及电子设备
CN113576452A (zh) * 2021-07-30 2021-11-02 深圳市商汤科技有限公司 基于热成像的呼吸率检测方法、装置及电子设备
EP3967215A1 (fr) 2020-09-15 2022-03-16 Alunos AG Dispositif de mesure et procédé d'évaluation de données vitales d'un sujet humain ou animal
RU2774146C1 (ru) * 2021-08-31 2022-06-15 Федеральное государственное бюджетное научное учреждение «Федеральный научный агроинженерный центр ВИМ» (ФГБНУ ФНАЦ ВИМ) Способ и устройство для бесконтактной диагностики заболевания конечностей крупного рогатого скота на ранней стадии
WO2023005469A1 (fr) * 2021-07-30 2023-02-02 上海商汤智能科技有限公司 Procédé et appareil pour déterminer une région de détection de respiration, support de stockage et dispositif électronique

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020108576A1 (en) * 2001-02-13 2002-08-15 Lely Enterprises Ag Device for and a method of milking an animal, a device for monitoring an animal
US20070093965A1 (en) * 2003-05-30 2007-04-26 Harrison Harry J Use of infrared thermography in live animals to predict growth efficiency
US20120289850A1 (en) * 2011-05-09 2012-11-15 Xerox Corporation Monitoring respiration with a thermal imaging system
US20130079658A1 (en) * 2011-09-27 2013-03-28 Xerox Corporation Minimally invasive image-based determination of carbon dioxide (co2) concentration in exhaled breath
US20130324875A1 (en) * 2012-06-01 2013-12-05 Xerox Corporation Processing a video for respiration rate estimation

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020108576A1 (en) * 2001-02-13 2002-08-15 Lely Enterprises Ag Device for and a method of milking an animal, a device for monitoring an animal
US20070093965A1 (en) * 2003-05-30 2007-04-26 Harrison Harry J Use of infrared thermography in live animals to predict growth efficiency
US20120289850A1 (en) * 2011-05-09 2012-11-15 Xerox Corporation Monitoring respiration with a thermal imaging system
US20130079658A1 (en) * 2011-09-27 2013-03-28 Xerox Corporation Minimally invasive image-based determination of carbon dioxide (co2) concentration in exhaled breath
US20130324875A1 (en) * 2012-06-01 2013-12-05 Xerox Corporation Processing a video for respiration rate estimation

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2017062125A (ja) * 2015-09-24 2017-03-30 株式会社Ｃｓソリューション 対象動物の体温を非接触で測定するための体温測定システム
WO2020025320A1 (fr) 2018-07-31 2020-02-06 Signify Holding B.V. Dispositif de commande permettant de détecter des animaux avec des états physiologiques
WO2020231250A1 (fr) * 2019-05-10 2020-11-19 Lely Patent N.V. Système de surveillance d'animaux ruminants
US11771062B2 (en) 2019-05-10 2023-10-03 Lely Patent N.V. Ruminant animal monitoring system
CN113727646A (zh) * 2019-05-10 2021-11-30 莱利专利股份有限公司 反刍动物监测系统
CN111387957B (zh) * 2020-03-13 2023-03-24 智方达(天津)科技有限公司 一种非接触式的体温与呼吸率联合检测方法
CN111387957A (zh) * 2020-03-13 2020-07-10 智方达(天津)科技有限公司 一种非接触式的体温与呼吸率联合检测方法
CN111707811A (zh) * 2020-06-29 2020-09-25 重庆市畜牧科学院 基于无菌环境下的羔羊呼吸气体浓度检测装置及方法
EP3967215A1 (fr) 2020-09-15 2022-03-16 Alunos AG Dispositif de mesure et procédé d'évaluation de données vitales d'un sujet humain ou animal
CN113576452A (zh) * 2021-07-30 2021-11-02 深圳市商汤科技有限公司 基于热成像的呼吸率检测方法、装置及电子设备
WO2023005403A1 (fr) * 2021-07-30 2023-02-02 上海商汤智能科技有限公司 Procédé et appareil de détection de fréquence respiratoire, et support d'enregistrement et dispositif électronique
WO2023005469A1 (fr) * 2021-07-30 2023-02-02 上海商汤智能科技有限公司 Procédé et appareil pour déterminer une région de détection de respiration, support de stockage et dispositif électronique
CN113576451A (zh) * 2021-07-30 2021-11-02 深圳市商汤科技有限公司 呼吸率检测方法、装置、存储介质及电子设备
RU2774146C1 (ru) * 2021-08-31 2022-06-15 Федеральное государственное бюджетное научное учреждение «Федеральный научный агроинженерный центр ВИМ» (ФГБНУ ФНАЦ ВИМ) Способ и устройство для бесконтактной диагностики заболевания конечностей крупного рогатого скота на ранней стадии

Similar Documents

Publication Publication Date Title
WO2015030611A1 (fr) Procédé et appareil de détermination des caractéristiques de respiration d'un animal
Costa et al. Symposium review: Precision technologies for dairy calves and management applications
DK2690948T3 (en) DEVICE AND PROCEDURE FOR USING INFRARED THERMOGRAPHY AND BEHAVIOR INFORMATION FOR IDENTIFICATION OF BIOLOGICAL IMPORTANT CONDITIONS OF ANIMALS
US8789494B2 (en) Detection apparatus for the monitoring of milking animals
Stewart et al. The use of infrared thermography and accelerometers for remote monitoring of dairy cow health and welfare
Hoffmann et al. Animal-related, non-invasive indicators for determining heat stress in dairy cows
Wijffels et al. Methods to quantify heat stress in ruminants: Current status and future prospects
US20150359200A1 (en) Infrared thermography and behaviour information for identification of biologically important states in animals
US9961883B2 (en) Rapid and automatic determination of metabolic efficiency in livestock
AU2010335065B9 (en) Detection method
Hoffmann et al. First investigations to refine video-based IR thermography as a non-invasive tool to monitor the body temperature of calves
Tzanidakis et al. Precision Livestock Farming (PLF) systems: Improving sustainability and efficiency of animal production
Süli et al. Body temperature and motion: evaluation of an online monitoring system in pigs challenged with porcine reproductive & respiratory syndrome virus
Hoffmann et al. 1. Monitoring the body temperature of cows and calves with a video-based infrared thermography camera
Cook Review on the use of infrared thermography to monitor the health of intensively housed livestock
CA2854344A1 (fr) Informations de thermographie infrarouge et de comportement pour determiner des etats biologiquement importants chez des animaux
NZ586888A (en) Apparatus for the detection of health conditions and oestrus in milking animals using temperature sensors
Yuan et al. Stress-free detection technologies for pig growth based on welfare farming: A review
Harty et al. Using big data and advanced analytics to optimise health and fertility
Mee State-of-the-art sensors to monitor/manage dairy calf birth and calf health
Puig et al. Technological Tools for the Early Detection of Bovine Respiratory Disease in Farms. Animals 2022, 12, 2623
Puig Escrig et al. Technological Tools for the Early Detection of Bovine Respiratory Disease in Farms
Singh PRECISION LIVESTOCK PRODUCTION: CLIMATE CHANGE PERSPECTIVE
Dowling et al. Infrared thermography for animal health and welfare monitoring: where to from here?
Jongman et al. DEVELOPING REMOTE MONITORING METHODS FOR EARLY DETECTION OF RESPIRATORY DISEASE IN PIGS

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14839577

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14839577

Country of ref document: EP

Kind code of ref document: A1