This is the author accepted manuscript. The final version is available from Elsevier via http://dx.doi.org/10.1016/j.applanim.2016.01.00

The use of technology to optimize the production and management of each individual animal is becoming key to good farming. There is a need for the real-time systematic detection and control of disease in animals in order to limit the impact on animal welfare and food supply. Diseases such as footrot and mastitis cause significant pain in sheep, and so early detection is vital to ensuring effective treatment and preventing the spread across the flock. Facial expression scoring to assess pain in humans and non-humans is now well utilized, and the Sheep Pain Facial Expression Scale (SPFES) is a tool that can reliably detect pain in this species. The SPFES currently requires manual scoring, leaving it open to observer bias, and it is also time-consuming. The ability of a computer to automatically detect and direct a producer as to where assessment and treatment are needed would increase the chances of controlling the spread of disease. It would also aid in the prevention of resistance across the individual, farm, and landscape at both national and international levels. In this paper, we present our framework for an integrated novel system based on techniques originally applied for human facial expression recognition that could be implemented at the farm level. To the authors’ knowledge, this is the first time that this technology has been applied to sheep to assess pain.

Pain is a sensory and emotional experience that significantly affects animal welfare and has negative impacts on the economics of farming. Pain is often associated with common production diseases such as lameness and mastitis, as well as introduced to the animal through routine husbandry practices such as castration and tail docking. Farm animals are prey species which tend not to overtly express pain or weakness, making recognizing and evaluating pain incredibly difficult. Current methods of pain assessment do not provide information on what the animal is experiencing at that moment in time, only that its experience is having a long term negative impact on its behavior and biological functioning. Measures that provide reliable information about the animals’ affective state in that moment are urgently required; facial expression as a pain assessment tool has this ability. Automation of the detection and analysis of facial expression is currently in development, providing further incentive to use these methods in animal welfare assessment.

Validated assessment protocols have been developed to quantify welfare states for intensively managed sport, pleasure, and working horses. There are few protocols for extensively managed or free-roaming populations. Here, we trialed welfare indicators to ascertain their feasibility, reliability, and repeatability using free-roaming Carneddau Mountain ponies as an example population. The project involved (1) the identification of animal and resource-based measures of welfare from both the literature and discussion with an expert group; (2) testing the feasibility and repeatability of a modified body condition score and mobility score on 34 free-roaming and conservation grazing Carneddau Mountain ponies; and (3) testing a prototype welfare assessment template comprising 12 animal-based and 6 resource-based welfare indicators, with a total of 20 questions, on 35 free-roaming Carneddau Mountain ponies to quantify inter-assessor reliability and repeatability. This pilot study revealed that many of the indicators were successfully repeatable and had good levels of inter-assessor reliability. Some of the indicators could not be verified for reliability due to low/absent occurrence. The results indicated that many animal and resource-based indicators commonly used in intensively managed equine settings could be measured in-range with minor modifications. This study is an initial step toward validating a much-needed tool for the welfare assessment of free-roaming and conservation grazing ponies.