| 1. |
- Haubro Andersen, Pia, et al.
(författare)
-
Towards Machine Recognition of Facial Expressions of Pain in Horses
- 2021
-
Ingår i: Animals. - : MDPI. - 2076-2615. ; 11:6
-
Forskningsöversikt (refereegranskat)abstract
- Simple Summary Facial activity can convey valid information about the experience of pain in a horse. However, scoring of pain in horses based on facial activity is still in its infancy and accurate scoring can only be performed by trained assessors. Pain in humans can now be recognized reliably from video footage of faces, using computer vision and machine learning. We examine the hurdles in applying these technologies to horses and suggest two general approaches to automatic horse pain recognition. The first approach involves automatically detecting objectively defined facial expression aspects that do not involve any human judgment of what the expression "means". Automated classification of pain expressions can then be done according to a rule-based system since the facial expression aspects are defined with this information in mind. The other involves training very flexible machine learning methods with raw videos of horses with known true pain status. The upside of this approach is that the system has access to all the information in the video without engineered intermediate methods that have filtered out most of the variation. However, a large challenge is that large datasets with reliable pain annotation are required. We have obtained promising results from both approaches. Automated recognition of human facial expressions of pain and emotions is to a certain degree a solved problem, using approaches based on computer vision and machine learning. However, the application of such methods to horses has proven difficult. Major barriers are the lack of sufficiently large, annotated databases for horses and difficulties in obtaining correct classifications of pain because horses are non-verbal. This review describes our work to overcome these barriers, using two different approaches. One involves the use of a manual, but relatively objective, classification system for facial activity (Facial Action Coding System), where data are analyzed for pain expressions after coding using machine learning principles. We have devised tools that can aid manual labeling by identifying the faces and facial keypoints of horses. This approach provides promising results in the automated recognition of facial action units from images. The second approach, recurrent neural network end-to-end learning, requires less extraction of features and representations from the video but instead depends on large volumes of video data with ground truth. Our preliminary results suggest clearly that dynamics are important for pain recognition and show that combinations of recurrent neural networks can classify experimental pain in a small number of horses better than human raters.
|
|
| 2. |
- Klütsch, Cornelya, et al.
(författare)
-
Regional occurrence, high frequency but low diversity of mitochondrial DNA haplogroup d1 suggests a recent dog-wolf hybridization in Scandinavia
- 2011
-
Ingår i: Animal Genetics. - : Wiley. - 0268-9146 .- 1365-2052. ; 42:1, s. 100-103
-
Tidskriftsartikel (refereegranskat)abstract
- P>The domestic dog mitochondrial DNA (mtDNA)-gene pool consists of a homogenous mix of haplogroups shared among all populations worldwide, indicating that the dog originated at a single time and place. However, one small haplogroup, subclade d1, found among North Scandinavian/Finnish spitz breeds at frequencies above 30%, has a clearly separate origin. We studied the genetic and geographical diversity for this phylogenetic group to investigate where and when it originated and whether through independent domestication of wolf or dog-wolf crossbreeding. We analysed 582 bp of the mtDNA control region for 514 dogs of breeds earlier shown to harbour d1 and possibly related northern spitz breeds. Subclade d1 occurred almost exclusively among Swedish/Finnish Sami reindeer-herding spitzes and some Swedish/Norwegian hunting spitzes, at a frequency of mostly 60-100%. Genetic diversity was low, with only four haplotypes: a central, most frequent, one surrounded by two haplotypes differing by an indel and one differing by a substitution. The substitution was found in a single lineage, as a heteroplasmic mix with the central haplotype. The data indicate that subclade d1 originated in northern Scandinavia, at most 480-3000 years ago and through dog-wolf crossbreeding rather than a separate domestication event. The high frequency of d1 suggests that the dog-wolf hybrid phenotype had a selective advantage.
|
|
| 3. |
- Li, Ci, et al.
(författare)
-
The Poses for Equine Research Dataset (PFERD)
- 2024
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 11:1
-
Tidskriftsartikel (refereegranskat)abstract
- Studies of quadruped animal motion help us to identify diseases, understand behavior and unravel the mechanics behind gaits in animals. The horse is likely the best-studied animal in this aspect, but data capture is challenging and time-consuming. Computer vision techniques improve animal motion extraction, but the development relies on reference datasets, which are scarce, not open-access and often provide data from only a few anatomical landmarks. Addressing this data gap, we introduce PFERD, a video and 3D marker motion dataset from horses using a full-body set-up of densely placed over 100 skin-attached markers and synchronized videos from ten camera angles. Five horses of diverse conformations provide data for various motions from basic poses (eg. walking, trotting) to advanced motions (eg. rearing, kicking). We further express the 3D motions with current techniques and a 3D parameterized model, the hSMAL model, establishing a baseline for 3D horse markerless motion capture. PFERD enables advanced biomechanical studies and provides a resource of ground truth data for the methodological development of markerless motion capture.
|
|
| 4. |
- Rashid, Maheen, et al.
(författare)
-
Equine Pain Behavior Classification via Self-Supervised Disentangled Pose Representation
- 2022
-
Ingår i: 2022 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2022). - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 152-162
-
Konferensbidrag (refereegranskat)abstract
- Timely detection of horse pain is important for equine welfare. Horses express pain through their facial and body behavior, but may hide signs of pain from unfamiliar human observers. In addition, collecting visual data with detailed annotation of horse behavior and pain state is both cumbersome and not scalable. Consequently, a pragmatic equine pain classification system would use video of the un-observed horse and weak labels. This paper proposes such a method for equine pain classification by using multi-view surveillance video footage of unobserved horses with induced orthopaedic pain, with temporally sparse video level pain labels. To ensure that pain is learned from horse body language alone, we first train a self-supervised generative model to disentangle horse pose from its appearance and background before using the disentangled horse pose latent representation for pain classification. To make best use of the pain labels, we develop a novel loss that formulates pain classification as a multi-instance learning problem. Our method achieves pain classification accuracy better than human expert performance with 60% accuracy. The learned latent horse pose representation is shown to be viewpoint covariant, and disentangled from horse appearance. Qualitative analysis of pain classified segments shows correspondence between the pain symptoms identified by our model, and equine pain scales used in veterinary practice.
|
|
| 5. |
- Amundin, Mats, et al.
(författare)
-
A proposal to use distributional models to analyse dolphin vocalisation
- 2017
-
Ingår i: Proceedings of the 1st International Workshop on Vocal Interactivity in-and-between Humans, Animals and Robots, VIHAR 2017. - 9782956202905 ; , s. 31-32
-
Konferensbidrag (refereegranskat)abstract
- This paper gives a brief introduction to the starting points of an experimental project to study dolphin communicative behaviour using distributional semantics, with methods implemented for the large scale study of human language.
|
|
| 6. |
- Falkenberg Hansen, Kjetil, 1972-
(författare)
-
The acoustics and performance of DJ scratching, Analysis and modelling
- 2010
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- This thesis focuses on the analysis and modeling of scratching, in other words, the DJ (disk jockey) practice of using the turntable as a musical instrument. There has been experimental use of turntables as musical instruments since their invention, but the use is now mainly ascribed to the musical genre hip-hop and the playing style known as scratching. Scratching has developed to become a skillful instrument-playing practice with complex musical output performed by DJs. The impact on popular music culture has been significant, and for many, the DJ set-up of turntables and a mixer is now a natural instrument choice for undertaking a creative music activity. Six papers are included in this thesis, where the first three approach the acoustics and performance of scratching, and the second three approach scratch modeling and the DJ interface. Additional studies included here expand on the scope of the papers. For the acoustics and performance studies, DJs were recorded playing both demonstrations of standard performance techniques, and expressive performances on sensor-equipped instruments. Analysis of the data revealed that there are both differences and commonalities in playing strategies between musicians, and between expressive intentions. One characteristic feature of scratching is the range of standard playing techniques, but in performances it seems DJs vary the combination of playing techniques more than the rendering of these techniques. The third study describes some of the acoustic parameters of typical scratch improvisations and looks at which musical parameters are typically used for expressive performances. Extracted acoustic and performance parameters from the data show the functional ranges within which DJs normally play. Unlike traditional musical instruments, the equipment used for scratching was not intended to be used for creating music. The interface studies focus on traditional as well as new interfaces for DJs, where parameter mappings between input gestures and output signal are described. Standard performance techniques have been modeled in software called Skipproof, based on results from the first papers. Skipproof was used for testing other types of controllers than turntables, where complex DJ gestures could be manipulated using simplified control actions, enabling even non-experts to play expressively within the stylistic boundaries of DJ scratching. The last paper describes an experiment of using an existing hardware platform, the Reactable, to help designing and prototyping the interaction between different sound models and instrument interfaces, including scratching and Skipproof. In addition to the included papers, studies were conducted of expressivity, description of the emotional contents of scratching, DJ playing activities, and the coupling between playing techniques and sample. The physical affordances of the turntable, mixer and samples, as well as genre conventions of hip-hop, are assumed to explain some of the findings that distinguish scratching from other instrumental sounds or practices.
|
|
| 7. |
- Bruno, Karl
(författare)
-
Use and users of artificial insemination in Swedish dairy cattle breeding, 1935–1955
- 2022
-
Ingår i: History & Technology. - : Informa UK Limited. - 0734-1512 .- 1477-2620. ; 38:4, s. 317-343
-
Tidskriftsartikel (refereegranskat)abstract
- From the middle of the twentieth century, the new technologies and techniques of artificial insemination (a.i.) transformed dairy cattle husbandry and breeding across dairy-producing countries. While there are nuanced and multi-faceted studies of early a.i., previous work has not engaged much with its material usage, meaning that we know little about how different techniques, practices, and animal reactions promoted or restricted a.i’.s use for particular purposes. Here, I address this aspect by studying early a.i. in Swedish cattle breeding as a concrete set of technical artefacts used by humans and animals, thus reframing a.i’.s early history as a problem of use and users. I focus specifically on the artificial vagina, the predominant instrument used to gather bull semen, and show how the modes of use and non-use of the artificial vagina not only helped shape a.i. itself but also veterinary expertise, the institutions of breeding, and parts of the Swedish agrarian economy.
|
|
| 8. |
- Ka, Sojeong, 1976-, et al.
(författare)
-
Extremely Different Behaviours in High and Low Body Weight Lines of Chicken are Associated with Differential Expression of Genes Involved in Neuronal Plasticity
- 2009
-
Ingår i: Journal of neuroendocrinology (Print). - : Wiley. - 0953-8194 .- 1365-2826 .- 0022-0795 .- 1479-6805. ; 21:3, s. 208-216
-
Tidskriftsartikel (refereegranskat)abstract
- Long-term selection (> 45 generations) for low or high body weight from the same founder population has generated two extremely divergent lines of chickens, the low (LWS) and high weight (HWS) lines, which at the age of selection (56 days) differs by more than nine-fold in body weight. The HWS line chickens are compulsive feeders, whereas, in the LWS line, some individuals are anorexic and others have very low appetites. The involvement of the central nervous system in these behavioural differences has been experimentally supported. We compared a brain region at 0 and 56 days of age containing the major metabolic regulatory regions, including the hypothalamus and brainstem, using a global cDNA array expression analysis. The results obtained show that the long-term selection has produced minor but multiple expression differences. Genes that regulate neuronal plasticity, such as actin filament polymerisation and brain-derived neurotrophic factor, were identified as being differentially expressed. Genes involved in lipid metabolism were over-represented among differentially expressed genes. The expression data confirm that neural systems regulating feeding behaviours in these lines are different. The results suggest that the lines are set in separate developmental trajectories equipped with slightly different nervous systems. We suggest that the lines adapt behaviourally different to changing situations post hatch, such as the transition from dependence on yolk to feeding, in order to obtain energy. The present study has identified and exemplifies the kind of changes that may underlie the extreme differences in such behaviours.
|
|
| 9. |
- Leblanc, Neil, et al.
(författare)
-
Development of a magnetic bead microarray for simultaneous and simple detection of four pestiviruses
- 2009
-
Ingår i: Journal of Virological Methods. - : Elsevier BV. - 0166-0934 .- 1879-0984. ; 155:1, s. 1-9
-
Tidskriftsartikel (refereegranskat)abstract
- This study reports a novel method for the rapid detection and identification of the four recognized species in the pestivirus genus of the Flaviviridae family, i.e. classical swine fever virus (CSFV), border disease virus (BDV), bovine viral diarrhoea virus type 1 (BVDV1) and type 2 (BVDV2). The analysis of pestivirus PCR products was performed on microarrays by means of magnetic bead detection. The process utilizes an oligonucleotide array, onto which 5' biotinylated PCR products were hybridized, followed by visualization with streptavidin-coated magnetic particles by the naked eye, microscope or biochip reader. The assay was tested on a collection of pestiviruses that included all four species and allowed a specific and sensitive detection. Sensitivity was compared with other post-PCR detection methods, namely gel electrophoresis and suspension microarray. The results indicate that due to its high sensitivity, specificity and simple detection procedure, the magnetic bead assay provides a powerful tool for detection and identification of viral pathogens. Considering the simplicity of the assay, the protocols for hybridization and magnetic bead detection offer an emerging application for molecular diagnoses in virology that is amenable for use in a modestly equipped laboratory.
|
|
| 10. |
- Rising, Anna, et al.
(författare)
-
Spider silk proteins : recent advances in recombinant production, structure-function relationships and biomedical applications
- 2011
-
Ingår i: Cellular and Molecular Life Sciences (CMLS). - : Springer. - 1420-682X .- 1420-9071. ; 68:2, s. 169-84
-
Forskningsöversikt (refereegranskat)abstract
- Spider dragline silk is an outstanding material made up of unique proteins-spidroins. Analysis of the amino acid sequences of full-length spidroins reveals a tripartite composition: an N-terminal non-repetitive domain, a highly repetitive central part composed of approximately 100 polyalanine/glycine rich co-segments and a C-terminal non-repetitive domain. Recent molecular data on the terminal domains suggest that these have different functions. The composite nature of spidroins allows for recombinant production of individual and combined regions. Miniaturized spidroins designed by linking the terminal domains with a limited number of repetitive segments recapitulate the properties of native spidroins to a surprisingly large extent, provided that they are produced and isolated in a manner that retains water solubility until fibre formation is triggered. Biocompatibility studies in cell culture or in vivo of native and recombinant spider silk indicate that they are surprisingly well tolerated, suggesting that recombinant spider silk has potential for biomedical applications.
|
|
