15AGRITECHCAT4: Development and validation of a system for automatic detection of lameness in sheep
Lead Research Organisation:
University of Nottingham
Department Name: School of Veterinary Medicine and Sci
Abstract
This project aims to address challenge of lameness in sheep, one of the biggest cause of poor health, welfare and economic loss on sheep farms in UK and globally. By developing innovative hardware (acclerometer ), integrated software and algorithms specifically suited for automated lameness detection in sheep we aim to reduce the lameness levels. Such system will be innnovative and will facilitate farmers to effectively manage lameness by providing an automated system for early detection. This will bring economic and welfare benefits across the food supply chain and creating new jobs in production of innovative hardware and software products.
Technical Summary
This project aims to address challenge of lameness in sheep, one of the biggest cause of poor health, welfare and economic loss on sheep farms in UK and globally. By developing innovative hardware (acclerometer), integrated software and algorithms specifically suited for automated lameness detection in sheep we aim to reduce the lameness levels. Such system will be innnovative and will facilitate farmers to effectively manage lameness by providing an automated system for early detection. This will bring economic and welfare benefits across the food supply chain and creating new jobs in production of innovative hardware and software products.
Planned Impact
Possible outcomes of this research include;
A) Development of algorithms to classify lameness in sheep and other behaviors
B) Validation of those algorithms with data, and quantify impact of lameness on various activities of sheep e.g lying , grazing
Main beneficiaries of the research and how will they benefit;
1. Farmers - through a system of automated early lameness detection , effective lameness control leading to significant economic savings and improved welfare
2. Businesses - Dunbia and Farm Wizard to have wider access to market by rendering the system and lameness alert services and by boosting the sale of existing platforms
3. The sheep industry and Dunbia - through improved sustainability, production and welfare by reducing lameness. It will allow farmers to follow recommendations and drive the national average of flock to <2% as recommended by the Farm Animal Welfare Council.
4. Veterinary surgeons and farm advisors - through being able to use this as a platform to implement advice
5. Consumers - through consuming a product associated with improved sheep welfare and with improved farming sustainability.
6. UK PLC - through improved efficiency of sheep farming and economic returns.
A) Development of algorithms to classify lameness in sheep and other behaviors
B) Validation of those algorithms with data, and quantify impact of lameness on various activities of sheep e.g lying , grazing
Main beneficiaries of the research and how will they benefit;
1. Farmers - through a system of automated early lameness detection , effective lameness control leading to significant economic savings and improved welfare
2. Businesses - Dunbia and Farm Wizard to have wider access to market by rendering the system and lameness alert services and by boosting the sale of existing platforms
3. The sheep industry and Dunbia - through improved sustainability, production and welfare by reducing lameness. It will allow farmers to follow recommendations and drive the national average of flock to <2% as recommended by the Farm Animal Welfare Council.
4. Veterinary surgeons and farm advisors - through being able to use this as a platform to implement advice
5. Consumers - through consuming a product associated with improved sheep welfare and with improved farming sustainability.
6. UK PLC - through improved efficiency of sheep farming and economic returns.
Publications
Jarchi D
(2021)
Lameness Detection in Cows Using Hierarchical Deep Learning and Synchrosqueezed Wavelet Transform
in IEEE Sensors Journal
Walton E
(2018)
Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour.
in Royal Society open science
Kaler J
(2020)
Automated detection of lameness in sheep using machine learning approaches: novel insights into behavioural differences among lame and non-lame sheep.
in Royal Society open science
Mansbridge N
(2018)
Feature Selection and Comparison of Machine Learning Algorithms in Classification of Grazing and Rumination Behaviour in Sheep.
in Sensors (Basel, Switzerland)
Vázquez-Diosdado JA
(2019)
A Combined Offline and Online Algorithm for Real-Time and Long-Term Classification of Sheep Behaviour: Novel Approach for Precision Livestock Farming.
in Sensors (Basel, Switzerland)
Description | * We have developed novel algorithms that can correctly classify sheep behaviour and lameness. Our research work specifically looked into following elements doing series of studies: * We started with the exploring the effect of different sampling rates (8, 16 and 32 Hz), position of sensors (ear or collar), window sizes (3s, 5s and 7s) for signal processing on accuracy of algorithms for sheep behaviour and also effect of energy consumption. From the magnitude of accelerometer and gyroscope eleven feature characteristics were extracted resulting in a total of 44 feature characteristics. More details of this study are in the journal of Royal society open science (Walton et al., 2018). We investigated classification of three activities 'lying', 'standing' and 'walking'. Our algorithms indicated high accuracy for classifying all activities. More importantly the study results also highlighted that above-mentioned elements impact accuracy and there are trade-offs (between these elements) one needs to consider for a real time monitoring solution * Next series of studies looked at development of novel machine learning algorithms to classify sheep lameness specifically choice of algorithms and number of features. (Kaler et al., 2020). * We then worked on the implementation of algorithms on the 'edge' and employing edge analytics for a classification of sheep behaviour and analysis. This is novel and first of its kind in precision livestock monitoring as all the 'thinking' is done on the device and it offer potential benefits from battery life perspective. (Vazquez et al., 2019) |
Exploitation Route | The approach we used have wider applicability to researchers and Industry working on precision livestock and digital technology, sports medicine etc. |
Sectors | Agriculture Food and Drink Digital/Communication/Information Technologies (including Software) Healthcare Manufacturing including Industrial Biotechology Pharmaceuticals and Medical Biotechnology |
Description | We have developed prototype for predicting sheep behaviour using technology and under discussion for licensing. More recently I demonstrated and discussed on BBC countryfile (6th Feb 2022) based on some of the methods developed on the project another technology for early disease detection in calves |
First Year Of Impact | 2018 |
Sector | Agriculture, Food and Drink,Digital/Communication/Information Technologies (including Software),Electronics |
Impact Types | Societal Economic |
Description | BBSRC Intelligent Sensing Workshop |
Geographic Reach | National |
Policy Influence Type | Contribution to a national consultation/review |
Description | Talk to veterinary practicitioners attended by 25 delgates |
Geographic Reach | National |
Policy Influence Type | Influenced training of practitioners or researchers |
Impact | Impact on delivering proactive services to sheep farmers in management of lameness |
Description | BBSRC |
Amount | £18,677 (GBP) |
Organisation | Biotechnology and Biological Sciences Research Council (BBSRC) |
Sector | Public |
Country | United Kingdom |
Start | 09/2018 |
End | 02/2019 |
Description | BBSRC ISCF agri-seedling award |
Amount | £20,000 (GBP) |
Organisation | Biotechnology and Biological Sciences Research Council (BBSRC) |
Sector | Public |
Country | United Kingdom |
Start | 09/2017 |
End | 02/2018 |
Description | BBSRC Icase PhD Studentships |
Amount | £80,000 (GBP) |
Organisation | Biotechnology and Biological Sciences Research Council (BBSRC) |
Sector | Public |
Country | United Kingdom |
Start | 09/2018 |
End | 09/2022 |
Description | DECIDE |
Amount | € 9,900,000 (EUR) |
Funding ID | R00327 |
Organisation | European Commission H2020 |
Sector | Public |
Country | Belgium |
Start | 06/2021 |
End | 06/2026 |
Description | ESRC Impact Accelerator fund |
Amount | £5,000 (GBP) |
Organisation | Economic and Social Research Council |
Sector | Public |
Country | United Kingdom |
Start | 11/2017 |
End | 10/2018 |
Title | Algorithm to detect sheep activity |
Description | A random forest algorithm to detect sheep activity using sensor data |
Type Of Material | Computer model/algorithm |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | To be evaluated |
URL | https://royalsocietypublishing.org/doi/full/10.1098/rsos.171442 |
Title | Algorithm to detect sheep grazing |
Description | Various algorithms that can detect sheep grazing behaviour using sensor data |
Type Of Material | Computer model/algorithm |
Year Produced | 2018 |
Provided To Others? | Yes |
Impact | To be evaluated |
URL | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6210268/ |
Description | Collaboartion on algorithm development |
Organisation | Hewlett Packard Enterprise (HPE) |
Country | United Kingdom |
Sector | Private |
PI Contribution | Joint discussion on approaches to algorithm development, sharing domain expertise, sharing of data and tools for this. |
Collaborator Contribution | Access to platforms for algorithm development. |
Impact | Multidisciplinary collaboration, data secientists, epidemiologist, engineers. |
Start Year | 2016 |
Description | Collaboration for precision livestock technology |
Organisation | Prognostix Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | Utilised methodologies for data pre- processing and algorithm development on sensor data |
Collaborator Contribution | Assess to new data and technologies |
Impact | In progress will be update soon Multidisciplinary: Engineering and Computing |
Start Year | 2017 |
Description | Partnership to predict lameness in cattle using sensor and non sensor data |
Organisation | CRV |
Country | Netherlands |
Sector | Private |
PI Contribution | Expertise input on analysis precision livestock data and training of PhD |
Collaborator Contribution | Assess to 'big data' on lameness in cattle with multiple variables and combination of sensor and non sensor |
Impact | Multidisciplinary |
Start Year | 2019 |
Title | EL4L prototype |
Description | A device with implement algorithms for sheep behaviour |
Type Of Technology | New/Improved Technique/Technology |
Year Produced | 2018 |
Impact | At the moment used within consortium and for demonstration purposes |
Title | EL4L software |
Description | The EL4L algorithms for behaviour detection if sheep |
Type Of Technology | Software |
Year Produced | 2017 |
Impact | Used internally at the moment |
Description | American Society of Animal Science - Invited talk |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Postgraduate students |
Results and Impact | Invited talk on Precision technology |
Year(s) Of Engagement Activity | 2019 |
Description | Article in Science and Technology Magazine |
Form Of Engagement Activity | A magazine, newsletter or online publication |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Industry/Business |
Results and Impact | Article Scitech Europe Quarterly Magazine (Page -234-235) |
Year(s) Of Engagement Activity | 2018 |
URL | http://edition.pagesuite-professional.co.uk/html5/reader/production/default.aspx?pubname=&edid=0e6f6... |
Description | BBC Countryfile |
Form Of Engagement Activity | A broadcast e.g. TV/radio/film/podcast (other than news/press) |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Media (as a channel to the public) |
Results and Impact | Broadcast on use of precision technologies for responsible use of antibiotics on farms |
Year(s) Of Engagement Activity | 2022 |
Description | BBSRC Innovation Hub Oxford Farming conference |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Industry/Business |
Results and Impact | Showcased animal health as part of BBSRC Innovation Hub, gave demonstration of the prototype for sheep behaviour and lameness detection at the Oxford farming conference |
Year(s) Of Engagement Activity | 2018 |
Description | Interview for the BBC East Midlands TV |
Form Of Engagement Activity | A press release, press conference or response to a media enquiry/interview |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Media (as a channel to the public) |
Results and Impact | The development of the prototype for sheep lameness behaviour was reported and filmed. |
Year(s) Of Engagement Activity | 2018 |
Description | Invited talk to Sheep Health and Welfare Group |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Industry/Business |
Results and Impact | Around 100 members of sheep Industry including Animal health development board, Retailers ,farmers and vets. |
Year(s) Of Engagement Activity | 2018 |
URL | http://beefandlamb.ahdb.org.uk/returns/health-and-welfare/sheep-health-and-welfare-group-shawg/ |
Description | New Scientist 2019 |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Public/other audiences |
Results and Impact | As part of Centre of Innovation for Livestock we were asked to demonstrate the sheep sensor at the New Scientist Live in London. |
Year(s) Of Engagement Activity | 2019 |
Description | Press release in Vet Times |
Form Of Engagement Activity | A magazine, newsletter or online publication |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | An article in Vet times |
Year(s) Of Engagement Activity | 2018 |
URL | https://www.vettimes.co.uk/news/smart-tech-aims-to-identify-earliest-signs-of-lameness/ |
Description | Report in Veterinary Record |
Form Of Engagement Activity | A magazine, newsletter or online publication |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Article published by Veterinary Record. |
Year(s) Of Engagement Activity | 2018 |
URL | http://veterinaryrecord.bmj.com/content/182/4/102.1 |
Description | Talks at British Society of Animal Science |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Policymakers/politicians |
Results and Impact | Oral talks on precision livestock health and algorithms from sensor data and technology adoption in farmers by Jasmeet Kaler, Jorge Vazquez Diosdado, Eliana Lima |
Year(s) Of Engagement Activity | 2018 |
Description | UK China Working Group Meeting and Business Partnering on Agritech Innovation - 21 Feb 2017 Beijing |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Policymakers/politicians |
Results and Impact | Discussion on UK- China agritech initiatives |
Year(s) Of Engagement Activity | 2017 |