Supplementary Materials for A systematic review on the role of livestock ontologies in animal health management and disease surveillance: A PRISMA 2020 and AI-assisted screening approach

Supplementary Materials for "A systematic review on the role of livestock ontologies in animal health management and disease surveillance: A PRISMA 2020 and AI-assisted screening approach"

Review Article DOI: https://doi.org/10.1016/j.atech.2026.101977 

This document provides detailed insights into the systematic review process conducted for the study titled "A Review on the role of livestock ontologies in animal health management and disease surveillance". The review focused on ontology-based approaches for disease surveillance and health management in livestock farming, aiming to address challenges and emerging trends in this domain. Using ASReview LAB, 286 records were systematically reviewed, resulting in the identification of 100 relevant records. This document also includes a description of the databases search queries and step-by-step instructions for replicating the review process.

Version 4: Final dataset after co-author eligibility review (96 label revisions). This is the version used in the published paper.

Databases and Search Strategy

The literature search utilized PubMed, IEEE Xplore, and Google Scholar to identify relevant peer-reviewed articles and literature. Since academic databases are continuously updated with new research, it is important to note that the number of articles retrieved at the time of search may change in future searches using the same queries. The review scope focused on research published between 2011 and 2025, and the search results reflect the number of articles available at the time of retrieval.

Search queries were designed to address infectious disease management in livestock (pigs, poultry, and cattle), ontology-based methods, and challenges in livestock health management. Keywords included "Health Ontology,” “Livestock Health Ontology," "Disease Surveillance," "Precision Livestock Farming," "Farming," "Agriculture," "Animal Health Surveillance,", "Health Surveillance,” “Infectious Diseases," "Disease Management," "Farm Management,” and "Data Management Challenges."

PubMed Search Queries and Keywords used

Searches conducted: December 2024. Date range filter: 2011–2024. Text availability: Abstract.

Query 1: ("Livestock Health Ontology" OR "Disease Surveillance" OR "Animal Health Surveillance" OR "Early Detection" OR "Infectious Diseases") AND ("Farm Animals" OR "Cattle" OR "Poultry" OR "Pig" OR "Salmon") AND (Agriculture OR "Animal Agriculture") AND "Ontology". This query targeted studies focusing on the role of ontologies in livestock health management, including disease surveillance and early detection for farm animals.

Query 2: (("ontologies" OR "ontology-based" ) AND ("animal" OR "livestock"  ) AND ("disease surveillance" OR "health surveillance" OR "Animal Trait " OR "infectious diseases")). This query explored ontology-based approaches and their application in disease surveillance for livestock species.

Query 3 : ("precision livestock farming" OR "PLF" OR "livestock management systems") AND ("farm management" OR "disease surveillance" OR "disease management") AND ("challenges" OR "barriers" OR "issues") AND ("pigs" OR "poultry" OR "cattle" OR "Salmon"). This query focused on identifying challenges and barriers in livestock management systems and PLF technologies for disease management.

Query 4: ("farm Health data management systems" OR "agriculture health data management" OR "livestock management systems") AND ("data access" OR "data sharing" OR "data governance" OR "data integration" OR "data privacy") AND ("challenges" OR "barriers" OR "issues"). This query examined challenges related to health data management in livestock farming, including issues of data sharing, governance, and privacy.

Query5: ("livestock health" OR "animal health" OR " Disease Surveillance") AND ("ontology" OR "ontologies") AND ("challenges" OR "limitations" OR "barriers"). This query focused on the challenges and limitations of ontology-based approaches in livestock health and disease surveillance.

After combining results from all five queries and removing within-database duplicates, 102 unique PubMed records were retained.

IEEE Xplore Search Queries

 Searches conducted: December 2024, Filter applied: 2011 - 2024

IEEE Xplore queries were modified slightly for specific focus areas:

Query 1: Ontology for animal infectious diseases. This query targeted studies addressing the development and use of ontologies for managing infectious diseases in animals.

Query 2: Livestock Farming ontology. This query focused on ontology applications in livestock farming.

Query 3: Livestock Farming data challenges. This query explored challenges related to data in livestock farming, including integration, governance, and privacy.

After combining results from all three queries and removing within-database duplicates, 112 unique IEEE Xplore records were retained.

Other methods: In addition to structured database searches, 80 records were identified through other methods, including the authors' domain expertise in livestock ontologies and animal health informatics, reference lists of key publications in the field, and recommendations from co-authors and collaborators. They include publications on topics such as aquatic species health (e.g., salmon disease surveillance) and cross-domain ontology applications that may not have been captured by the structured database queries alone. Following PRISMA 2020 guidance, these records are reported as identified through "other methods" using the dual-pathway identification framework.

Post-Search Processing and Analysis

Records from all three sources (PubMed, IEEE Xplore, and other methods) were compiled using the Zotero reference manager and exported as a single RIS file. Cross-source duplicates (records appearing in more than one source) were identified and removed, resulting in the removal of 8 duplicate records.The final dataset contained 286 unique records for screening (PubMed: 102 + IEEE Xplore: 112 + Other methods: 80 = 294, minus 8 cross-source duplicates = 286). The RIS file, titled "Collected papers 286 - A Review on the Role of Ontologies in Modern Agriculture.ris", contains the bibliographic data for the 286 records that were screened after removing duplicates. This file can be loaded into reference managers or ASReview for systematic screening and analysis.

Note: Individual search queries within a database may return overlapping results; the per-database totals reported above (102 and 112) reflect the unique records after within-database deduplication. The complete dataset of 286 records is available on Zenodo (https://doi.org/10.5281/zenodo.15397012).

ASReview Project File Details

To streamline the review process, the ASReview LAB tool was utilized. The project file " a-review-on-the-role-of-ontologies.asreview" contains all the review settings, records, and inclusion/exclusion decisions. The Prior knowledge to warmup the AI is 28 relevant and 10 irrelevant records

 Researchers can replicate the review process by following these steps.

1.     Install ASReview:
Run the command: pip install asreview.

2.     Open ASReview lab:
Start the interface by typing asreview lab in the command prompt.

3.     Import the project file:
Load the project file into the ASReview LAB interface to view the screening process and decisions made during the review.

4.     Export Relevant datasets/records:
Use the export section to download records for further analysis.

The following datasets were generated during the review process using ASReview LAB:

1. asreview_dataset_all_a-review.ris/xlsx:
   This dataset contains all 286 records, including both labeled and unlabeled data.
2. asreview_dataset_relevant_a-review.ris/xlsx:
   This dataset includes only the 100 records identified as relevant during the review process.

These datasets provide a comprehensive view of the reviewed literature and can be used for further analysis or replication of the study. These resources are available on Zenodo https://doi.org/10.5281/zenodo.18817798 .

ASReview Analytics

The ASReview LAB tool generated analytics and visualizations to summarize the review process. Screenshots of these analytics, such as Supplementary Figure F1: ASReviewAnalytics.png shows relevant or irrelevant papers with percentage of relevant papers. Supplementary Figure F2 ASReviewLABprogressDensity.png shows number of review records, and Supplementary Figure F3 ASReviewLABprogressRecall.png provide further insights into the systematic review process.

Supplementary Figure F1: ASReview analytics.png

Supplementary Figure F2: Progress density.png

Supplementary Figure F3: progress recall.png

Metadata of Active Ontologies

This table provides additional metadata for the 15 active ontologies identified in this review (Table 1), including repository URLs, data formats, licenses, and last known update information.

Supplementary Table S1: Metadata of Active Ontologies