Source: MICHIGAN STATE UNIV submitted to
PHENOMICS AND GENOMICS OF BEHAVIORAL AND GROWTH TRAITS IN GROUP-HOUSED PIGS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
AFRI COMPETITIVE GRANT
Reporting Frequency
Annual
Accession No.
1011783
Grant No.
2017-67007-26176
Project No.
MICL08529
Proposal No.
2016-07995
Multistate No.
(N/A)
Program Code
A5171
Project Start Date
Apr 1, 2017
Project End Date
Mar 31, 2021
Grant Year
2017
Project Director
Steibel, J. P.
Recipient Organization
MICHIGAN STATE UNIV
(N/A)
EAST LANSING,MI 48824
Performing Department
Animal Science
Non Technical Summary
In livestock farming, production efficiency, health and welfare of group-housed animals may be compromised due to harmful social interactions between group members. For example, aggressive behavior among group-housed pigs significantly and routinely compromises productivity and welfare. Genetic selection against undesirable social behavior shows great promise for solving this problem. However, its implementation presents great challenges such as:a)social genetic models are extremely complex to elicit and fit and b) behavioral phenotyping of behavioraltraits is time consuming and labor intensive.In this project, we will study social interactions between group housed pigs and the genetic basis of those social interactions. We will improve models used for genetic evaluation of pigs by incorporating information on individual pig behavior and interactions. This will allow to predict not only each animal's individual performance, but also the effect of each animal in the social group the animal is housed with. Furthermore, we will also focus on the efficient collection of behavioral data. So far, behavioral phenotyping requires direct observation of video recording, which is impossible to implement in large scale pig breeding. We will compare computer programs for automatic video decoding and we will design optimal pens to make automatic video decoding more reliable. We will also study the use of automatic feeder data and thermographic images.All this work will ultimately increase pork production efficiency through the selection of animals that perform better under group housing conditions.
Animal Health Component
0%
Research Effort Categories
Basic
33%
Applied
34%
Developmental
33%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
3033599108150%
3153510108050%
Knowledge Area
303 - Genetic Improvement of Animals; 315 - Animal Welfare/Well-Being and Protection;

Subject Of Investigation
3510 - Swine, live animal; 3599 - Swine, general/other;

Field Of Science
1081 - Breeding; 1080 - Genetics;
Goals / Objectives
The overall goal of this project is to enhance global food security through improved pig well-being, by using phenomics to model direct and social genetic effects for performance, behavioral and welfare traits in group-housed pigs. Specifically, we propose:1) Estimate intensity of competition in order to parameterize genomic evaluation models to estimate direct and indirect (social) genetic effects of growth and lesion score traits in pigs.2) Examine automated methods for behavioral phenomic profiling of group-housed pigs to identify approaches which have the greatest potential for high throughput automated capture of behavior data on farm.3) Develop and deliver computational tools for analysis of genomics and phenomic data related to animal behavior.This proposal addresses Priority Area 2: Breeding and Phenomics of Food Crops and Animals of the Agriculture and Food Research Initiative Food Security Challenge Area: to develop locally adapted animal breeds that contribute to rural economic development and prosperity while enhancing regional and national resilience and food security. Specifically, our project will lead to identification of pigs that are able to enhance diversity and resilience of production systems by addressing the pork industry need of evaluating behavioral phenotypes related to group housing.
Project Methods
Objective 1: We will fit indirect genetic effects models using a number of different interaction matrices, including those based on behavioral analyses of group housed pigs.Objective 2a: We will conducta comprehensive review will examine the various approaches being used for automatic video analysis that could be suitable for behavioral analyses.Objective 2b: We will perform correlation analyses of different patterns of feeding behavior and aggressive behavior.Objective 2c: We will perform correlation analyses of thermographic image output and aggressive behavior in pigs.Objective 2d: We will use information from objectives 2a,b,c to design pens for optimal phenomics of pig behavior. We will implement those pen design and collect video, thermography and feeding behavior data.Objective 3: we will deliver programs for 1) IGE modeling, 2) decoding of behavioral phenotyping, 3) Algorithms for detecting aggression in video data.

Progress 04/01/17 to 03/31/21

Outputs
Target Audience: Animal Geneticists Biological Engineers Livestock production experts Livestock industry managers Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Students were cross trained accross disciplines. Belcy Angarita Barajas and Junjie Han are animal science students who were cross trained on phenomics, quantitative genetics and machine learning/computer vision. Chen Chen, a biolotical engineerring PhD student in Catholic University of Leuven was trained on computer vision and machine learning applied to behavioral phenomics. Kaitlin Wurtz and Csrly O'Malley, animal behavior postdocs, were trained on the use of comptuer vision for animal tracking and behavioral phenoyping and data analysis Two undergraduate students were trained in behavioral analyses. How have the results been disseminated to communities of interest?It was tough with last year's covid restrictions. However, we were able to report and disseminate results in multiple venues: 1) Before lock-down a) PI Juan Steibel presented this work at the NRSP8 meeting in PAG in January. The audience was mostly animal genomics and genetics researchers b) PI Juan Steibel co-organized and chaired a PLF symposium at Mid west ASAS in Omaha NE, It included results of this grant and other groups. The audience included Biological engineers, livestock management and production researchers as well as private sector PLF and genomics firms. c) PhD Student Junjie Han presented part of this work in MW-ASAS too. 2) After lock down a) We reported via teleconference in industry and academic meetings. PI Juan Steibel received an invitation to present at a ToPigs anual meeting. And there was also a virtual presentation at NRSP-8. b) PI Steibel and coPI Siegford helped to co-organize and presented at an MSU extension event (Field Days) dedicated to Precision Livestok Farming and Livestock Phenomics that was attended by an audience of almost 100 persons with a strong international precense from academic and corporate institutions. What do you plan to do during the next reporting period to accomplish the goals?This is the final report. We fulfulled and evene exceeded all the proposed aims.

Impacts
What was accomplished under these goals? All proposed goals have been accomplished: 1) We published a final paper on the estimation of social effects (Angarita et al 2021 accepted by JAS) using automatic feeding station data. With this, it is a total of two papers published under this objective durting the life of the paper. Note this paper also includes activities under 2d 2) An invited book chapter was submited under objective 2a and 2b (wurtz et al 2020). Three papers were published for objectives 2a and 2b). For 2d) a paper on automatic feeding data streams for estimating social effects is now accepted (Angarita et al 2021 accepted by JAS). 3) Under this objective a paper and companion github have been published on the use of differential evolution to optimize deep learning models usied in phenomics (Han et al 2021). Finally a general opinion paper on phenomics that cuts across all the objectices of this grant was published in GSE (Perez Enciso and Steibel GSE 2021). With this we accomplished every single proposed objective. There will be more publications coming out in the next years that are the results of this grant activity, but those are "extra production" above and beyond what we promised when this grant was fundded. We will acknowledge the funding in every single publication, even those that occur after completion of this funding.

Publications

  • Type: Journal Articles Status: Accepted Year Published: 2021 Citation: Han, J.J., Reid, K., Gondro, C., & Steibel, J.P. (2021). Heuristic hyperparameter optimization of deep learning models for genomic prediction. G3. Accepted.
  • Type: Journal Articles Status: Accepted Year Published: 2021 Citation: Angarita, B. K., Han, J.J.; Cantet, R. J., Chewning, S.K., Wurtz, K. E., Siegford, J. M., Ernst, C. W., & Steibel, J. P. (2020). Estimation of direct and social effects of feeding duration in growing pigs using records from automatic feeding stations. Journal of Animal Science. Accepted.
  • Type: Book Chapters Status: Submitted Year Published: 2020 Citation: Wurtz, K. Norton, T. Siegford, J. and Steibel, J. Assessment of open-source programs for automated tracking of individual pigs within a group.
  • Type: Journal Articles Status: Published Year Published: 2021 Citation: P�rez-Enciso, Miguel, and Juan P. Steibel. "Phenomes: the current frontier in animal breeding." Genetics Selection Evolution 53, no. 1 (2021): 1-10.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Chen, Chen, Weixing Zhu, Juan Steibel, Janice Siegford, Junjie Han, and Tomas Norton. "Recognition of feeding behaviour of pigs and determination of feeding time of each pig by a video-based deep learning method." Computers and Electronics in Agriculture 176 (2020): 105642.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Chen, Chen, Weixing Zhu, Juan Steibel, Janice Siegford, Junjie Han, and Tomas Norton. "Classification of drinking and drinker-playing in pigs by a video-based deep learning method." Biosystems Engineering 196 (2020): 1-14.


Progress 04/01/19 to 03/31/20

Outputs
Target Audience:1) Researchers in the area of animal breeding and genetics interested in genetic improvement in animal welfare. 2) Researchers in pig welfare and behavior. 3) researchers in the area of precision livestock farming Changes/Problems:There were no major changes or problems, except for one graduate student (Chewning) leaving her program due to personal/health reeasons early in 2019. We had no time to recruitanother student and we adapted by dividing the student's task among a PI (Steibel) and a vissiting student Angarita, who extended her stay at MSU. We asked for an extension to finish the work left unfinished by the graduate student. What opportunities for training and professional development has the project provided?Several undergraduate students were trained in behavioral phenotyping by Dr. Janice Siegford. A graduate students (Junjie Han) was trained by Dr. Steibel on automatic phenotyping using computer vision and on-farm data collection for behavioral phenotypinv. An Internationalvissiting graduate student (Belcy Angarita) was trained by Dr. Steibel and Dr. Cantet on estimation of social genetic effects and on on-farm data collection for behavioral phenotyping. A graduate student was trained by Dr. Norton (Chen Chen) on computer vision for behavioral phenotyping. A postdoc (Kaitlin Wurtz) was trained by Dr. Siegford, Dr Steibel and Dr. Norton on animal tracking using computer vision. How have the results been disseminated to communities of interest?We published papers, delivered invited and contributed talks in scientific comferences. We also co-organized two workshops on Precision LivestockFarming: One in the European Conference on Precision Livestock Farming (Organizer: Siegford) in August 2010 and one in Midwest ASAS (Organizer Steibel) in march 2020. In these symposia we disseminated the work funded under this grant and invited presenters working on similar problems to exchange results and experiences. What do you plan to do during the next reporting period to accomplish the goals?Almost all experimental work is done now. Next resporting period will be focused on finishing analyses and publishing results. Objective 2 is almost completed, except for a few papers to be submitted or under review. Objective 1 is almost complete, we will publish one more paper on the use automatic feeders for inferring social genetic effects. Objective 3 will be completed in the next period when we release code for behavioral phenotyping using computer vision.

Impacts
What was accomplished under these goals? We published results from objective 1 incorporating behavioral observations to predict lession scores. From Objective 2 we published a systematic review on animal tracking using computer vision (Objective 2a is now completed). We performed experiments to test effects of lighting conditions and pen floowing color to track pigs (Objective 2c is now completd). We performed experiments and published two papers on automatic detection of aggression (Objective 2b). We delivered programs for incorporating behavioral measures into the estimation of social genetic effects (Objective 3).

Publications

  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Wurtz, K., Camerlink, I., D'Eath, R. B., Fern�ndez, A. P., Norton, T., Steibel, J. P., & Siegford, J. (2019). Recording behaviour of indoor-housed farm animals automatically using machine vision technology: A systematic review. PloS one, 14(12), e0226669.
  • Type: Journal Articles Status: Published Year Published: 2020 Citation: Chen, C., Zhu, W., Steibel, J.P., Siegford, J., Wurtz, K., Han, J., & Norton, T. (2020). Recognition of aggressive episodes of pigs based on convolutional neural network and long short-term memory. Computers and Electronics in Agriculture, 169, 105166.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Steibel, J. P., J. Han, S. Chewning, K. Wurtz, J.M. Siegford (2019). Combining automatic feeding records with image analyses to study feeder occupancy in grow-finish. Proceedings of the 9th European Congress on Precision Livestock Farming.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Wurtz, K., I. Camerlink, R. DEath, A. Pe�a Fern�ndez, J. Siegford and Steibel, J. P. (2019). Automated phenotyping of swine behavior using image analysis: A systematic review. Proceedings of the 9th European Congress on Precision Livestock Farming.
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: OMalley, C. I., Turner, S. P., DEath, R. B., Steibel, J. P., Bates, R. O., Ernst, C. W., & Siegford, J. M. (2019). Animal personality in the management and welfare of pigs. Applied Animal Behaviour Science, 218:104821
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Angarita, B. K., Cantet, R. J., Wurtz, K. E., Omalley, C. I., Siegford, J. M., Ernst, C. W., & & Steibel, J. P. (2019). Estimation of indirect social genetic effects for skin lesion count in group-housed pigs by quantifying behavioral interactions. Journal of animal science, 97(9), 3658-3668.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Steibel, J. P. (October 2019). Estimation of social Genetic effects by integrating behavioral phenomics. XVI Annual meeting of the Argentinian Chapter of the International Biometrical Society.
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Chen, C., Zhu, W., Liu, D., Steibel, J., Siegford, J., Wurtz, K., ... & Norton, T. (2019). Detection of aggressive behaviours in pigs using a RealSence depth sensor. Computers and Electronics in Agriculture, 166, 105003.


Progress 04/01/18 to 03/31/19

Outputs
Target Audience: 1) Researchers in the area of animal breeding and genetics interested in genetic improvement in animal welfare. 2) Researchers in pig welfare and behavior. Changes/Problems:One major challenge has been to recruit students to undertake all the work. Especially the analysis of feeder data, where we recruited a student and then she left the group. Accessing the software for tracking has been another one. But we have reassigned the work and we are confident that we will be able to accomplish all the proposed objectives. What opportunities for training and professional development has the project provided?Ample opportunities to train students were provided by this project. Under Objective 1: Belcy Angarita Barajas is a PhD student at UBA jointly adviced by Dr. Cantet and Dr. Steibel. Belcy worked as a vissiting student at MSU during the whole reporting period. Under Objective 2: Kaitlin Wurtz is a postdoctoral scientist at MSU, she is the lead person in objective 2a. Extra funds were secured for Kaitling to complete a 2 month intership at KUL to finish the review.Simone Foister (PhD student, SRUC) worked part-time in objective 2b. Sarah Chewning, PhD student at MSU worked part time in objective 2b utilizing MSU feeder data. Irene Camerlink, a post-doctoral Scientistat SRUC completed objective 2c. Junjie Han (PhD student at MSU), Dong Liu (PhD Student at KUL) are being trained in objective 2d and 3. Two part time undergraduate students were trained in decoding of video for behavioral phenotyping. How have the results been disseminated to communities of interest?We published papers and we presented talks and poster. Among those, we can include three invited presentations. What do you plan to do during the next reporting period to accomplish the goals?Objective 1: Apply social genetic effects of lesion scores to a larger dataset collected by the industry. Extend social genetic effects analysis to feed intake data. Objective 2: Test competing programs for animal tracking. Test programs for pig identification from static images. Objective 3: release final version of program to estimate social genetic effect. Release of programs for analysis of decoded video. Release improved version of animal tracking programs.

Impacts
What was accomplished under these goals? Objective 1: Models for estimation of social genetic effects were fit to existing MSU data. The incorporation of behavioral observations allowed to increase the recovery of genetic variance (direct and social genetic variances) for the trait lesion count. The results are very promising and they were communicated already in PAG and in MWASAS. A full paper is under preparation. Objective 2: Objective2a (perform a systematic review) is is almost complete. A team of three post-docs/graduatestudentes from MSU, SRUC and KUL have just finished the first draft. Objective 2b (using automatic feeding records to perform behavioral phenomics) is underway: a team at SRUC is analyzing an existing dataset with feeding records and agression and a team at MSU installed two feeders and it is collecting feeding data and performing video recordings for autmatic behavioral phenotyping, two abstracts has been presented. Analysis of thermal images for objectives 2c is completed and a paper is under review. Objective 2c . Objective 2d is underway. Testing of tracking technologies selected from accomplishing 2a has started. twoabstracts has been submitted. Also, under this objective, further funding have been secured between two collaborating institutions (MSU and SRUC) to recruit a joint Phd Student between:'Precision livestock technologies to improve welfare: needs and constraints of farmers and veterinary surgeons'. To begin Oct 2019. Objective 3: gwaR was expanded to accomodate social genetic effects. An EM algorithm have been

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Kaitlin Wurtz, Irene Camerlink, Rick DEath, Alberto Pe�a Fern�ndez, Janice Siegford, Juan Steibel. Automated phenotyping of swine behavior using image analysis: A systematic review. The European Conference on Precision Livestock Farming, Cork, Ireland, 2019
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Juan P. Steibel, Junjie Han, Sarah Chewning, Kaitlin Wurtz, Janice M. Siegford. Combining automatic feeding records and image analyses to study feeder occupancy in growing-finishing pigs. The European Conference on Precision Livestock Farming, Cork, Ireland, 2019
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Kaitlin Wurtz, Tomas Norton, Janice Siegford, Juan Steibel. Assessment of open-source programs for automated tracking of individual pigs within a group 53rd Congress of the International Society for Applied Ethology, Bergen, Norway, 2019
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Junjie Han, Sarah K. Chewning, Kaitlin E. Wurtz, Janice Siegford, Juan P. Steibel. Using 3D images and deep learning to predict feeder occupancy in grow-finish pigs. 2019 MWASAS Meeting. Omaha Nebraska. March 11-13 2019.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Belcy Karine. Angarita Barajas, Rodolfo Cantet, Kaitlin E. Wurtz, Carly OMalley, Janice Siegford, Catherine Ernst, Juan P. Steibel. Improved estimation of indirect social genetic effects in group-housed pigs by quantifying behavioral interactions. 2019 MWASAS Meeting. Omaha Nebraska. March 11-13 2019.
  • Type: Journal Articles Status: Published Year Published: 2018 Citation: Foister, S., Doeschl-Wilson, A., Roehe, R., Arnott, A., Boyle, L., Turner, S. 2018. Social network properties predict chronic aggression in commercial pig systems. PlosONE, 13 (10): e0205122; https://doi.org/10.1371/journal.pone.0205122
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Juan P. Steibel, Belcy K. Angarita Barajas, Junjie Han, Sarah Chewning, Kaitlin Wurtz, Carly O'Malley, Rodolfo Cantet, Janice M. Siegford and Catherine W. Ernst. MSU Station Report: Behavioral Phenotyping and Modeling of Social Genetic Effects. XXVII PAG. San Diego, USA Jan. 12-16, 2019.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Foister, S., Doeschl-Wilson, A., Roehe, R., Boyle, L. and Turner, S. 2018. Piggy in the middle: The cost - benefit trade-off of a central network position in regrouping aggression Proc. ISAE, Canada.
  • Type: Journal Articles Status: Published Year Published: 2018 Citation: Wurtz, K., J.M. Siegford, C.W. Ernst, N.E. Raney, R.O. Bates, J. P. Steibel. Genome-wide association analyses of lesion scores in group-housed pigs. 49(6):628-631
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Turner, S.P., Foister, S., Roehe, R., Boyle, L., Doeschl-Wilson, A. 2018. Social network structure predicts the outcome of aggressive interactions in pigs. Proc EAAP, Dubrovnik, Croatia.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: Steibel, J.P. (September 2018) Genetics of Aggressive Behavior in Group Housed Pigs. XIV National Congress on Porcine Production. Cordoba, Argentina.
  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2018 Citation: Steibel, J.P. (May 2018) Quantitative Genomics Research on Meat Quality and Behavioral Traits. Workshop: Technological Platform for Swine Breeding Programs. Curitiba, Brazil.


Progress 04/01/17 to 03/31/18

Outputs
Target Audience:1) Researchers in the area of animal breeding and genetics interested in genetic improvement in animal welfare. 2) Researchers in animal welfare and behavior. 3) Animal Breeding companies. Changes/Problems:There are no major changes to goals and proposed approach of this grant. and there are only minimal changes to the timeline. Our major challenge was recruiting personel. Consequently, we changed the proposed order of some tasks that will not affect the final outcomes of this project. In particular, we concentrated more heavily on the manual video decoding and the developement of programs for modeling social effects in this fist year and the actual fitting of those models will occur in the second year using already developed software. In the developement of image analysis programs, KUL team run into some technical diffuculties in processing large chunks of video in real time, and they are working on the developing of algorithms to reduce the field of analysis in images, through automatic detection of area of interest. This is not an unexpected challenge and it will not affect the final proposed outcomes. What opportunities for training and professional development has the project provided?UBA: Belcy Angarita Barajas was recruited as a PhD student. She received formal formal training in quantitative genetics at Univeristy of Buenos Aires and sent to MSU for performing analysis related to objective 1. MSU: Belcy Angarita Barajas has been perfoming all analyses related to Objective 1 under the supervision of Juan Steibel. Maria Arceo is a Central Michigan University graduate students (MS in Statistics) who is performing an internship with Juan P. Steibel. Her work focuses in the analysis of behavioral data, writing functions that will be incorporated into the analyses required by Objective 1 and into programs generated under objective 2. Kaitlin Wurtz is a post-doctoral researcher working under the supervision of Siegford and Steibel. She is performing the systematic review under objective 2. She is also leading video decodingefforts. Leuven: Graduate students A. Diana and Z. Dong were cross-trained on real time video analysis for automatic feature detection. Diana had experience in animal behavior and was trained in the use of algoritthm for automatic detection of agression. Dong had a strong background in computer science and was trainined on animal behavior applications of monitoring algororithms. SRUC: Irene Camerlink, A postdoctoral Fellow working under the supervision of Simon Turner, is performing analyses of thermal images. She was trained in the use and technical of thermal images related to assessments of agression. How have the results been disseminated to communities of interest?Most of the production for this past year has been in the form of abstracts and conference presentations, because they inclide preliminary results from the first year of work. Papers have also been submitted. What do you plan to do during the next reporting period to accomplish the goals?Objective 1: With all MSU video data decoded, we will be able to fit model with intensity of competition in the estimation of indirect social genetic effects to welfare and other traits. Objective 2: We will finish the Systematic review of software for automatic analysis of behavior and tracking of animals in groups. We will be able to perform testing of some of those software. We will conduct designed experiments at the MSU farm for further evaluation of those programs and the incorporation of other types of data, such as feeding behavior data from automatic feeders. Analysis of thermal images from SRUC will be completed. Objective 3: an improved version of gwaR will be publicly available to model social genetic effects. Improved behavioral detection programs by KUL will be released.

Impacts
What was accomplished under these goals? Accomplishments related to all objectives: Manual (human driven) video decoding of behavior in 1000 pigs in the MSU population is almost complete (estimated to March 25th: 85% complete). This is a major accomplishment that feeds all objctives (and especially objectives 1 and 2 with gold standard data). Objective 1: Models for the use of intensity of competition measures have been developed by J.P. Steibel, R.J.C. Cantet and their students. First a simulation study was performed to determine the estimability of indirect genetic effects. Second, estimation of indirect genetic effects using traditional competition models was performed. Use of intensity of competition in modeling indirect genetic effects is underway. Objective 2: Completion of objective 2a (perform a systematic review) is underway. A team of three post-docs/graduate studentes from MSU, SRUC and KUL has already defined criteria according to PRISMA standards. Analysis of thermal images for objectives 2b is also underway. Objective 3: Progress has been done by KLU in iomproving their algorithms for automatic detection of pig aggression.They determined that automatic video segmentation and definition of the region of interest is the bottleneck to improve algorithm for real time analysis of video in thedetectionof pig to pig agression.MSU has expanded gwaR for the modeling of indirect genetic effects. And the program is currently being tested with real data.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: J.P. Steibel, R.J.C. Cantet, Janice Siegford. 2017. Modelling intensity of interaction to estimate direct and indirect genetic effects. Proceedings of the 7th WAFL conference. Ede, The Netherlands.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Diana, A., Carpentier, L., Piette, D., Boyle, L., Tullo, E., Guarino, M., Norton, T. and Berckmans, D., 2017, January. An ethogram of biter and bitten weaner pigs during ear biting events. In Book of Abstract of the 68th Annual Meeting of the European Federation of Animal Science (Vol. 23, p. 129).
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Foister, S., Doeschl-Wilson, A., Roehe, R., Boyle, L. and Turner, S. 2017. Can aggressive network structures at mixing be used to predict lesion outcomes in pigs? Proc. 7th International Conference on Welfare Assessment at Farm and Group Level.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Carpentier, L., Diana, A., Boyle, L., Norton, T. and Berckmans, D., 2017, January. Behaviour of weaner pigs during ear biting as possible indicators for automatic detection. In Proceedings of the ISAE Benelux conference 2017 (p. 23).
  • Type: Journal Articles Status: Under Review Year Published: 2018 Citation: Wurtz, K., J.M. Siegford, C.W. Ernst, N.E. Raney, R.O. Bates, J. P. Steibel. Genome-wide association analyses of lesion scores in group-housed pigs. Under review. Animal Genetics