Source: FORT VALLEY STATE UNIVERSITY submitted to NRP
GEO-INFORMATION TECHNOLOGY AND ARTIFICIAL INTELLIGENCE APPLICATION BASED ANIMAL HEALTH MANAGEMENT DECISION SUPPORT SYSTEM DEVELOPMENT TO SUPPORT SMALL-HOLDING FARMERS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
OTHER GRANTS
Reporting Frequency
Annual
Accession No.
1028574
Grant No.
2022-38821-37299
Cumulative Award Amt.
$749,873.00
Proposal No.
2021-12820
Multistate No.
(N/A)
Project Start Date
May 15, 2022
Project End Date
May 14, 2025
Grant Year
2022
Program Code
[EQ]- Research Project
Recipient Organization
FORT VALLEY STATE UNIVERSITY
1005 STATE UNIVERSITY DRIVE
FORT VALLEY,GA 31030
Performing Department
Agricultural Research Station
Non Technical Summary
Production of sheep and goats is growing in the United States, mainly with small and limited resource farmers in the Southeast.However,several things are limiting the growth of the small ruminant industry in this region. These include marginal soils unsuitable for high-input forage production, infection with internal parasites, and lack of farmer access to expert assistance on pasture production, animal health, and parasite management. Through artificial intelligence (AI) techniques, a team of plant and animal scientists, parasitologists, extension specialists, agriculture engineer and computer modelers will work together to develop an automated decision support system (aDSS) for forage and parasite management that farmers can access through their cellphones. The work will include a computer model to show farmers the best places to grow sericea lespedeza, an anti-parasitic forage plant, based upon soil properties, rainfall, and other climatic parameters. We will also develop a smartphone APP for farmers to identify and treat parasite-infected animals based upon their movement on pastures. We will train farmers on how to use the aDDS to grow high-quality forages that can keep their animals healthy, and finally, we will install a new AI computer laboratory for use by scientists and students at Fort Valley State University to support this work.
Animal Health Component
25%
Research Effort Categories
Basic
25%
Applied
25%
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
31316991110100%
Knowledge Area
313 - Internal Parasites in Animals;

Subject Of Investigation
1699 - Pasture and forage crops, general/other;

Field Of Science
1110 - Parasitology;
Goals / Objectives
The overall goal of the project is to develop an automated decision support system (aDSS) farmers can access through their cellphones. The specific objectives will be to i) evaluate growth and quality of sericea lespedeza, an anti-parasitic plant, to build a pasture production model supported by artificial intelligence (AI) to show farmers where to grow SL and how to improve forage yield and quality; ii) test a telemetry system to monitor animal activity related to disease outbreaks and provide real-time treatment information through a smartphone APP; iii) through Farmer Field Schools, train farmers in the use of the aDSS website to access the information they need to grow high-quality forages and keep their animals healthy, and iv) establish a Geospatial Technology Laboratory at Fort Valley State University to support these project objectives.
Project Methods
The Site Specific Forage Management Decision Support System (SSFMDSS) will be developed as an automated model for use by farmers in the southeastern U.S. for efficient production of sericea lespedeza (SL), a tannin-rich, anti-parasitic plant. Data related to land-use, topography, soils, temperature, precipitation, moisture content, water activity, and other SL growth parameters will be acquired for the study area, which will be existing SL pastures at multiple producer sites in South Carolina, Georgia, and Alabama. The SSFMDSS model will be strengthened with the inclusion of separate models correlating SL forage biomass, moisture content, extractable condensed tannin (ECT), protein-bound CT (PBCT), total CT (TCT), and mg protein-binding CT with field topography, soil quality, and climate parameters, as well as Remote Sensing (RS) data obtained with an Unmanned Aerial Vehicle (UAV; drone) that will collect images of SL fields at the FVSU Research Station, as well as others from fields of cooperating farmersat various locations throughoutthe proposed study area. We will use the tested and efficacy-confirmed Radial Basis Function Network (RBFN) modeling approach for our study. A step-by-step model optimization procedure will be developed for this study to obtain the best correlation between input and output parameters in RBFN. The learning rate, momentum term, and iteration rates will be changed alternately to optimize the RBFN models so that optimal prediction accuracies will be obtained. Once developed, the SSFMDSS will be useful in enhancing farmers' SL production with higher yield and forage quality to enhance animal (small ruminant) growth andprovide high quality nutraceutical hay for potential marketing.Radio Frequency Identification (RFID) assisted transponder and collars will be used on 100 animals (goats and sheep) identified from various locations/farms in the proposed study area, including the on-station FVSU agriculturalresearch farm and several on-farm collaborators' sites. Data will be streamed to Microsoft Cloud (Azure) at the University of North Georgia through the telemetry process and API coding. These data will be analyzed with the 3 years of on-farm animal movement RFID data available from the project's PI from South Africa (Van Wyk, UP) and from our U.S. based study sites to develop the signal ranges for detecting disease-related behavior in the animals. For the next step, we will develop a cellphone APP for the Android operating system and later for other OS using the software developed. The cellphone APP will be designed to directly communicate with the UNG AZURE server that streams data from the farms. We will also design the APP to generate warnings about each animal's status based on its movement. Then we will modify the APP to create a calling option for farmers to provide feedback to personnel at FVSU.We will create laboratory space in the FVSU Ag. Research Stationand equip it with appropriate hardware and software related to GIS, RS, GNSS, and IT (Database Management, WebGIS Server Technology) as part of this project. Research results will be disseminated to clientele groups through refereed journal articles, presentations at field days, scientific and producer meetings, and through a project page on the American Consortium for Small Ruminant Parasite Control website (wormx.info).

Progress 05/15/24 to 05/14/25

Outputs
Target Audience:The target audiences reached during this reporting period included producers in the southern USA (South Carolina) and South Africa, as well as extension personnel and researchers at field research sites in South Africa. As part of this project, a web-based decision support system (DSS) model for predicting optimal sites for production of sericea lespedeza (Lespedeza cuneata) as an antiparasitic nutraceutical forage based upon weather, soil conditions, ground cover, elevation, etc., was developed and validated through ground-truthing on producer farms and at University research sites in the southern USA. In the thirdyear of this project, contact was made with producers, scientists, and extension personnel at these sites in the USA for project-related research and outreach planning. Changes/Problems:We have faced numerous challenges during the accomplishment of the work for 2024-2025Some of the challenges that we have faced were: 1. Procurement of software and software renewals as needed on time. 3. We are facing some serious issues in purchasing top of the line drones (UAV) and drone accessories due to the out of stock problems. 4. Purchasing and delivery of items that are required for the whole project is getting difficult due to the changes taking place due to change in rules and regulation. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Results of this work has been published in several refereed journal articles, conference proceedings, and abstracts, the work has been presented in posters and oral presentations at national and international conferences, and information on our research findings have been included in the website of the American Consortium for Small Ruminant Parasite Control (ASCRPC). What do you plan to do during the next reporting period to accomplish the goals?Complete work with radio frequency identification transponders/global positioning system tags to monitor animal activity and movement of grazing goats with varying levels of parasitic infection.

Impacts
What was accomplished under these goals? Our Geospatial Technology Laboratory at FVSU was successfully established with appropriate hardware and software installed. Students have been using the laboratory to process machine learning modeling related to animal health monitoring and field identification of bioactive (anti-parasitic) forages. Machine learning models were successfully developed for rapid anemia detection in goats using analysis of patterns of blood drops on glycerine-soaked filter paper, image analysis of lower eyelid conjunctiva of goats to determine varying levels of infection with gastrointestinal nematodes, and identification of sericea lespedeza (SL) compared with weeds and other plants in the field, including a developed cellphone APP for farmers to use for SL identification.

Publications

  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Siddique, A., Panda, S. S., Khan, S., Dargan, S. T., Lewis, S., Carter, I., ... & Terrill, T. H. (2024). Innovations in animal health: artificial intelligence-enhanced hematocrit analysis for rapid anemia detection in small ruminants. Frontiers in Veterinary Science, 11, 1493403.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Neha, A., Shaik, A., Chelkapally, S. C., Kolikapongu, R. S., Namani, S. C., Erukulla, T., ... & Terrill, T. H. (2024). Effect of feeding a blackseed meal-sericea lespedeza leaf meal pellet on gastrointestinal nematode and coccidia infection and animal performance in young goats. Veterinary Parasitology, 331, 110253.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Siddique, A., K. Cook, Y. Holt, S.S. Panda, A.K. Mahapatra, E.R. Morgan, J.A. van Wyk, and T.H. Terrill. 2024. From plants to pixels: The role of artificial intelligence in identifying sericea lespedeza in field-based studies. Agronomy 14:992.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Panda, S.S., T.H. Terrill, A. Siddique, A.K. Mahapatra, E.R. Morgan, A.A. Pech-Cervantes, and J.A. van Wyk. 2024. Development of a decision support system for animal health management using geo-information technology: A novel approach to precision livestock management. Agriculture 14:696.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Aftab Siddique, Andres A. Pech-Cervantes, Jan A. van Wyk, Sudhanshu S. Panda, Ajit Mahapatra, Eric R. Morgan, and Thomas H. Terrill. 2024. Innovative strategies in image classification: Targeting sericea lespedeza identification. Proceedings of the American Forage and Grassland Conference, January 7-10, 2024, Mobile, AL.
  • Type: Other Status: Published Year Published: 2024 Citation: Aftab Siddique, Thomas H. Terrill, Sudhanshu S. Panda, Ajit K. Mohapatra, Eric Morgan, Andres A. Pech-Cervantes, Zhanyou Xu, and Jan A. Van Wyk. 2024. Enhanced Image Classification of Agricultural Forage Plants and Weeds through a CNN Model Utilizing RMSprop Optimization. AI in Agriculture conference, April 15-17, 2023, Texas A &M University, College Station, TX.
  • Type: Books Status: Submitted Year Published: 2025 Citation: Terrill, T.H. and O.M. Samples (Eds.). 2024. The One Health Model as Applied to Zoonotic Diseases. Wiley Publishing.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Sudhanshu Panda, Thomas Terrill, Ajit Mahapatra, Eric Morgan, Aftab Siddique, Jan Van Wyk. 2024. Climate Smart and Pathogen Impact Limiting Sericea Lespedeza Fodder Production Spatial Suitability Analysis in the Southern African Development Community Countries with Geospatial Engineering and Technology Support. IAEC (2024) International Conference, May 22-24, 2024, Bankok, Thailand.
  • Type: Other Status: Published Year Published: 2024 Citation: Aftab Siddique, Andres A. Pech-Cervantes, Jan A. van Wyk, Sudhanshu S. Panda, Ajit Mahapatra, Eric R. Morgan, and Thomas H. Terrill. 2024. Innovative strategies in image classification: targeting sericea lespedeza identification. American Forage and Grassland Council meeting, January 7-10, 2024, Mobile AL.


Progress 05/15/23 to 05/14/24

Outputs
Target Audience:The target audiences reached during this reporting period included producers in the southern USA (South Carolina) and South Africa, as well as extension personnel and researchers at field research sites in South Africa. As part of this project, a web-based decision support system (DSS) model for predicting optimal sites for production of sericea lespedeza (Lespedeza cuneata) as an anti-parasitic nutraceutical forage based upon weather, soil conditions, ground cover, elevation, etc., was developed and validated through ground-truthing on producer farms and at University research sites in the southern USA. In the second year of this project, contact was made with producers, scientists, and extension personnel at these sites in the USA for project-related research and outreach planning. Changes/Problems:Challenges: We have faced numerous challenges during the accomplishment of the work for 2023-2024 Some of the challenges that we have faced were: 1. Procurement of RFID tags, GPS tags from the collaborative company in South Africa. The transportation company was non cooperative during the shipment and delivery of the equipment. 2. Procurement of software and software renewals as needed on time. 3. We are facing some serious issues in purchasing top of the line drones (UAV) and drone accessories due to the out of stock problems. 4. Purchasing and delivery of items that are required for the whole project is getting difficult due to the changes taking place due to change in rules and regulation. What opportunities for training and professional development has the project provided?We have trained 12 undergraduate students and two master's students under the project. How have the results been disseminated to communities of interest?We have published three manuscripts in peer-reviewed scientific journals, published 12 abstracts, and two conference proceedings papers. We made ten oral or poster presentations at scientific meetings as part of this project (Two by PIs, eight by students). What do you plan to do during the next reporting period to accomplish the goals?We plan to fully incorporate the use of UAV's (drones) and RFID-GPS tags to monitor animal movement related to heath and nutraceutical forages, including big data analytics with AI-assisted technologies. Our goal is to publish three additional peer-reviewed manuscripts and make at least three presentations at national or internationalconferences.

Impacts
What was accomplished under these goals? Publications: 1. Panda, S., Terrill, T., Mahapatra, A., Morgan, E., Siddique, A., Pech-Cervantes, A. A., & Van Wyk, J. (2023, May). Geospatial Engineering and Technology Supported Climate Sensitive Sericea Lespedeza Fodder Production Suitability Analysis Modeling in the Southeastern United States. In 2023 IST-Africa Conference (IST Africa) (pp. 1-12). IEEE. 2. Panda, S. S., Terrill, T. H., Mahapatra, A. K., Morgan, E. R., Siddique, A., Pech-Cervantes, A. A., & van Wyk, J. A. (2023). Optimizing Sericea Lespedeza Fodder Production in the Southeastern US: A Climate-Informed Geospatial Engineering Approach. Agriculture, 13(9), 1661. 3. Panda, S., Terrill, T., Mahapatra, A., Morgan, E., Siddique, A., Pech-Cervantes A. A., Wyk, J. V. (2024). Development of a Decision Support System for Animal Health Management Using Geo Information Technology: A Novel Approach to Precision Livestock Management. We were able to publish three research articles in peer-reviewed journals. Conference Abstracts: 1. Randall, R., Siddique, A., Pech-Cervantes, A. A., van Wyk, J., Panda, S., Mahapatra, A., ... & Terrill, T. H. (2023). PSXIII-28 Advancing Anemia Detection in Small Ruminants with Convolutional Neural Networks and Smartphone Integration. Journal of Animal Science, 101(Supplement_3), 528-529. 2. Sudhanshu PANDA, Thomas TERRILL, Ajit MAHAPATRA, Eric MORGAN, Aftab SIDDIQUE, Jan VAN WYK (2024). Climate Smart and Pathogen Impact Limiting Sericea Lespedeza Fodder Production Spatial Suitability Analysis in the Southern African Development Community Countries with Geospatial Engineering and Technology Support. IAEC (2024) International Conference. 3. Aftab Siddique, Andres A. Pech-Cervantes, Jan A. van Wyk, Sudhanshu S. Panda, Ajit Mahapatra, Eric R. Morgan, and Thomas H. Terrill (2024). Innovative strategies in image classification: targeting sericea lespedeza identification. American Forage and Grassland Council meeting, Mobile AL, 2024. 4. A. V. Rubio*, K. Cook, A. Siddique, S. S. Panda, E. R. Morgan, J. A. van Wyk, A. K. Mahapatra, T. H. Terrill (2024). Empowering Farmers with AI: From Image Categorization to Precision Agriculture: Nutraceutical Forage Plant Identification. ARD meeting 2024. 5. D. Brown*, Y. Holt, A. Siddique, A.A. Pech-Cervantes, S. S. Panda, E. R. Morgan, J. A.van Wyk, A. K. Mahapatra, T. H. Terrill (2024). Deep Learning in Agriculture: An Analysis of Forage Plant Identification with Convolutional Neural Networks. 6. I. Carter*, A. Siddique, A.A. Pech-Cervantes, S. S. Panda, E. R. Morgan, J. A. van Wyk, A. K. Mahapatra, T. H. Terrill (2024). Development of a Hematocrit Biosensor for Rapid Anemia Detection in Small Ruminants using different input variables 7. Randall, R., Stegall, K., Siddique, A., Cervantes, A., Mahapatra, K, A., Morgan, E., Jan A. Wyk., Terrill, T (2024). Smartphone integration in anemia detection for small ruminants with Convolutional Neural Networks. ASAS Southeastern Branch Meeting 2024 8. S. Dargan, Siddique, A., Cervantes, A., Mahapatra, K, A., Morgan, E., Jan A. Wyk., Terrill, T (2024). AI-Enhanced Hematocrit Analysis: Revolutionizing rapid anemia detection in small ruminants. ASAS southeastern branch meeting 2024. 9. Randall, R., Stegall, K., Siddique, A., Cervantes, A., Mahapatra, K, A., Morgan, E., Jan A. Wyk., Terrill, T (2023). Smartphone Integration in Anemia Detection in Small Ruminants with Convolutional Neural Networks. ASAS 2023 (Program book) 10. Aftab Siddique*, Thomas H. Terrill, Sudhanshu S. Panda, Ajit K. Mohapatra, Eric Morgan, Andres A. Pech-Cervantes, Zhanyou Xu, and Jan A. Van Wyk (2024). Enhanced Image Classification of Agricultural Forage Plants and Weeds through a CNNModel Utilizing RMSprop Optimization. AI in Agriculture conference, Texas A &M University, 2024 11. Panda, S.S., Jat, P., Mahapatra, A.K., Terrill, T., Siddique, A., (2023). Site-Specific Fodder Management Decision Support System Development: Lespedeza Cuneata Biomass Production Analysis with Artificial Neural Network Modeling Approach. AI in Agriculture Conference, FL 12. Panda, S.S., Jat, P., Mahapatra, A.K., Terrill, T., Siddique, A., (2023). Site-Specific Fodder Management Decision Support System Development: Lespedeza Cuneata Biomass Production Analysis with Artificial Neural Network Modeling Approach. AI in Agriculture Conference, FL From the given objectives For objective one, our researcher team was able to publish one peer reviewed journal manuscript in agriculture journal in MDPI entitled as " Optimizing Sericea Lespedeza Fodder Production in the Southeastern US: A Climate-Informed Geospatial Engineering Approach" and one conference proceeding paper published in IEEE Xplore entitled as " Geospatial Engineering and Technology Supported Climate Sensitive Sericea Lespedeza Fodder Production Suitability Analysis Modeling in the Southeastern United States. We were also able to publish and present abstracts in AI in Agriculture conference 2023, in Orlando FL entitled as "Site-Specific Fodder Management Decision Support System Development: Lespedeza Cuneata Biomass Production Analysis with Artificial Neural Network Modeling Approach" and "Climate Smart and Pathogen Impact Limiting Sericea Lespedeza Fodder Production Spatial Suitability Analysis in the Southern African Development Community Countries with Geospatial Engineering and Technology Support" We were also able to publish a conference proceeding paper in AFGC conference entitled as "Innovative strategies in image classification: targeting sericea lespedeza identification. For objective 2: We were able to publish one research article in MDPI Agriculture Journal in special issue entitled as "Development of a Decision Support System for Animal Health Management Using Geo Information Technology: A Novel Approach to Precision Livestock Management". We were also able to publish and present at national and international meetings and conferences. We published 5 abstracts entitled as " Advancing Anemia Detection in Small Ruminants with Convolutional Neural Networks and Smartphone Integration.", "Development of a Hematocrit Biosensor for Rapid Anemia Detection in Small Ruminants using different input variables", "Smartphone integration in anemia detection for small ruminants with Convolutional Neural Networks", "AI-Enhanced Hematocrit Analysis: Revolutionizing rapid anemia detection in small ruminants", "Smartphone Integration in Anemia Detection in Small Ruminants with Convolutional Neural Networks." Objective 3: We are still working on arranging farmer's school for the decimation of aDSS models and providing training for livestock management. Objective 4: We were able to procure the computer systems for the establishments for our GIS lab. Systems are in place and ready to be used by students and farmer's school participants for training purposes. We have ordered the software and software renewal but are waiting for them to arrive and get them installed on the systems.

Publications

  • Type: Journal Articles Status: Published Year Published: 2023 Citation: Panda, S.S., T.H. Terrill, A.K. Mahapatra, E.R. Morgan, A. Siddique, A.A. Pech-Cervantes, and J.A. van Wyk. 2023. Optimizing Sericea Lespedeza Fodder Production in the Southeastern US: A Climate-Informed Geospatial Engineering Approach. Agriculture 2023, 13, 1661. Panda, Sudhanshu, Thomas Terrill, Ajit Mahapatra, Eric Morgan, Aftab Siddique, Andres A. Pech-Cervantes, and Jan Van Wyk. "Geospatial Engineering and Technology Supported Climate Sensitive Sericea Lespedeza Fodder Production Suitability Analysis Modeling in the Southeastern United States." In 2023 IST-Africa Conference (IST-Africa), pp. 1-12. IEEE, 2023.


Progress 05/15/22 to 05/14/23

Outputs
Target Audience:The target audiences reached during this reporting period included producers in the southern USA and South Africa, as well as extension personnel and researchers at field research sites in South Africa. As part of this project, a modelfor predicting optimal sites for production of sericea lespedeza (Lespedeza cuneata) as an anti-parasitic nutraceutical forage based upon weather, soil conditions, ground cover, elevation, etc., will be validated through ground-truthing on producer farms and at University research sites in both the USA and South Africa. In the first year of this project, contact was made with producers, scientists, and extension personnel at these sites in both countries for project-related research and outreach planning. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The FVSU PIs and Post-Doctoral Researcher attended the Artificial Intelligence (AI) Conference held during April 17-19, 2023, in Orlando, FL to learn more about AI and collection and analysis of big data related to plant and animal agriculture. How have the results been disseminated to communities of interest?Information on the project research and outreach plans were shared with producers, scientists, and extension specialists in face-to-face visits in both the southern USA and South Africa. Initial project results were shared with scientists and industry representatives in poster and oral presentations at the Artificial Intelligence Conference held in Orlando, FL from 17-19 April, 2023. Information on establishment and management of sericea lespedeza has been made available to a world-side audience online on the web site of the American Consortium for Small Ruminant Parasite Control (wormx.info). What do you plan to do during the next reporting period to accomplish the goals?We will use drones to collect images of sericea lespedeza fields on-farm and at University Research sites, as well as collecting forage and soil samples for laboratory analysis, forage production (yield) data, and weather-related data at each site in the USA (Objectives 1 and 2). We will collect RFID data on activity of parasitized goats and sheepin grazing trials at Research Station and/or on-farms sites in both the USA and South Africa, in addition to collecting FAMACHA data and blood and fecal sample analysis data related to parasite infection level (Objective 3). We will analyze collected data in the GIS/AI laboratory at FVSU and begin development of a cell phone-based automated decision support model APP for forage management and animal health recommendations (Objective 4).

Impacts
What was accomplished under these goals? The Post-doctoral researcher with experience in computer modelling and big data collection and analysis was hired at Fort Valley State University (FVSU). With the assistance of this individual, the location site for the new Geographic Information Systems (GIS)/Artificial Intelligence (AI)at th FVSU Agricultural Research Station was identified and equipped with the required computer hardware and software for big data analysis and model development. Project PIs from FVSU, the University of North Georgia, Queen's University Belfast (Northern Ireland), and the University of Pretoria (South Africa) met in South Africa in January, 2022, for project research and outreach planning and to visit potential project research sites in different parts of the country. Project-related abstracts were submitted and poster and oral presentations prepared for the 2023 Innovation, Science, and Technology (IST) Africa Conference (May 31-June 2, Tshwane, South Africa) and the annual American Society of Animal Science (ASAS) meeting (July 16-20, Albuquerque, NM). A paper was also prepared and submitted to be included in the 2023 IST-Africa Conference Proceedings. Abstracts were submitted and oral and poster presentations were also presented at the Artificial Intelligence (AI) Conference held on April 17-19, 2023, in Orlando, FL. A grazing trial with 80 yearling male Spanish goats was initiated in April, 2023at the FVSU Agricultural Research Station for collection of FAMACHA(anemia detection) data and other blood and fecal analysisdata related to infection with internal parasites, as well as radio frequency indentification (RFID) data related to animal activity. These data will be used in the development of automated animal health and forage production decision support systems in the project (Objectives 1-4).

Publications