Introduction

In modern poultry production, disease threatens the health of poultry every year, which not only seriously affects the animal welfare of poultry but also directly results in serious economic losses. Early warning of poultry diseases is very important for poultry health. Avian diseases are highly diverse and can be classified based on several parameters. In Diseases, such as avian flu, infectious bronchitis, and Newcastle disease, pose significant threats to poultry health and economic stability. These diseases primarily affect the respiratory system, causing symptoms like coughing, sneezing, nasal discharge, and rales (abnormal breathing sounds). Early detection is crucial for effective management and reducing losses. Precision Poultry Farming (PPF) systems have emerged as vital tools for monitoring poultry health by integrating sensing, analysis, and intervention functions. They utilize advanced technologies like image processing, sound analysis, and artificial intelligence to automate disease detection, reducing the need for manual observation and enabling efficient management of large flocks (Neethirajan et al., 2017; Wolfert et al., 2017). PPF systems collect data noninvasively using sensors like microphones to monitor animal sounds such as coughs and sneezes. The syrinx in birds plays a key role in producing diverse vocal frequencies, which can be analysed for respiratory disease detection. Microphones are cost-effective tools that allow large-scale monitoring without being restricted by sight or light conditions. Studies have demonstrated the effectiveness of sound technology in identifying respiratory issues and evaluating poultry health through vocalization analysis (Chelotti et al., 2016; Carpentier et al., 2018; Mahdavian et al., 2020).

Automated systems equipped with acoustic signal processing and machine learning can detect anomalies early and alert farmers to potential outbreaks or equipment failures. This proactive approach enables veterinarians to intervene promptly, often before farmers notice symptoms. Such systems improve animal welfare by reducing disease-related losses and minimizing antibiotic usage (Carpentier et al., 2019; Liu et al., 2020). Respiratory diseases remain a critical concern in poultry farming due to their impact on both animal welfare and human health. By leveraging PPF technologies, farmers can enhance disease detection and ensure better flock management while addressing these challenges effectively (Aydin et al., 2017; Bishop et al., 2019; Zhuang et al., 2018; Fang et al., 2020).

How technologies play an important role in disease identification?

Diagnosing Newcastle Disease (ND), Bronchitis Virus (BV), and Avian Influenza (AI) based solely on clinical signs like diarrhoea or nasal discharge is unreliable, as these signs are not pathognomonic and can vary due to other factors. Rapid and accurate diagnosis is critical to minimizing economic losses and preventing disease spread (Rahimian et al., 2011). Diagnostic methods such as RT-PCR, real-time RT-PCR, and ELISA are commonly used to detect these diseases. While RT-PCR is effective for amplifying viral genomes, it can only identify one virus at a time. Multiplex RT-PCR addresses this limitation by detecting multiple viruses simultaneously, improving efficiency (Corman et al., 2013; Haryanto et al., 2013,). However, methods like ELISA are time-consuming, and many diagnostic approaches require skilled personnel and expensive equipment, limiting their practicality in some settings (Kataria et al., 1998; Soltan et al., 2016). Smart poultry management systems integrating automatic detection devices offer a promising solution. These systems can monitor symptoms like respiratory distress in real time and alert farmers to take preventive measures. For instance, isolating sick birds or vaccinating healthy ones can prevent disease spread. Precision Poultry Farming (PPF) facilitates faster detection and treatment, reducing the risk of outbreaks and enhancing poultry health management. Such advancements are crucial for sustainable poultry production.

Novel devices in diagnosis:

Chicken Boy is a lightweight, rail-mounted device that gathers data on ambient conditions such as humidity, temperature, airspeed, and CO₂ levels in poultry houses. It uses the animal itself as a sensor to quickly identify growth challenges. The robot takes thermographic images to distinguish between live and dead birds and assesses dropping color, which can predict disease two to three days earlier. Chicken Boy is part of a broader trend in precision poultry farming (PPF), which emphasizes monitoring environmental conditions to optimize production and welfare. Environmental factors such as temperature, air velocity, ventilation rate, litter quality, humidity, and gas concentrations (e.g., ammonia and carbon dioxide) are crucial for maintaining healthy conditions (Dallimore, 2017). High levels of these gases can lead to decreased production and health issues in birds (David et al., 2015).In addition to Chicken Boy, other sensor innovations are being developed. For instance, the ALIS Greenhouse Sensor monitors ammonia, carbon dioxide, and humidity in real time, while the ALIS Ambient Sensor tracks temperature and lighting to optimize yields and welfare. The ALIS Cluster Sensor uses thermal imaging to monitor flock mobility and alert farmers to potential health issues like heat stress, which can trigger disease or infection (Dallimore, 2017).

Chicken Boy and its Functioning

Artificial intelligence and machine learning are increasingly applied in poultry health monitoring. Studies have shown that sound analysis can detect diseases by analysing bird vocalizations. For example, Sadeghi et al. (2015) proposed a method to detect Clostridium perfringens type A infections based on chicken vocalization. Similarly, Banakar et al. (2016) used statistical features from frequency and time-frequency domains to diagnose avian diseases through vocalization.

Wearable sensors for avian influenza virus detection

The threat of influenza virus infection is a major concern for poultry farms, as early detection is crucial to prevent its spread. Conventional diagnostic methods, such as virus culture or PCR, are labor-intensive and slow, requiring samples from symptomatic chickens (Neethirajan et al., 2017). To address these delays, wearable sensors have been developed to detect physiological and movement abnormalities in chickens infected with highly virulent strains like H5N1. Okada et al. (2009) created a lightweight sensor (~5 g) that monitors fever and weakness before death. Later, Okada et al. (2014) demonstrated that accelerometers measuring movement patterns provide faster and more accurate detection compared to body temperature alone. Accelerometers are cost-effective, widely used in smart devices, and suitable for poultry diagnostics. These advancements enable rapid response to infections, reducing transmission within and between farms while minimizing economic losses and zoonotic risks.

  Fig 2: Wearable sensor

Biosensors for avian influenza virus

Biosensors and wearable sensors are emerging as essential tools for detecting poultry diseases like avian influenza, Newcastle disease (ND), and infectious bronchitis. Biosensors utilize bioreceptors, such as proteins or nucleic acids, to recognize pathogens and convert the recognition into electrical or visual signals (Luka et al., 2015). For instance, Chen & Neethirajan (2015) developed a biosensor capable of detecting hemagglutinin (HA) proteins of influenza viruses, including H1 and H5 subtypes, using a two-probe system. Biosensors provide precise diagnoses but require manual sampling and operation, making them suitable for use after clinical symptoms appear. Wearable sensors, on the other hand, offer real-time data transfer and continuous monitoring of physiological symptoms linked to infections, such as fever or movement abnormalities in chickens. However, these sensors lack specificity and may produce false positives due to other factors affecting bird physiology. For ND and infectious bronchitis, Mahdavian et al. (2021) demonstrated that vocalization changes in infected birds could be detected using a wavelet entropy method. This approach successfully distinguished between healthy and infected birds based on vocal pattern variations. The method achieved 87% detection accuracy on the third day and 94% on the fourth day post-infection. Combining wearable sensors for rapid detection with biosensors for accurate diagnosis could enhance poultry disease management. Wearable sensors enable early alerts to physiological changes, while biosensors confirm specific pathogens like avian influenza virus or ND. These technologies reduce economic losses and improve poultry health by enabling timely interventions.

Conclusion

The integration of precision poultry farming (PPF) technologies, such as wearable sensors and biosensors, presents a promising strategy for early and accurate detection of poultry diseases. Wearable sensors facilitate real-time monitoring of physiological and behavioural changes, enabling immediate alerts to potential infections like avian influenza. Simultaneously, biosensors offer precise diagnoses of specific pathogens, ensuring targeted and effective responses. The combined use of these technologies, alongside methods like vocalization analysis for diseases such as Newcastle disease and infectious bronchitis, enhances disease management in poultry farms. This approach not only mitigates economic losses but also promotes animal welfare by enabling timely interventions and reducing the spread of infectious diseases. Ultimately, these advancements contribute to more sustainable and efficient poultry production.