← Back

Artificial Intelligence Applications in Poultry Health and Welfare Monitoring: A Systematic Review and Meta-analysis of Model Performance

Authors
Pouneh Hajipour 2 iD
Affiliations
1Department of Viral Diseases, SANA Institute for Avian Health and Diseases Research, Tehran, Iran
2Department of Avian Diseases, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran
Overview

Abstract

Artificial intelligence (AI), machine learning, computer vision, sensor-based systems, and Internet of Things (IoT) technologies are increasingly used for automated monitoring of poultry health and welfare, but reported model performance varies across tasks, data sources, validation strategies, and production settings. This systematic review and meta-analysis evaluated the reported performance of AI-based models for poultry health and welfare monitoring. PubMed/MEDLINE, Scopus, and Web of Science were searched for peer-reviewed studies involving poultry, reporting health- or welfare-related monitoring outcomes, and providing extractable model-performance data. Data were extracted independently by two reviewers, and pooled accuracy was estimated using a Restricted Maximum Likelihood random-effects model in STATA version 17. Twenty-one studies published between 2012 and 2025 were included, of which ten were eligible for meta-analysis. Included systems used convolutional neural networks, random forests, support vector machines, and other machine-learning approaches for behavior recognition, disease detection, lesion assessment, mortality detection, environmental monitoring, and welfare assessment. The pooled accuracy estimate was 90.39% (95% CI: 84.89–95.89; p<0.001), with substantial heterogeneity among studies. Reported performance varied by application domain, data source, model type, assessment dimension, sample size, and integration with IoT-based systems. AI-based technologies show promise for automated poultry health and welfare monitoring; however, heterogeneous methods and limited external validation restrict the generalizability of pooled estimates. Future studies should prioritize standardized reporting, open datasets, external validation, and testing under commercial farm conditions.

Artificial intelligenceMachine learningPoultryWelfareHealth monitoringPrecision livestock farming

Introduction

Continuous monitoring in poultry farms is essential for the timely identification of abnormal behavior, environmental stress, and early signs of disease (Fan, Peng, and Yang 2025; Cruz et al. 2025). Due to the high density and flocks' sensitivity to environmental changes and diseases, early detection is crucial; delays in identifying problems can lead to mortality and economic losses. Early detection and timely alarms enable rapid intervention, reduce mortality, and improve welfare and productivity. Traditional monitoring methods entail high costs, require significant staff, and entail the risk of human mistake . The assessments are often time-consuming, labor-intensive, and costly, especially in large-scale poultry . In addition, manual evaluations are highly subjective, and inter- or intra-observer variation or errors may affect the assessments . Artificial intelligence is an effective solution to these limitations by enabling continuous, automated, and objective monitoring of poultry .

Artificial intelligence and machine learning technologies are increasingly used across all fields . Poultry production, health surveillance, welfare assessment, growth prediction, and automated decision-making are areas where artificial intelligence provides accurate assessments (Srinivasagan et al. 2025; Kalita et al. 2025). The literature has reported high performance of AI applications in poultry. Tasks such as disease detection from chicken images, behavioral monitoring through video analysis, sound-based stress recognition, and prediction of growth or productivity outcomes or outbreaks are common applications of AI in modern poultry . However, due to rapid expansion in this field, the evidence-based review remains unstructured. The literature on the application of AI in poultry management varies widely in sample sizes, data types, analytical models, validation methods, and reporting, which makes it difficult to draw consistent conclusions about overall AI performance. Although previous reviews have summarized AI, IoT, and computer vision applications in poultry production, a quantitative synthesis of reported model performance estimates across poultry welfare and health monitoring tasks remains limited. The aim of this systematic review and meta-analysis was to evaluate the reported performance of AI- and machine-learning-based models for poultry welfare and health monitoring, with particular attention to model accuracy, heterogeneity, application domains, data sources, and validation-related limitations.

Material and Methods

Reporting Framework

This study was designed as a systematic review and meta-analysis of peer-reviewed studies evaluating AI-based methods for poultry welfare and health monitoring. The review followed the PRISMA 2020 reporting guideline. The review protocol was not prospectively registered; eligibility criteria and the synthesis plan were defined prior to screening.

The research question was structured according to the PICOS framework. The population comprised poultry, including broilers, layers, breeders, and chicks. The intervention or index technology included artificial intelligence-based, machine-learning, deep-learning, computer-vision, sensor-based, Internet of Things, or related computational monitoring systems. A comparator was not mandatory because many eligible AI-monitoring studies reported model performance against reference labels or validation datasets rather than against a conventional monitoring method. The outcomes of interest were model-performance metrics and poultry welfare- or health-related monitoring outcomes, including accuracy, sensitivity, specificity, precision, area under the curve, behavior recognition, disease detection, lesion assessment, mortality detection, and environmental or welfare monitoring. Eligible study designs were peer-reviewed original studies reporting extractable AI model-performance data; reviews, editorials, commentaries, conference abstracts, book chapters, studies without full-text availability, and non-English studies were excluded.

Search Strategy

This systematic review was conducted using PubMed/MEDLINE, Scopus, and Web of Science. The objective was to identify peer-reviewed studies evaluating AI-based approaches for poultry welfare and health monitoring. The search strategy followed the PRISMA 2020 guidelines (Page et al. 2021; Sadeghi et al. 2023) and combined poultry-related terms, including “broiler,” “layer,” “chicken,” “poultry,” and “farm,” with AI-related terms, including “artificial intelligence,” “machine learning,” “deep learning,” “neural networks,” “computer vision,” “image processing,” “automated detection,” “behavior analysis,” “prediction,” and “precision agriculture.” Boolean operators were used to combine terms within each concept block using “OR” and to combine poultry- and AI-related blocks using “AND.” No date restrictions were applied. All records were imported into EndNote and Rayyan for deduplication and screening.

Inclusion and exclusion criteria

Studies were included if they met the following criteria: (1) they involved poultry, including broilers, layers, breeders, or chicks; (2) they applied AI, machine learning, deep learning, computer vision, sensor-based, or related computational methods; (3) they reported poultry welfare- or health-related monitoring outcomes; and (4) they were published in English in peer-reviewed journals. Studies were excluded if they did not provide extractable performance data, were reviews, editorials, commentaries, conference abstracts, or book chapters, lacked full-text availability, or were not published in English.

Study selection and data extraction

The data extraction process was independently conducted by two reviewers using Rayyan. At the first step, the titles, abstracts, and keywords of all included records were screened. In the next step, the full texts of potentially eligible studies were assessed. The data were extracted by using a standardized Excel extraction sheet. The extracted variables including AI or machine learning methods, data sources, sample size, model performance metrics, and relevant outcomes. The discrepancies between reviewers were resolved through group discussion and adjudicated by the corresponding author.

Quality assessment of included studies

The methodological quality and reporting completeness of the included studies were assessed using a structured appraisal framework adapted for AI-based animal monitoring studies. Each study was evaluated across the following domains: clarity of dataset description; poultry population and production setting; data source and acquisition method; ground-truth or labeling procedure; model development approach; validation strategy; reporting of performance metrics; risk of overfitting; real-world applicability; and reproducibility. Particular attention was given to whether studies used independent test datasets, cross-validation, or external validation, and whether model performance was evaluated under commercial or field conditions. Each domain was rated as low, moderate, or high concern. Disagreements between reviewers were resolved by discussion, and unresolved disagreements were adjudicated by the corresponding author.

Statistical analysis

STATA Version 17 (StataCorp, College Station, Texas, USA) was used for statistical analysis. The primary outcome measure was the accuracy of the models. The Restricted Maximum Likelihood (REML) method was used to calculate the pooled effect sizes and their 95% of confidence intervals (CIs). In addition, the heterogeneity of the included studies was assessed with the Higgins I-squared (I²) statistic and the between-study variance estimate (τ²). The publication bias of the included studies was visually assessed using funnel plots, as well as with the Begg's rank correlation test and Egger's linear regression test. In addition, the trim-and-fill test was used to estimate the possible impact of missing studies on the pooled effect size. All analyses were conducted as two-tailed, and a p-value less than 0.05 was statistically considered significant.

Results

In this systematic review, 21 studies were included. The included studies, published between 2012 and 2025, focused on poultry health and welfare monitoring (Figure 1). The characteristics of included studies were reported in Supplementary Table S1. The study period ranged from short-term experimental assessments (a few days) to long-term observational studies lasting up to 3.5 years. The geographical distribution of the included studies spanned across the world, including Europe (Netherlands, Belgium, UK, Italy), Asia (China, Saudi Arabia, Republic of Korea, Japan, Thailand), Africa (Egypt, Ethiopia), North America (USA), and South America (Brazil).

Overall, the included studies varied considerably in methodological quality and completeness of reporting. Most studies described the AI model architecture and input data type, but reporting of ground-truth labeling procedures, external validation, dataset availability, and reproducibility was inconsistent. Several studies relied on controlled or curated datasets, whereas fewer evaluated model performance under commercial farm conditions. These quality-related limitations were taken into account when interpreting the pooled estimates, particularly given the substantial heterogeneity across studies. The detailed study-level quality assessment is presented in Supplementary Table S2.

Figure 1PRISMA 2020 flow diagram of study identification, screening, eligibility assessment, and inclusion.
C:\Users\IRANDO~1.K\AppData\Local\Temp\Rar$DIa0.120\Figure 1.PNG

The included studies assessed different domains of poultry health and welfare monitoring. These domains include behavior classification and movement algorithm (Shimmura et al. 2024; Rocha Merenda et al. 2024; Jaihuni et al. 2023; De Montis et al. 2013; Guo et al. 2020), leg health assessment (Fodor et al. 2025; Aydin 2017), lesion and bone health detection (Castro Júnior et al. 2022; Kaewtapee et al. 2022; Vanderhasselt et al. 2013), disease identification (Degu and Simegn 2023; Huang, Wang, and Zhang 2019; Elmessery et al. 2023), dead-bird detection and inactivity monitoring (Bao et al. 2021; Khanal, Wu, and Lee 2024), distress recognition and welfare assessment (Srinivasagan et al. 2025; Du et al. 2020), and production performance prediction in breeder farms (Ji, Xu, and Teng 2025).

Also, the sample sizes and dataset scales varied across studies, ranging from small laboratory cohorts (Huang, Wang, and Zhang 2019) to large commercial datasets of flock-level production records (Hepworth et al. 2012; Ji, Xu, and Teng 2025). The populations under study were included broilers (Fodor et al. 2025; Ji, Xu, and Teng 2025; De Montis et al. 2013; Rocha Merenda et al. 2024; Jaihuni et al. 2023; Aydin 2017; Hepworth et al. 2012; Bao et al. 2021; Khanal, Wu, and Lee 2024; Guo et al. 2020; Elmessery et al. 2023; Kaewtapee et al. 2022; Vanderhasselt et al. 2013), laying hens (Huang, Wang, and Zhang 2019; Shimmura et al. 2024; Srinivasagan et al. 2025), broiler breeders (Ji, Xu, and Teng 2025), and newly hatched chicks (Goldman and Wood 2015). Data sources were collected by CCTV video recordings (Fodor et al. 2025; Shimmura et al. 2024; Jaihuni et al. 2023; De Montis et al. 2013; Khanal, Wu, and Lee 2024; Guo et al. 2020; Elmessery et al. 2023; Vanderhasselt et al. 2013), wearable inertial sensors (Bao et al. 2021; Shimmura et al. 2024), environmental IoT sensors measuring temperature, humidity, and THI (Ji, Xu, and Teng 2025), farm management and production records (Ji, Xu, and Teng 2025; Hepworth et al. 2012), and bioacoustics recordings (Srinivasagan et al. 2025; Du et al. 2020; Huang, Wang, and Zhang 2019).

In terms of the modeling methods, the included studies used machine learning (ML), deep learning (DL), computer vision (CV), TinyML, and explainable AI frameworks. In addition, the YOLO variants and CNN-based models were used for Vision-based assessment systems (Fodor et al. 2025; Rocha Merenda et al. 2024; Jaihuni et al. 2023; Khanal, Wu, and Lee 2024; Guo et al. 2020; Degu and Simegn 2023; Elmessery et al. 2023; Kaewtapee et al. 2022; Vanderhasselt et al. 2013). Also, the MFCC, SVM or 1D CNN were used for audio-based monitoring approaches (Srinivasagan et al. 2025; Du et al. 2020; Huang, Wang, and Zhang 2019; Fontana et al. 2017). In term of the environmental and production, the datasets were analyzed by using Random Forest, XGBoost, LightGBM, SVR, MLP, ANN, and kNN (Hepworth et al. 2012; Ji, Xu, and Teng 2025; Shimmura et al. 2024; Castro Júnior et al. 2022; Aydin 2017; Bao et al. 2021). The most common development platforms were python-based environments (TensorFlow, PyTorch, scikit-learn, OpenCV) (Srinivasagan et al. 2025; Fodor et al. 2025; Ji, Xu, and Teng 2025; Rocha Merenda et al. 2024; Jaihuni et al. 2023; Khanal, Wu, and Lee 2024; Guo et al. 2020; Degu and Simegn 2023), MATLAB (Castro Júnior et al. 2022; Du et al. 2020; Aydin 2017; Fontana et al. 2017), Edge Impulse (Srinivasagan et al. 2025; Bao et al. 2021), TensorFlow Lite Micro, and Darknet-based implementations (Degu and Simegn 2023).

Also, the performance metrics, the accuracy, sensitivity, specificity, precision, and AUC, are reported in Table 1. The meta-analysis included 10 studies and reported a pooled estimate of accuracy for health & welfare monitoring of 90.39% (95% CI: 84.89-95.89; p<0.001). Heterogeneity across studies was statistically significant (τ²=69.2591; I²=99.86%; p<0.001). The CNN model achieves the best performance, particularly in AUC and behavior detection task accuracy. Also, some studies report that TinyML for real-time audio classification enables early detection of poultry distress or abnormal behavior, and the XGBoost predicts production performance in broiler breeders.

The regression-based Egger test reports potential small-study effects (p=0.015); however, Begg's test reports no significant effect (p=1.0000). In addition, trim-and-fill reports show no change after potential imputation for missing data (Table 1, Figure 2).

Table 1Accuracy of artificial intelligence-based technologies in poultry welfare and health monitoring

Study / Summary

Effect Size (Accuracy %)

95% CI

Weight (%)

Srinivasagan et al.

96.60

95.01 – 98.19

11.25

Shimmura et al.

96.30

86.90 – 100

8.53

Castro Júnior et al.

73.00

67.19 – 78.81

10.08

Aydin et al.

93.00

89.84 – 96.16

10.95

Goldman and Wood

65.00

41.63 – 88.37

3.72

Guo et al.

95.44

92.52 – 98.36

11.01

Degu and Simegn

98.70

98.48 – 98.92

11.36

Huang et al.

90.00

88.53 – 91.47

11.27

Kaewtapee et al.

82.40

77.62 – 87.18

10.46

Bao et al.

95.60

95.40 – 95.80

11.36

Pooled Effect (Random-Effects Model)

90.39

84.89 – 95.89

—

Heterogeneity: τ² = 69.2591; I² = 99.86%; H² = 720.83

C:\Users\IRANDO~1.K\AppData\Local\Temp\Rar$DIa0.004\Figure 2.jpgFigure 2. Forest plot of studies assessing artificial intelligence-based poultry welfare and health monitoring.

Table 2The results of the subgroup analysis of AI and technologies in Welfare and Health monitoring

Subgroup Analysis

Group

studies

Effect size

95% CI

p for pooled estimate

Sample Size

under 200

4

82.98

70.03-95.94

<0.001

over 200

4

92.22

86.17-98.26

<0.001

Domain of AI Application

Welfare

7

87.83

79.92-95-74

<0.001

Disease/Mortality

3

94.81

89.86-99.77

<0.001

Assessment Dimension

Whole-Animal Monitoring

7

93.96

91.57-96.36

<0.001

Anatomical / Biological Substra

3

84.93

70.11-99.75

<0.001

Data Source Type

field/real world

6

93.30

89.13-97.48

<0.001

controlled curated

4

83.85

69.66-98.03

<0.001

AI Model Type

deep learning

3

92.82

83.02-100.00

<0.001

classical ML/non DL

7

89.03

81.77-96.29

<0.001

Data Acquisition Strategy (Hybrid vs. Fully Automated AI)

Hybrid Systems (AI + IoT)

3

95.77

95.05-96.50

<0.001

Just AI

7

87.45

79.77-95.13

<0.001

Subgroup analyses were exploratory because of the small number of studies in each category and the substantial overall heterogeneity. Disease/mortality studies showed numerically higher pooled accuracy than welfare studies (94.81%, 95% CI: 89.86–99.76 vs. 87.83%, 95% CI: 79.92–95.74). Higher pooled estimates were also observed for larger datasets, whole-animal monitoring approaches, field/real-world datasets, deep-learning models, and hybrid AI+IoT systems. However, most between-subgroup differences were not statistically significant, and these findings should be interpreted as hypothesis generating rather than confirmatory. The clearest subgroup difference was observed for data acquisition strategy, where hybrid AI+IoT systems showed higher pooled accuracy than AI-only systems.

Figure 3Forest plot of subgroup analyses of artificial intelligence-based poultry welfare and health monitoring.
C:\Users\IRANDO~1.K\AppData\Local\Temp\Rar$DIa0.756\figure 3.jpg

Discussion

The aim of this systematic review was to evaluate the application of Artificial Intelligence (AI) and IoT for monitoring poultry welfare and health. The results showed a significant increase in research focused on integrating Artificial Intelligence (AI) and technologies for poultry monitoring, especially after 2017. This increase in publications can be attributed to rapid advancements in machine learning algorithms and the growing use of technologies such as cameras, sensors, and wearable devices. These advancements enable poultry to collect real-time data more efficiently and to develop automated systems that monitor poultry welfare and health. In addition, this trend suggests the industry's rising demand for more efficient, automated solutions to address challenges such as labor shortages, disease prevention, and welfare management (Yang et al. 2024; Attia et al. 2024). The studies in this review used various platforms, including Python, MATLAB, and TensorFlow, for model development and implementation. The combination of these platforms with IoT sensors, wearable devices, and cameras enables real-time data collection, which is crucial for monitoring poultry management (Lin and Suhendra 2025; Modak et al. 2025).

The time range of methodologies used in studies spans from short-term experimental setups to long-term observational studies that assess the sustained impacts of AI and IoT technologies. The meta-analysis showed a pooled accuracy estimate of 90.39% for artificial intelligence-based poultry health and welfare monitoring. The high accuracy of AI models in poultry monitoring suggests significant benefits for poultry management and for early detection of health issues and behavioral changes . This enables farmers and farm managers to monitor in real time, identify distress signals or early symptoms of disease, and intervene promptly, reducing the need for costly medical treatments or large-scale interventions (Cruz et al. 2025; Naeem et al. 2025; Abdel-Wareth et al. 2025).

Also, since these models can monitor environmental factors such as temperature, humidity, and ventilation in addition to chicken assessments (Pereira et al. 2020; Godinho et al. 2025; Elwakeel 2025), the use of AI models helps achieve sustainable poultry production. Using AI to automate tasks reduced staff errors and costs to a minimum, enabling farmers to implement a 24/7 monitoring program . This transition to AI-powered, data-driven systems enables us to collect longitudinal, high-coverage, reliable, and valid poultry farming data with high efficiency (Essien and Neethirajan 2025; Kumari et al. 2025). These data enable accurate decision-making and management, improve overall poultry welfare and farm productivity, support sustainable farming practices, and help to solve today's agricultural challenges, such as staff shortages, resource optimization, and environmental sustainability (Makapela, Alexander, and Tshelane 2025; Niloofar et al. 2021; Sajid et al. 2024).

To maximize the benefits of AI and IoT technologies in large poultry operations, it is recommended to focus on scalable, integrated systems capable of handling large volumes of real-time data . Implementing AI-powered farm management tools can help reduce labor costs and ensure consistent welfare standards. In addition, emerging models such as TinyML and edge computing are gaining attention for their ability to enable real-time data processing and early detection of poultry health issues (Srinivasagan et al. 2025; Bhattad et al. 2025). Future research should focus on integrating advanced AI models, improving data quality, classifying disease severity in individual chickens, and exploring the long-term impacts of AI and IoT applications on poultry welfare and farm sustainability.

While AI implementation has a high impact, implementing AI and IoT models in poultry comes at a high cost. Primarily implementing AI; need advanced hardware, such as IoT devices, sensors, AI-powered systems, and monitoring systems. Also, it needs specialized training staff or consultants to operate and interpret data from these advanced technologies. In addition, some ongoing expenses for system maintenance, software updates, and data storage require expensive investments that can be a barrier for many farm (Leong 2024; Issa et al. 2024; Natsir et al. 2025). However, automation management reduces labor costs and minimizes human error. These systems enable better resource optimization, improving energy efficiency, and minimizing environmental impact. Over time, these technologies lead to higher productivity and healthier poultry, ultimately resulting in greater profitability and long-term financial improvements and efficiencies.

This systematic review has several limitations. The first limitation of this systematic review is that it did not assess potential barriers to the use of AI and IoT technologies in poultry farming, such as the financial burdens associated with implementation, maintenance, and training. The studies included in this review varied in their methodologies, including sample sizes, data sources, study duration, and analytical methods. This variation may lead to excessive heterogeneity and reduced generalizability of the findings. In addition, although the review used statistical methods to assess publication bias and heterogeneity, potential small-study effects or missing studies could still affect the pooled estimates.

Conclusion

In conclusion, AI- and IoT-enabled systems achieve high reported accuracy in poultry welfare and health monitoring, particularly in automated, sensor-integrated applications. However, the evidence remains heterogeneous, and the generalizability of pooled model performance is limited by differences in datasets, monitoring tasks, validation approaches, and reporting standards. Future studies should prioritize standardized performance reporting, external validation, open datasets, and evaluation under commercial farm conditions before broad implementation.

Conflict of Interest

The authors declare no conflicts of interest.

Funding

This research received no external funding.

Data Availability Statement

The data extracted and analyzed in this study are included in the article and Supplementary Materials. Additional information is available from the corresponding author upon reasonable request.

Ethics Statement

This article is a review/meta-analysis and did not involve any new experiments on animals or humans.

Author Contributions

M.J: Conceptualization, Investigation, Data curation, Formal analysis, Resources, Validation, Project administration, Supervision, Writing – original draft. P.H: Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing.

Declaration of Generative AI and AI-assisted technologies

The authors declare that generative AI and AI-assisted technologies were only used to assist with data extraction and to improve grammar, language, and clarity of the manuscript. These tools were not used for the generation or interpretation of scientific content, and the authors take full responsibility for the accuracy, integrity, and originality of the work.

References

  • Abdel-Wareth, Ahmed AA, Ahmed Ahmed, Md Salahuddin, and Jayant Lohakare. 2025. Application of Artificial Intelligence in Rabbit Husbandry: From Reproductive Moni-toring to Precision Farming. Frontiers in Veterinary Science 12: 1679630. https://doi.org/10.3389/fvets.2025.1679630
  • Attia, Youssef A., Ahmed K. Aldhalmi, Islam M. Youssef, Fulvia Bovera, Vincenzo Tufarelli, Mohamed E. Abd El-Hack, et al. 2024. Climate change and its effects on poultry industry and sustainability. Discover Sustainability 5 (1): 397. 10.1007/s43621-024-00627-2
  • Aydin, Arda. 2017. Using 3D vision camera system to automatically assess the level of inactivity in broiler chickens. Computers and Electronics in Agriculture 135: 4-10. https://doi.org/10.1016/j.compag.2017.01.024
  • Bao, Yiqin, Hongbing Lu, Qiang Zhao, Zhongxue Yang, Wenbin Xu, and Y Bao. 2021. Detection system of dead and sick chickens in large scale farms based on artificial intelligence. Math. Biosci. Eng 18 (5): 6117-6135. https://doi.org/10.3934/mbe.2021306
  • Bhattad, Sejal, Ahmed Abdelmoamen Ahmed, Ahmed A. A. Abdel-Wareth, and Jayant Lohakare. 2025. An IoT-Based System for Measuring Diurnal Gas Emissions of Laying Hens in Smart Poultry Farms. AgriEngineering 7 (8): 267. https://www.mdpi.com/2624-7402/7/8/267
  • Castro Júnior, Sérgio L de, Iran JO da Silva, Aérica C Nazareno, and Mariana de O Mota. 2022. Computer vision for morphometric evaluation of broiler chicken bones. Engenharia Agrícola 42: e20210150. https://doi.org/10.1590/1809-4430-Eng.Agric.v42nepe20210150/2022
  • Cruz, Edmanuel, Miguel Hidalgo-Rodriguez, Adiz Mariel Acosta-Reyes, José Carlos Rangel, Keyla Boniche, and Franchesca Gonzalez-Olivardia. 2025. ACMSPT: Automated Counting and Monitoring System for Poultry Tracking. AgriEngineering 7 (3): 86. https://www.mdpi.com/2624-7402/7/3/86
  • De Montis, A, A Pinna, M Barra, and E Vranken. 2013. Analysis of poultry eating and drinking behavior by software eYeNamic. Journal of Agricultural Engineering 44 (s2). https://doi.org/10.4081/jae.2013.275
  • Degu, Mizanu Zelalem, and Gizeaddis Lamesgin Simegn. 2023. Smartphone based detection and classification of poultry diseases from chicken fecal images using deep learning techniques. Smart Agricultural Technology 4: 100221. https://doi.org/10.1016/j.atech.2023.100221
  • Du, Xiaodong, Lenn Carpentier, Guanghui Teng, Mulin Liu, Chaoyuan Wang, and Tomas Norton. 2020. Assessment of laying hens’ thermal comfort using sound technology. Sensors 20 (2): 473. https://doi.org/10.3390/s20020473
  • Elmessery, Wael M, Joaquín Gutiérrez, Gomaa G Abd El-Wahhab, Ibrahim A Elkhaiat, Ibrahim S El-Soaly, Sadeq K Alhag, et al. 2023. YOLO-based model for automatic detection of broiler pathological phenomena through visual and thermal images in intensive poultry houses. Agriculture 13 (8): 1527. https://doi.org/10.3390/agriculture13081527
  • Elwakeel, Abdallah Elshawadfy. 2025. A smart automatic control and monitoring system for environmental control in poultry houses integrated with earlier warning system. Scientific Reports 15 (1): 31630. 10.1038/s41598-025-17074-2
  • Essien, Daniel, and Suresh Neethirajan. 2025. Multimodal AI systems for enhanced laying hen welfare assessment and productivity optimization. Smart Agricultural Technology 12: 101564. 10.1016/j.atech.2025.101564
  • Fan, Weiqin, Hui Peng, and Diqi Yang. 2025. Review: The application and challenges of advanced detection technologies in poultry farming. Poultry Science 104 (11): 105870. 10.1016/j.psj.2025.105870
  • Fodor, István, Marjaneh Taghavi, Esther D Ellen, and Malou van der Sluis. 2025. Top-view characterization of broiler walking ability and leg health using computer vision. Poultry Science 104 (2): 104724. https://doi.org/10.1016/j.psj.2024.104724
  • Fontana, Ilaria, Emanuela Tullo, Lenn Carpentier, Dries Berckmans, Andy Butterworth, Erik Vranken, Tomas Norton, Daniel Berckmans, and Marcella Guarino. 2017. Sound analysis to model weight of broiler chickens. Poultry Science 96 (11): 3938-3943. https://doi.org/10.3382/ps/pex215
  • Godinho, António, Romeu Vicente, Sérgio Silva, and Paulo Jorge Coelho. 2025. Wireless Environmental Monitoring and Control in Poultry Houses: A Conceptual Study. IoT 6 (2): 32. https://www.mdpi.com/2624-831X/6/2/32
  • Goldman, Jason G, and Justin N Wood. 2015. An automated controlled-rearing method for studying the origins of movement recognition in newly hatched chicks. Animal Cognition 18 (3): 723-731. https://doi.org/10.1007/s10071-015-0839-3
  • Guo, Yangyang, Lilong Chai, Samuel E. Aggrey, Adelumola Oladeinde, Jasmine Johnson, and Gregory Zock. 2020. A Machine Vision-Based Method for Monitoring Broiler Chicken Floor Distribution. Sensors 20 (11): 3179. 10.3390/s20113179
  • He, Pengguang, Zhonghao Chen, Hongwei Yu, Khawar Hayat, Yefan He, Jinming Pan, et al. 2022. Research Progress in the Early Warning of Chicken Diseases by Monitoring Clinical Symptoms. Applied Sciences 12 (11): 5601. https://doi.org/10.3390/app12115601
  • Hepworth, Philip J, Alexey V Nefedov, Ilya B Muchnik, and Kenton L Morgan. 2012. Broiler chickens can benefit from machine learning: support vector machine analysis of observational epidemiological data. Journal of the Royal Society Interface 9 (73): 1934-1942. https://doi.org/10.1098/rsif.2011.0852
  • Huang, Junduan, Wenqing Wang, and Tiemin Zhang. 2019. Method for detecting avian influenza disease of chickens based on sound analysis. Biosystems engineering 180: 16-24. https://doi.org/10.1016/j.biosystemseng.2019.01.015
  • Issa, Ali Ashoor, Safa Majed, Abdul Ameer, and Hassan M Al-Jawahry. 2024. IoT and AI in livestock management: A game changer for farmers. E3S Web of Conferences. https://doi.org/10.1051/e3sconf/202449102015
  • Jaihuni, Mustafa, Hao Gan, Tom Tabler, Maria Prado, Hairong Qi, and Yang Zhao. 2023. Broiler mobility assessment via a semi-supervised deep learning model and neo-deep sort algorithm. Animals 13 (17): 2719. https://doi.org/10.3390/ani13172719
  • Jebari, Hakim, Meriem Hayani Mechkouri, Siham Rekiek, and Kamal Reklaoui. 2023. Poultry-Edge-AI-IoT System for Real-Time Monitoring and Predicting by Using Artificial Intelligence. Int. J. Interact. Mob. Technol. 17 (12): 149-170. https://doi.org/10.3991/ijim.v17i12.38095
  • Ji, Hengyi, Yidan Xu, and Guanghui Teng. 2025. Predicting egg production rate and egg weight of broiler breeders based on machine learning and Shapley additive explanations. Poultry Science 104 (1): 104458. https://doi.org/10.1016/j.psj.2024.104458
  • Kaewtapee, C, S Thepparak, C Rakangthong, C Bunchasak, and A Supratak. 2022. Objective scoring of footpad dermatitis in broiler chickens using image segmentation and a deep learning approach: camera-based scoring system. British poultry Science 63 (4): 427-433. https://doi.org/10.1080/00071668.2021.2013439
  • Kalita, Arnab Jyoti, Mirash Subba, Sheikh Adil, Manzoor A Wani, Yasir Afzal Beigh, and Majid Shafi. 2025. Application of artificial intelligence and machine learning in poultry disease detection and diagnosis: A review: Ai and machine learning in poultry disease diagnosis. Letters In Animal Biology 5 (1): 01-06. https://doi.org/10.62310/liab.v5i1.155
  • Khanal, Ridip, Wenqin Wu, and Joonwhoan Lee. 2025. Automated Dead Chicken Detection in Poultry Farms Using Knowledge Distillation and Vision Transformers. Applied Sciences 15 (1): 136. https://doi.org/10.3390/app15010136
  • Kumari, Karishma, Ali Mirzakhani Nafchi, Salman Mirzaee, and Ahmed Abdalla. 2025. AI-Driven Future Farming: Achieving Climate-Smart and Sustainable Agriculture. AgriEngineering 7 (3): 89. https://www.mdpi.com/2624-7402/7/3/89
  • Leong, W. Y., Leong, Y. Z., & Leong, W. S. (2025). IoT and smart manufacturing for poultry farm. In T.-H. Meen, C.-F. Yang, & C.-Y. Chang (Eds.), Proceedings of the 7th International Conference on Knowledge Innovation and Invention, Volume 2 (pp. 205–213). Springer. 10.1007/978-981-95-1941-5_22
  • Li, N., Z. Ren, D. Li, and L. Zeng. 2020. Review: Automated techniques for monitoring the behaviour and welfare of broilers and laying hens: towards the goal of precision livestock farming. Animal 14 (3): 617-625. 10.1017/s1751731119002155
  • Lin, Hao-Ting, and Suhendra. 2025. Development and Implementation of an IoT-Enabled Smart Poultry Slaughtering System Using Dynamic Object Tracking and Recognition. Sensors 25 (16): 5028. https://www.mdpi.com/1424-8220/25/16/5028
  • Makapela, Mzuhleli, Gregg Alexander, and Molaodi Tshelane. 2025. Enhancing Agricultural Productivity Among Emerging Farmers Through Data-Driven Practices. Sustainability 17 (10): 4666. https://www.mdpi.com/2071-1050/17/10/4666
  • Merenda, V. R., Bodempudi, V. U. C., Pairis-Garcia, M. D., & Li, G. (2024). Development and validation of machine-learning models for monitoring individual behaviors in group-housed broilers. Poultry Science, 103(12), 104374. 10.1016/j.psj.2024.104374
  • Michels, E. A. M., S. Gilbert, I. Koval, and M. K. Wekenborg. 2025. Alarm fatigue in healthcare: a scoping review of definitions, influencing factors, and mitigation strategies. BMC Nurs 24 (1): 664. 10.1186/s12912-025-03369-2
  • Modak, Mrinmoy, Muin Mustahasin Pritom, Sajal Chandra Banik, and Md Sanaul Rabbi. 2025. Internet of Things-Based Health Surveillance Systems for Livestock: A Review of Recent Advances and Challenges. IET Wireless Sensor Systems 15 (1): e70013. 10.1049/wss2.70013
  • Naeem, M., Z. Jia, J. Wang, S. Poudel, S. Manjankattil, Y. Adhikari, M. Bailey, and D. Bourassa. 2025. Advancements in machine learning applications in poultry farming: a literature review. Journal of Applied Poultry Research 34 (4): 100602. 10.1016/j.japr.2025.100602
  • Natho, P., S. Boonying, P. Bonguleaum, N. Tantidontanet, and L. Chamuthai. 2025. An enhanced machine vision system for smart poultry farms using deep learning. Smart Agricultural Technology 12: 101083. 10.1016/j.atech.2025.101083
  • Natsir, Muhammad Halim, Wayan Firdaus Mahmudy, Mochamad Tono, and Yuli Frita Nuningtyas. 2025. Advancements in artificial intelligence and machine learning for poultry farming: Applications, challenges, and future prospects. Smart Agricultural Technology: 101307. https://doi.org/10.1016/j.atech.2025.101307
  • Niloofar, Parisa, Deena P Francis, Sanja Lazarova-Molnar, Alexandru Vulpe, Marius-Constantin Vochin, George Suciu, et al. 2021. Data-driven decision support in livestock farming for improved animal health, welfare and greenhouse gas emissions: Overview and challenges. Computers and Electronics in Agriculture 190: 106406. https://doi.org/10.1016/j.compag.2021.106406
  • Ojo, Rasheed O., Anuoluwapo O. Ajayi, Hakeem A. Owolabi, Lukumon O. Oyedele, and Lukman A. Akanbi. 2022. Internet of Things and Machine Learning techniques in poultry health and welfare management: A systematic literature review. Computers and Electronics in Agriculture 200: 107266. 10.1016/j.compag.2022.107266
  • Page, Matthew J, Joanne E McKenzie, Patrick M Bossuyt, Isabelle Boutron, Tammy C Hoffmann, Cynthia D Mulrow, et al. 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372. https://doi.org/10.1136/bmj.n71
  • Pereira, Wariston Fernando, Leonardo da Silva Fonseca, Fernando Ferrari Putti, Bruno César Góes, and Luciana de Paula Naves. 2020. Environmental monitoring in a poultry farm using an instrument developed with the internet of things concept. Computers and Electronics in Agriculture 170: 105257. 10.1016/j.compag.2020.105257
  • Sadeghi, M., A. Banakar, S. Minaei, M. Orooji, A. Shoushtari, and G. Li. 2023. Early Detection of Avian Diseases Based on Thermography and Artificial Intelligence. Animals (Basel) 13 (14). 10.3390/ani13142348
  • Sajid, M. A., Akhtar, I., Rauf, U., Hafeez, S., Tahir, Y., Abdullah, M., Ather, N., Yasin, N., & Asghar, R. (2024). Smart sensors and robotics in poultry farming: Transforming operational efficiency—A review. Scholars Journal of Agriculture and Veterinary Sciences, 11(7), 133–145. 10.36347/sjavs.2024.v11i07.003
  • Shimmura, Tsuyoshi, Itsufumi Sato, Ryo Takuno, and Kaori Fujinami. 2024. Spatiotemporal understanding of behaviors of laying hens using wearable inertial sensors. Poultry Science 103 (12): 104353. https://doi.org/10.1016/j.psj.2024.104353
  • Srinivasagan, Ramasamy, Mohammed Shawky El Sayed, Mohammed Ibrahim Al-Rasheed, and Ali Saeed Alzahrani. 2025. Edge intelligence for poultry welfare: Utilizing tiny machine learning neural network processors for vocalization analysis. PloS One 20 (1): e0316920. https://doi.org/10.1371/journal.pone.0316920
  • Vanderhasselt, RF, Margot Sprenger, Luc Duchateau, and FAM Tuyttens. 2013. Automated assessment of footpad dermatitis in broiler chickens at the slaughter-line: Evaluation and correspondence with human expert scores. Poultry Science 92 (1): 12-18. https://doi.org/10.3382/ps.2012-02153
  • Yang, Xiao, Ramesh Bahadur Bist, Bidur Paneru, Tianming Liu, Todd Applegate, Casey Ritz, et al. 2024. Computer vision-based cybernetics systems for promoting modern poultry farming: a critical review. Computers and Electronics in Agriculture 225: 109339. https://doi.org/10.1016/j.compag.2024.109339