Automated Observations of Dogs' Resting Behaviour Patterns Using Artificial Intelligence and Their Similarity to Behavioural Observations.

Although direct behavioural observations are widely used, they are time-consuming, prone to error, require knowledge of the observed species, and depend on intra/inter-observer consistency. As a result, they pose challenges to the reliability and repeatability of studies. Automated video analysis is becoming popular for behavioural observations. Sleep is a biological metric that has the potential to become a reliable broad-spectrum metric that can indicate the quality of life and understanding sleep patterns can contribute to identifying and addressing potential welfare concerns, such as stress, discomfort, or health issues, thus promoting the overall welfare of animals; however, due to the laborious process of quantifying sleep patterns, it has been overlooked in animal welfare research. This study presents a system comparing convolutional neural networks (CNNs) with direct behavioural observation methods for the same data to detect and quantify dogs' sleeping patterns. A total of 13,688 videos were used to develop and train the model to quantify sleep duration and sleep fragmentation in dogs. To evaluate its similarity to the direct behavioural observations made by a single human observer, 6000 previously unseen frames were used. The system successfully classified 5430 frames, scoring a similarity rate of 89% when compared to the manually recorded observations. There was no significant difference in the percentage of time observed between the system and the human observer (p > 0.05). However, a significant difference was found in total sleep time recorded, where the automated system captured more hours than the observer (p < 0.05). This highlights the potential of using a CNN-based system to study animal welfare and behaviour research.

Environmental enrichments and data-driven welfare indicators for sheltered dogs using telemetric physiological measures and signal processing.

Shelters are stressful environments for domestic dogs which are known to negatively impact their welfare. The introduction of outside stimuli for dogs in this environment can improve their welfare and life conditions. However, our current understanding of the influence of different stimuli on shelter dogs' welfare is limited and the data is still insufficient to draw conclusions. In this study, we collected 28 days (four weeks) of telemetry data from eight male dogs housed in an Italian shelter for a long period of time. During this period, three types of enrichment were introduced into the dogs' pens for one week each: entertaining objects, intraspecific, and interspecific social enrichment, by means of the presence of female conspecifics and the presence of a human. To quantify their impact, we introduce novel metrics as indicators of sheltered dogs' welfare based on telemetry data: the variation of heart rate, muscle activity, and body temperature from an average baseline day, quality of sleep, and the regularity for cyclicity of the aforementioned parameters, based on the day-night cycle. Using these metrics, we show that while all three stimuli statistically improve the dogs' welfare, the variance between individual dogs is large. Moreover, our findings indicate that the presence of female conspecific is the best stimulus among the three explored options which improves both the quality of sleep and the parameters' cyclicity. Our results are consistent with previous research findings while providing novel data-driven welfare indicators that promote objectivity. Thus, this research provides some useful guidelines for managing shelters and improving dogs' welfare.

BovineTalk: machine learning for vocalization analysis of dairy cattle under the negative affective state of isolation.

There is a critical need to develop and validate non-invasive animal-based indicators of affective states in livestock species, in order to integrate them into on-farm assessment protocols, potentially via the use of precision livestock farming (PLF) tools. One such promising approach is the use of vocal indicators. The acoustic structure of vocalizations and their functions were extensively studied in important livestock species, such as pigs, horses, poultry, and goats, yet cattle remain understudied in this context to date. Cows were shown to produce two types of vocalizations: low-frequency calls (LF), produced with the mouth closed, or partially closed, for close distance contacts, and open mouth emitted high-frequency calls (HF), produced for long-distance communication, with the latter considered to be largely associated with negative affective states. Moreover, cattle vocalizations were shown to contain information on individuality across a wide range of contexts, both negative and positive. Nowadays, dairy cows are facing a series of negative challenges and stressors in a typical production cycle, making vocalizations during negative affective states of special interest for research. One contribution of this study is providing the largest to date pre-processed (clean from noises) dataset of lactating adult multiparous dairy cows during negative affective states induced by visual isolation challenges. Here, we present two computational frameworks-deep learning based and explainable machine learning based, to classify high and low-frequency cattle calls and individual cow voice recognition. Our models in these two frameworks reached 87.2 and 89.4% accuracy for LF and HF classification, with 68.9 and 72.5% accuracy rates for the cow individual identification, respectively.

Digitally-enhanced dog behavioral testing.

Behavioral traits in dogs are assessed for a wide range of purposes such as determining selection for breeding, chance of being adopted or prediction of working aptitude. Most methods for assessing behavioral traits are questionnaire or observation-based, requiring significant amounts of time, effort and expertise. In addition, these methods might be also susceptible to subjectivity and bias, negatively impacting their reliability. In this study, we proposed an automated computational approach that may provide a more objective, robust and resource-efficient alternative to current solutions. Using part of a 'Stranger Test' protocol, we tested n = 53 dogs for their response to the presence and neutral actions of a stranger. Dog coping styles were scored by three dog behavior experts. Moreover, data were collected from their owners/trainers using the Canine Behavioral Assessment and Research Questionnaire (C-BARQ). An unsupervised clustering of the dogs' trajectories revealed two main clusters showing a significant difference in the stranger-directed fear C-BARQ category, as well as a good separation between (sufficiently) relaxed dogs and dogs with excessive behaviors towards strangers based on expert scoring. Based on the clustering, we obtained a machine learning classifier for expert scoring of coping styles towards strangers, which reached an accuracy of 78%. We also obtained a regression model predicting C-BARQ scores with varying performance, the best being Owner-Directed Aggression (with a mean average error of 0.108) and Excitability (with a mean square error of 0.032). This case study demonstrates a novel paradigm of 'machine-based' dog behavioral assessment, highlighting the value and great promise of AI in this context.

Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data.

The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

BrachySound: machine learning based assessment of respiratory sounds in dogs.

The early and accurate diagnosis of brachycephalic obstructive airway syndrome (BOAS) in dogs is pivotal for effective treatment and enhanced canine well-being. Owners often do underestimate the severity of BOAS in their dogs. In addition, traditional diagnostic methods, which include pharyngolaryngeal auscultation, are often compromised by subjectivity, are time-intensive and depend on the veterinary surgeon's experience. Hence, new fast, reliable assessment methods for BOAS are required. The aim of the current study was to use machine learning techniques to bridge this scientific gap. In this study, machine learning models were employed to objectively analyze 366 audio samples from 69 Pugs and 79 other brachycephalic breeds, recorded with an electronic stethoscope during a 15-min standardized exercise test. In classifying the BOAS test results as to whether the dog is affected or not, our models achieved a peak accuracy of 0.85, using subsets from the Pugs dataset. For predictions of the BOAS results from recordings at rest in Pugs and various brachycephalic breeds, accuracies of 0.68 and 0.65 were observed, respectively. Notably, the detection of laryngeal sounds achieved an F1 score of 0.80. These results highlight the potential of machine learning models to significantly streamline the examination process, offering a more objective assessment than traditional methods. This research indicates a turning point towards a data-driven, objective, and efficient approach in canine health assessment, fostering standardized and objective BOAS diagnostics.

Deep learning for video-based automated pain recognition in rabbits.

Despite the wide range of uses of rabbits (Oryctolagus cuniculus) as experimental models for pain, as well as their increasing popularity as pets, pain assessment in rabbits is understudied. This study is the first to address automated detection of acute postoperative pain in rabbits. Using a dataset of video footage of n = 28 rabbits before (no pain) and after surgery (pain), we present an AI model for pain recognition using both the facial area and the body posture and reaching accuracy of above 87%. We apply a combination of 1 sec interval sampling with the Grayscale Short-Term stacking (GrayST) to incorporate temporal information for video classification at frame level and a frame selection technique to better exploit the availability of video data.

Explainable automated pain recognition in cats.

Manual tools for pain assessment from facial expressions have been suggested and validated for several animal species. However, facial expression analysis performed by humans is prone to subjectivity and bias, and in many cases also requires special expertise and training. This has led to an increasing body of work on automated pain recognition, which has been addressed for several species, including cats. Even for experts, cats are a notoriously challenging species for pain assessment. A previous study compared two approaches to automated 'pain'/'no pain' classification from cat facial images: a deep learning approach, and an approach based on manually annotated geometric landmarks, reaching comparable accuracy results. However, the study included a very homogeneous dataset of cats and thus further research to study generalizability of pain recognition to more realistic settings is required. This study addresses the question of whether AI models can classify 'pain'/'no pain' in cats in a more realistic (multi-breed, multi-sex) setting using a more heterogeneous and thus potentially 'noisy' dataset of 84 client-owned cats. Cats were a convenience sample presented to the Department of Small Animal Medicine and Surgery of the University of Veterinary Medicine Hannover and included individuals of different breeds, ages, sex, and with varying medical conditions/medical histories. Cats were scored by veterinary experts using the Glasgow composite measure pain scale in combination with the well-documented and comprehensive clinical history of those patients; the scoring was then used for training AI models using two different approaches. We show that in this context the landmark-based approach performs better, reaching accuracy above 77% in pain detection as opposed to only above 65% reached by the deep learning approach. Furthermore, we investigated the explainability of such machine recognition in terms of identifying facial features that are important for the machine, revealing that the region of nose and mouth seems more important for machine pain classification, while the region of ears is less important, with these findings being consistent across the models and techniques studied here.

Explainable automated recognition of emotional states from canine facial expressions: the case of positive anticipation and frustration.

In animal research, automation of affective states recognition has so far mainly addressed pain in a few species. Emotional states remain uncharted territories, especially in dogs, due to the complexity of their facial morphology and expressions. This study contributes to fill this gap in two aspects. First, it is the first to address dog emotional states using a dataset obtained in a controlled experimental setting, including videos from (n = 29) Labrador Retrievers assumed to be in two experimentally induced emotional states: negative (frustration) and positive (anticipation). The dogs' facial expressions were measured using the Dogs Facial Action Coding System (DogFACS). Two different approaches are compared in relation to our aim: (1) a DogFACS-based approach with a two-step pipeline consisting of (i) a DogFACS variable detector and (ii) a positive/negative state Decision Tree classifier; (2) An approach using deep learning techniques with no intermediate representation. The approaches reach accuracy of above 71% and 89%, respectively, with the deep learning approach performing better. Secondly, this study is also the first to study explainability of AI models in the context of emotion in animals. The DogFACS-based approach provides decision trees, that is a mathematical representation which reflects previous findings by human experts in relation to certain facial expressions (DogFACS variables) being correlates of specific emotional states. The deep learning approach offers a different, visual form of explainability in the form of heatmaps reflecting regions of focus of the network's attention, which in some cases show focus clearly related to the nature of particular DogFACS variables. These heatmaps may hold the key to novel insights on the sensitivity of the network to nuanced pixel patterns reflecting information invisible to the human eye.

Evaluation of shelter dog activity levels before and during COVID-19 using automated analysis.

Animal shelters have been found to represent stressful environments for pet dogs, both affecting behavior and influencing welfare. The current COVID-19 pandemic has brought to light new uncertainties in animal sheltering practices which may affect shelter dog behavior in unexpected ways. To evaluate this, we analyzed changes in dog activity levels before COVID-19 and during COVID-19 using an automated video analysis within a large, open-admission animal shelter in New York City, USA. Shelter dog activity was analyzed during two two-week long time periods: (i) just before COVID-19 safety measures were put in place (Feb 26-Mar 17, 2020) and (ii) during the COVID-19 quarantine (July 10-23, 2020). During these two periods, video clips of 15.3 second, on average, were taken of participating kennels every hour from approximately 8 am to 8 pm. Using a two-step filtering approach, a matched sample (based on the number of days of observation) of 34 dogs was defined, consisting of 17 dogs in each group (N1/N2 = 17). An automated video analysis of active/non-active behaviors was conducted and compared to manual coding of activity. The automated analysis validated by comparison to manual coding reaching above 79% accuracy. Significant differences in the patterns of shelter dog activity were observed: less activity was observed in the afternoons before COVID-19 restrictions, while during COVID-19, activity remained at a constant average. Together, these findings suggest that 1) COVID-19 lockdown altered shelter dog in-kennel activity, likely due to changes in the shelter environment and 2) automated analysis can be used as a hands-off tool to monitor activity. While this method of analysis presents immense opportunity for future research, we discuss the limitations of automated analysis and guidelines in the context of shelter dogs that can increase accuracy of detection, as well as reflect on policy changes that might be helpful in mediating canine stress in changing shelter environments.

Measurement of Canine Ataxic Gait Patterns Using Body-Worn Smartphone Sensor Data.

Ataxia is an impairment of the coordination of movement or the interaction of associated muscles, accompanied by a disturbance of the gait pattern. Diagnosis of this clinical sign, and evaluation of its severity is usually done using subjective scales during neurological examination. In this exploratory study we investigated if inertial sensors in a smart phone (3 axes of accelerometer and 3 axes of gyroscope) can be used to detect ataxia. The setting involved inertial sensor data collected by smartphone placed on the dog's back while walking in a straight line. A total of 770 walking sessions were evaluated comparing the gait of 55 healthy dogs to the one of 23 dogs with ataxia. Different machine learning techniques were used with the K-nearest neighbors technique reaching 95% accuracy in discriminating between a healthy control group and ataxic dogs, indicating potential use for smartphone apps for canine ataxia diagnosis and monitoring of treatment effect.

Clustering for Automated Exploratory Pattern Discovery in Animal Behavioral Data.

Traditional methods of data analysis in animal behavior research are usually based on measuring behavior by manually coding a set of chosen behavioral parameters, which is naturally prone to human bias and error, and is also a tedious labor-intensive task. Machine learning techniques are increasingly applied to support researchers in this field, mostly in a supervised manner: for tracking animals, detecting land marks or recognizing actions. Unsupervised methods are increasingly used, but are under-explored in the context of behavior studies and applied contexts such as behavioral testing of dogs. This study explores the potential of unsupervised approaches such as clustering for the automated discovery of patterns in data which have potential behavioral meaning. We aim to demonstrate that such patterns can be useful at exploratory stages of data analysis before forming specific hypotheses. To this end, we propose a concrete method for grouping video trials of behavioral testing of animal individuals into clusters using a set of potentially relevant features. Using an example of protocol for testing in a "Stranger Test", we compare the discovered clusters against the C-BARQ owner-based questionnaire, which is commonly used for dog behavioral trait assessment, showing that our method separated well between dogs with higher C-BARQ scores for stranger fear, and those with lower scores. This demonstrates potential use of such clustering approach for exploration prior to hypothesis forming and testing in behavioral research.

Automated recognition of pain in cats.

Facial expressions in non-human animals are closely linked to their internal affective states, with the majority of empirical work focusing on facial shape changes associated with pain. However, existing tools for facial expression analysis are prone to human subjectivity and bias, and in many cases also require special expertise and training. This paper presents the first comparative study of two different paths towards automatizing pain recognition in facial images of domestic short haired cats (n = 29), captured during ovariohysterectomy at different time points corresponding to varying intensities of pain. One approach is based on convolutional neural networks (ResNet50), while the other-on machine learning models based on geometric landmarks analysis inspired by species specific Facial Action Coding Systems (i.e. catFACS). Both types of approaches reach comparable accuracy of above 72%, indicating their potential usefulness as a basis for automating cat pain detection from images.

Objective Video-Based Assessment of ADHD-Like Canine Behavior Using Machine Learning.

Canine ADHD-like behavior is a behavioral problem that often compromises dogs' well-being, as well as the quality of life of their owners; early diagnosis and clinical intervention are often critical for successful treatment, which usually involves medication and/or behavioral modification. Diagnosis mainly relies on owner reports and some assessment scales, which are subject to subjectivity. This study is the first to propose an objective method for automated assessment of ADHD-like behavior based on video taken in a consultation room. We trained a machine learning classifier to differentiate between dogs clinically treated in the context of ADHD-like behavior and health control group with 81% accuracy; we then used its output to score the degree of exhibited ADHD-like behavior. In a preliminary evaluation in clinical context, in 8 out of 11 patients receiving medical treatment to treat excessive ADHD-like behavior, H-score was reduced. We further discuss the potential applications of the provided artifacts in clinical settings, based on feedback on H-score received from a focus group of four behavior experts.

Exploring the dog-human relationship by combining fMRI, eye-tracking and behavioural measures.

Behavioural studies revealed that the dog-human relationship resembles the human mother-child bond, but the underlying mechanisms remain unclear. Here, we report the results of a multi-method approach combining fMRI (N = 17), eye-tracking (N = 15), and behavioural preference tests (N = 24) to explore the engagement of an attachment-like system in dogs seeing human faces. We presented morph videos of the caregiver, a familiar person, and a stranger showing either happy or angry facial expressions. Regardless of emotion, viewing the caregiver activated brain regions associated with emotion and attachment processing in humans. In contrast, the stranger elicited activation mainly in brain regions related to visual and motor processing, and the familiar person relatively weak activations overall. While the majority of happy stimuli led to increased activation of the caudate nucleus associated with reward processing, angry stimuli led to activations in limbic regions. Both the eye-tracking and preference test data supported the superior role of the caregiver's face and were in line with the findings from the fMRI experiment. While preliminary, these findings indicate that cutting across different levels, from brain to behaviour, can provide novel and converging insights into the engagement of the putative attachment system when dogs interact with humans.

Computational Analysis of Movement Patterns of Dogs with ADHD-Like Behavior.

Computational approaches were called for to address the challenges of more objective behavior assessment which would be less reliant on owner reports. This study aims to use computational analysis for investigating a hypothesis that dogs with ADHD-like (attention deficit hyperactivity disorder) behavior exhibit characteristic movement patterns directly observable during veterinary consultation. Behavioral consultations of 12 dogs medically treated due to ADHD-like behavior were recorded, as well as of a control group of 12 dogs with no reported behavioral problems. Computational analysis with a self-developed tool based on computer vision and machine learning was performed, analyzing 12 movement parameters that can be extracted from automatic dog tracking data. Significant differences in seven movement parameters were found, which led to the identification of three dimensions of movement patterns which may be instrumental for more objective assessment of ADHD-like behavior by clinicians, while being directly observable during consultation. These include (i) high speed, (ii) large coverage of space, and (iii) constant re-orientation in space. Computational tools used on video data collected during consultation have the potential to support quantifiable assessment of ADHD-like behavior informed by the identified dimensions.