Archivos Latinoamericanos de Producción Animal. 2025 (OctubreDiciembre). 33 (4)
Noninvasive monitoring in guinea pig production to assist in
establishing their health status: a literature review
Recibido: 20250523. Revisado: 20251002. Aceptado 20251116
1Corresponding author: johnbarco@gmail.com
2 Universidad CESMAG, Facultad de Ingeniería, Pasto, Colombia
199
John BarcoJiménez1,2
Abstract. Guinea pigs have significant economic and cultural importance in Latin America. However, their
breeding is affected by various diseases that reduce productivity and animal welfare. Early detection of these
diseases is crucial, but traditional methods such as palpation and visual observation can be imprecise and
subjective. Additionally, guinea pigs are very sensitive to human contact, and improper handling can negatively
affect their health. In this context, the question arises: Is it possible to develop a noninvasive monitoring system
that combines temperature and weight measurements in guinea pigs to determine their health status more
accurately than traditional methods? To answer this question, a comprehensive review of the existing literature is
proposed. The central focus of this review is to identify and analyze the most relevant noninvasive technologies for
measuring temperature and weight in guinea pigs. Given the scarcity of studies specifically conducted in this
species, the review also examines noninvasive technologies applied to other variables (such as activity, behaviour,
and environmental conditions) and in other livestock species, discussing their potential adaptation to guinea pig
production systems. Furthermore, methodologies for combining these measurements to assess health status, as well
as current trends in artificial intelligence, particularly machine learning, for sensor data analysis are explored. The
review methodology follows a systematic approach, establishing inclusion and exclusion criteria to ensure the
quality and relevance of the information collected. Keywords are defined and relevant research databases are
selected. The article selection process is carried out in three stages: removal of duplicates, exclusion of articles with
restricted access, and filtering according to the research topic. Finally, the quality of the selected articles is evaluated
to deepen the discussion and draw conclusions on the feasibility and potential benefits of implementing non
invasive monitoring systems to improve guinea pig health and welfare in production settings.
Keywords: Cavia porcellus, guinea pig, cuy, health status, animal monitoring.
https://doi.org/10.53588/alpa.330402
Escuela de Ciencias Básicas, Tecnología e Ingeniería,
Universidad Nacional Abierta y a Distancia. Pasto, Colombia
Monitoreo no invasivo en la producción de cuyes para la evaluación
de su estado de salud: una revisión bibliográfica
Resumen. Los cuyes tienen una gran importancia económica y cultural en Latinoamérica. Sin embargo, su cría se ve
afectada por diversas enfermedades que reducen la productividad y el bienestar animal. La detección temprana de estas
enfermedades es crucial, pero los métodos tradicionales, como la palpación y la observación visual, pueden ser imprecisos y
subjetivos. Además, los cuyes son muy sensibles al contacto humano, y un manejo inadecuado puede afectar negativamente
su salud. En este contexto, surge la pregunta: ¿Es posible desarrollar un sistema de monitoreo no invasivo que combine la
medición de la temperatura y el peso en cuyes para determinar su estado de salud con mayor precisión que los métodos
tradicionales? Para responder a esta pregunta, se propone una revisión exhaustiva de la literatura existente. El objetivo
principal de esta revisión es identificar y analizar las tecnologías no invasivas más relevantes para la medición de la
temperatura y el peso en cuyes. Dada la escasez de estudios realizados específicamente en esta especie, la revisión también
examina tecnologías no invasivas aplicadas a otras variables (como la actividad, el comportamiento y las condiciones
ambientales) y en otras especies de ganado, analizando su potencial adaptación a los sistemas de producción de cuyes.
Además, se exploran metodologías para combinar estas mediciones en la evaluación del estado de salud, así como las
tendencias actuales en inteligencia artificial, particularmente en aprendizaje automático, para el análisis de datos de
sensores. La metodología de revisión sigue un enfoque sistemático, estableciendo criterios de inclusión y exclusión para
William Arévalo2
Héctor Mora2Miller Ruales2José Camilo Eraso2
200
Introduction
Guinea pigs have significant economic and cultural
importance in Latin America. However, their breeding
is affected by various diseases that reduce productivity
and animal welfare. Early detection of these diseases is
crucial, but traditional methods such as palpation and
visual observation can be imprecise and subjective
(Tedeschi et al., 2021). Additionally, guinea pigs are
very sensitive to human contact, and improper
handling can negatively affect their health. In intensive
production systems, where large numbers of animals
are concentrated, diseases can spread rapidly, leading
to significant economic losses and affecting animal
welfare (BarcoJiménez, Martínez, y Solarte 2021).
The traditional methods of assessing the health of
guinea pigs, such as physical examination and
observation of behavior, have limitations in terms of
accuracy and objectivity. Palpation, for example,
depends on the experience and sensitivity of the
operator and may not detect subtle changes in body
condition. Additionally, visual observation can be
influenced by subjective factors and may not be
effective in identifying early signs of disease (Zhao et
al. 2024). In this context, the development of non
invasive monitoring systems that provide objective and
quantitative data on the health status of guinea pigs is
of great interest. Such systems would allow for early
detection of diseases, more effective management of
animal health, and improved productivity in guinea
pig production systems. Technological advances in the
field of sensors and data analysis have opened up new
possibilities for noninvasive monitoring of animals.
The use of sensors to measure physiological
parameters such as temperature and weight can
para garantizar la calidad y relevancia de la información recopilada. Se definen palabras clave y se seleccionan bases de
datos de investigación relevantes. El proceso de selección de artículos se lleva a cabo en tres etapas: eliminación de
duplicados, exclusión de artículos con acceso restringido y filtrado según el tema de investigación. Finalmente, se evalúa la
calidad de los artículos seleccionados para profundizar en el análisis y extraer conclusiones sobre la viabilidad y los
beneficios potenciales de implementar sistemas de monitoreo no invasivos para mejorar la salud y el bienestar de las
cobayas en entornos de producción.
Palabras clave: Cavia porcellus, cobaya, cuy, estado de salud, monitoreo animal.
Monitorização não invasiva na produção de cobaias para auxiliar na determinação
do seu estado de saúde: uma revisão da literatura
Resumo. As cobaias têm uma grande importância económica e cultural na América Latina. No entanto, a sua criação é
afetada por diversas doenças que reduzem a produtividade e o bemestar animal. A deteção precoce destas doenças é
crucial, mas os métodos tradicionais, como a palpação e a observação visual, podem ser imprecisos e subjetivos. Além disso,
as cobaias são muito sensíveis ao contacto humano e uma gestão inadequada pode afetar negativamente a sua saúde. Neste
contexto, surge a questão: é possível desenvolver um sistema de monitorização não invasivo que combine medições de
temperatura e peso em cobaias para determinar o seu estado de saúde com maior precisão do que os métodos tradicionais?
Para responder a esta questão, propõese uma revisão abrangente da literatura existente. O foco central desta revisão é
identificar e analisar as tecnologias não invasivas mais relevantes para a medição da temperatura e do peso em cobaias.
Dada a escassez de estudos conduzidos especificamente nesta espécie, a revisão examina também as tecnologias não
invasivas aplicadas a outras variáveis (como a atividade, o comportamento e as condições ambientais) e noutras
espécies de animais de produção, discutindo a sua potencial adaptação aos sistemas de produção de cobaias. Além
disso, são exploradas metodologias para combinar estas medições na avaliação do estado de saúde, bem como as
tendências atuais em inteligência artificial, particularmente a aprendizagem automática, para a análise de dados de
sensores. A metodologia de revisão segue uma abordagem sistemática, estabelecendo critérios de inclusão e exclusão
para garantir a qualidade e a relevância da informação recolhida. As palavraschave são definidas e as bases de dados
de pesquisa relevantes são selecionadas. O processo de seleção dos artigos é realizado em três etapas: remoção de
duplicados, eliminação de artigos com acesso restrito e filtragem de acordo com o tema de investigação. Por fim, a
qualidade dos artigos selecionados é avaliada para aprofundar a discussão e tirar conclusões sobre a viabilidade e os
potenciais benefícios da implementação de sistemas de monitorização não invasivos para melhorar a saúde e o bem
estar dos porquinhosdaíndia em ambientes de produção.
Palavraschave: Cavia porcellus, porquinhodaíndia, cuy, estado de saúde, monitorização animal.
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
201
Noninvasive monitoring in guinea pig production
Theoretical Concepts
The guinea pig (Cavia porcellus) is a rodent
domesticated thousands of years ago in the Andes of
South America. Its domestication originated from wild
guinea pigs such as the Cavia cutleri and the Cavia
tschudii. Archaeological evidence suggests that the
domestication of the guinea pig began 25003600 years
ago, and that its meat was consumed since 250 BC. The
guinea pig breeding was massified in the Peruvian
highlands and then diversified with the migration to
the coast. Nowadays, there is a great demand for its
meat at national and international level (Isabel Ramos
2014), (RSPCA 2015).
Guinea pig breeding
Technical guinea pig breeding aims for efficient meat
production through planned methods and the use of
appropriate technology. It relies on genetically
improved animals and is implemented in valleys near
urban areas to facilitate marketing. Labor seeks
efficiency and animals are grouped by class, sex, and
age. Production records are kept to ensure profitability.
This system differs from empirical breeding in the use
of technology and the pursuit of efficiency. It differs
from commercial breeding in the scale of production
and the investment in infrastructure.
provide valuable information on the health status of
guinea pigs.
Furthermore, the combination of these measurements
with artificial intelligence techniques, such as machine
learning, can allow for the development of predictive
models that identify animals at risk of developing
diseases (Yin et al. 2020). The implementation of non
invasive monitoring systems in guinea pig production
could have a significant impact on the sustainability
and profitability of these systems. Early detection of
diseases would allow for timely intervention, reducing
the need for antibiotic treatments and minimizing
economic losses. Additionally, improved animal welfare
would increase consumer confidence and contribute to a
more sustainable and ethical production system.
The potential benefits of implementing noninvasive
monitoring systems in guinea pig production are
numerous. Such systems can improve animal welfare,
increase productivity, reduce the use of antibiotics, and
contribute to a more sustainable and ethical production system.
The existing literature indicates that several studies
have explored the use of noninvasive monitoring
techniques to assess the health status of guinea pigs.
For instance, research by (BarcoJiménez et al. 2021)
proposed an automated weighing system that
significantly facilitates the handling of guinea pigs,
reducing stress and the risk of injury during weighing.
This innovative system represents a valuable
contribution to improving the welfare and safety of
guinea pigs in production settings. Other studies, such
as the one by (Jebari et al. 2023), have demonstrated the
potential of combining Artificial Intelligence, Internet
of Things, and Edge Computing to create efficient and
intelligent standalone systems for realtime
monitoring, prediction, and advanced automation. This
paper, for example, aimed to monitor and predict the
environmental conditions in poultry barns using an
artificial intelligence algorithm. However, these studies
have generally focused on specific aspects of animal
health or have not been specifically designed for
guinea pigs. The implementation of noninvasive
monitoring systems in guinea pig production could
have a significant impact on the sustainability and
profitability of these systems (Tekın et al. 2021).
Furthermore, the research by (Chung et al. 2020)
suggests a broader industry trend towards intelligent
management and automated health monitoring systems,
indicating the potential for analogous advancements in
the domain of guinea pig production (Liu et al. 2023).
In this context, the question arises: Is it possible to
develop a noninvasive monitoring system that
combines temperature and weight measurements in
guinea pigs to determine their health status more
accurately than traditional methods? To answer this
question, a comprehensive review of the existing
literature is proposed. This review aims to identify and
analyze the most relevant noninvasive technologies for
measuring the temperature and weight of these animals.
Furthermore, methodologies for combining these
measurements to assess the health status of guinea pigs
will be explored. The review will also consider current
trends in artificial intelligence, specifically in machine
learning, to analyze the data collected by sensors. The
review methodology follows a systematic approach,
establishing inclusion and exclusion criteria to ensure
the quality and relevance of the information collected.
Keywords are defined, and relevant research
databases are selected. The article selection process is
carried out in three stages: elimination of duplicates,
elimination of articles with access restrictions, and
filtering according to the research topic. Finally, the
quality of the selected articles is evaluated to deepen
the proposed discussion.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
202
Semicommercial breeding is found between
empirical and technical breeding. It is characterized by
the use of incipient technology and investment in
infrastructure, forage planting, and family labor. It is intended
for selfconsumption and the marketing of surpluses.
Guinea pig variety by origin
2.2.1 Peruvian breed
The Peruvian breed, originating from the northern
highlands of Peru, particularly in Cajamarca, is
distinguished by its short and straight fur, displaying
variable coloring such as white, black, brown, or
spotted patterns. This robust breed exhibits
adaptability to diverse climates and altitudes. With a
mediumsized build, approximately 1 kg in weight, it
demonstrates commendable prolificacy, yielding 34
offspring per birth. Primarily cultivated for local and
family consumption, the Peruvian breed stands as a
versatile and resilient choice for breeders in the region
(Cesar Guerra León 2009).
2.2.2 Andean breed
Developed at the National Institute of Agricultural
Research (INIA) in Junín, the Andean breed is
distinguished by its short, straight fur, mirroring the
color palette of the Peruvian breed. This precocious
breed exhibits rapid growth and boasts excellent meat
production qualities. Surpassing the Peruvian breed in
size, it reaches weights of up to 1.5 kg, coupled with
high prolificacy, yielding 45 offspring per birth. Its
primary orientation is towards commercial meat
production, positioning the Andean breed as a
strategic choice for those involved in the meat industry
(Hever Castro 2002).
2.2.3 Inti breed
Established at the National University of San Marcos
(UNMSM), the Inti breed is defined by its short,
straight fur, exhibiting a diverse range of colors. This
mediumsized breed, weighing between 1 and 1.2 kg,
demonstrates commendable prolificacy, yielding 34
offspring per birth. Renowned for its high meat content
and superior quality, the Inti breed serves a dual
purpose, catering to both family consumption and
commercial production. Its versatility and desirable
attributes position the Inti breed as a valuable choice
for those seeking a balanced and productive cavy
variety (Isabel Ramos 2014).
2.2.4 INCA line
Engineered by INIA Cajamarca, the INCA line stands
as a synthetic creation resulting from the strategic
crossbreeding of the Peruvian, Andean, and Inti
breeds. This distinctive line showcases short, straight
fur in a variety of colors. Merging the desirable traits
inherited from its parent breeds—namely, robustness,
precocity, size, and prolificacy—the INCA line is
purposefully tailored for largescale commercial meat
production. Its carefully selected genetic characteristics
position it as an efficient and effective choice for those
involved in the expansive realm of meat production
(Isabel Ramos 2014).
Most common diseases in guinea pig breeding
According to (Cesar Guerra León 2009; Hever Castro
2002; Isabel Ramos 2014), in guinea pig breeding
various diseases can impact the health of these animals,
categorized across different systems:
Respiratory Diseases: Pneumonia, the most preva
lent ailment, is caused by bacteria, viruses, or fungi,
leading to symptoms such as nasal discharge,
breathing difficulties, appetite loss, and weight
reduction. Pasteurellosis, primarily affecting the lungs
and caused by Pasteurella multocida, manifests as
fever, coughing, difficult breathing, and nasal
discharge. Rhinotracheitis, resulting from viruses,
induces symptoms like sneezing, coughing,
conjunctivitis, and nasal discharge.
Digestive Diseases: Coccidiosis, stemming from
protozoa (Eimeria), affects the digestive system,
causing diarrhea, loss of appetite, weight loss,
dehydration, and anemia. Salmonellosis, caused by
Salmonella spp., manifests as diarrhea, fever, loss of
appetite, and weight loss. Enterotoxemia, associated
with Clostridium perfringens, leads to bloody
diarrhea, dehydration, and fatality.
Parasitic Diseases: Scabies, caused by mites, results
in itching, skin irritation, and hair loss. External
parasites like fleas and lice contribute to anemia,
weight loss, and decreased meat production.
Ringworm, a fungal infection, produces skin lesions
with associated hair loss.
Viral Diseases: Aujeszky's disease induces tremors,
paralysis, and death; it is zoonotic, posing a risk to
human transmission. Guinea pig pox, characterized by
skin pustules, is highly contagious.
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
203
Methodology
Planification
Guinea pigs (Cavia porcellus) are animals of significant
economic and cultural importance in many regions of
Latin America. Their breeding is adversely affected by
various diseases that negatively impact productivity
and animal welfare. Early detection of these diseases is
crucial for proper management, although traditional
methods such as palpation and visual observation can
be imprecise and subjective. Additionally, guinea pigs
are highly sensitive to human contact. Repeated
interaction without necessary care (by nonexpert
personnel) can lead to a decline in their health or well
being. For instance, processes such as cage changing,
weighing, or counting can cause injuries if the animals
are improperly handled during transport, lifting, or
weighing.
In this context, the research question arises: Is it
possible to develop a noninvasive monitoring system
that combines temperature and weight measurements
in guinea pigs to determine their health status with
greater accuracy than traditional methods?
To address this research question, a comprehensive
review of the available literature will be conducted.
The primary objective of this review is to identify and
analyze the most relevant noninvasive technologies
for the precise measurement of temperature and
weight in these animals. Additionally, various existing
methodologies for combining temperature and weight
measurements will be explored to obtain an assessment
of the animal's health status. The review will also
encompass current trends in artificial intelligence,
specifically in the field of machine learning, for
analyzing the data collected by sensors.
The review methodology is described in Figure 1
and is based on a systematic approach. In the planning
stage, the review protocol is defined based on the
primary research question to guide the search and
selection of relevant literature. Additionally, inclusion
and exclusion criteria for the studies are established
(see Table 1) to ensure the quality and relevance of the
information collected.
The process of searching for scientific articles begins
with the definition of keywords, which are selected
based on the research question. Subsequently, similar
terms are sought to broaden the search and construct a
search string that balances precision and breadth.
Finally, relevant research databases for the research
topic are selected, where the search for articles that
meet the criteria is conducted.
Once the database of articles has been compiled, the
selection proceeds through a threestage process. First,
duplicates are eliminated. Second, articles with access
restrictions are removed. Finally, exhaustive filtering is
Other Diseases: Malnutrition, often caused by a
deficient diet in vitamins, minerals, or proteins,
presents symptoms like weight loss, dull fur, and
weakness. Dental problems, resulting from excessive
tooth growth, can lead to eating difficulties and mouth
injuries. Dystocia denotes difficulty during childbirth.
These comprehensive categories highlight the range of
health challenges in guinea pig breeding.
As the reviewed information indicates, seemingly
unremarkable symptoms such as fever and weight loss
may mask various underlying health concerns in
guinea pigs. These can range from respiratory
infections to infestations with parasites or viral
diseases. Therefore, continuous observation and
monitoring of these early warning signs are crucial for
prompt diagnosis and intervention, potentially
safeguarding the guinea pig's wellbeing.
Table 1. Inclusion/Exclusion criteria. Source: Created by the
authors.
Figure 1. Diagram of the review methodology. Source:
Created by the authors.
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
204
conducted according to the research topic, using
predefined criteria to ensure the quality and relevance
of the selected articles.
The final stage of the process involves evaluating the
quality of the selected articles. This evaluation is
conducted from the perspective of specific research
questions. During this process, relevant data and
information are extracted to deepen the discussion
proposed by the review.
Search process
Definition of keywords and similar terms based on
the research question. Subsequently, relevant research
databases for the research topic are selected.
Accordingly, the following search equation is
constructed:
(“guinea pigs” OR livestock OR cattle OR rodents
OR “cavia porcellus”) AND (temperature OR weight)
AND (“monitoring device” OR IOT OR internetof
things) AND “health status”
Finally, based on the topics, the following databases
are selected: Scopus, Web of Science (WOS), Google Scholar,
ScienceDirect, IEEE, and PubMed, as shown in Table 3.
Selection (duplicate elimination, exclusion, quality
assessment, research questions, and evaluation)
The selection of articles according to the search string
was performed based on the title, keywords, and
abstract. Additionally, the inclusion/exclusion criteria
from Table 1 were applied, considering only journal
articles within the period 2019 – 2024, in English or
Spanish, within the fields of Veterinary Science,
Engineering, Computer Science, and Artificial Intelligence.
The selection of documents was carried out in three
specific stages: i) searching using the equation, ii)
filtering articles according to inclusion/exclusion
criteria, iii) eliminating duplicates based on the digital
object identifier (DOI), and iv) removing non
downloadable articles. The search resulted in the
following numbers of articles from the various
databases, as shown in Table 4.
Quality assessment and synthesis
To evaluate the quality of the articles, the review
methodologies proposed by (ReveloSánchez,
CollazosOrdóñez, y JiménezToledo 2018) and
(Kitchenham et al. 2009) are used as references. Based
on these, the following evaluation criteria are
established:
1. Relevance of the article to the topics of interest.
2. Rigor in data analysis.
3. Presentation of simulation or implementation results.
Each selected article is reviewed in relation to each of
these criteria. A score from 0 to 2 is assigned to each
criterion, where 0 indicates that it is not met, 1
indicates it is partially met, and 2 indicates it is
satisfactorily met. Only articles that obtain a total score
of 4 or higher are selected. Following the quality
assessment, a final group of 49 articles is obtained.
To perform consistent data extraction focused on the
topic of the article, the following four research
questions have been formulated based on the primary
research question presented in Section 3.1:
Table 2. Keywords and similar terms. Source: Created by
the authors.
Table 3. Databases consulted. Source: Created by the
authors.
Table 4. Number of selected articles. Source: Created by the
authors.
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
205
QA1: What are the latest trends in noninvasive
monitoring systems applied to guinea pig or livestock
production to improve management practices and
preserve animal health, and which technology or
technological innovation has been most applied in this area?
QA2: What are the fundamental characteristics that a
monitoring system in guinea pig or livestock
production must possess for the precise measurement
of weight and temperature, in order to improve
management practices and preserve animal health, and
how do these contribute to the improvement of
management practices and preservation of animal health?
QA3: What are the main challenges identified in
relation to noninvasive monitoring systems for
temperature and weight in guinea pig (or livestock)
production?
The final group of selected articles was reviewed from
the perspective of the aforementioned questions,
evaluating the contributions they can make to answer
them. In this way, relevant information was extracted
to construct a discussion around the posed questions,
resulting in a wellfounded synthesis.
The following sections are structured as follows:
Section 4.1 corresponds to QA1, Section 4.2 to QA2,
and Section 4.3 to QA3.
Results y Discussion
Main technologies used in noninvasive monitoring
4.1.1. Wearable sensors
Wearable collars represent a pivotal advancement in
animal monitoring technology, serving as versatile
tools for tracking health indicators and behavioral
patterns in various species (Chung et al. 2020; Zhang et
al. 2020). Specifically designed for animals, these collars
integrate a multitude of functionalities, including GPS
capabilities for precise location tracking (Okokpujie et
al. 2023). For instance, in the context of livestock
management, GPSenabled collares de seguimiento play a
crucial role in optimizing grazing practices and ensuring
the welfare of herds. By harnessing technologies such as
RFID and GPS, these collars empower farmers to
streamline their operations, enhancing efficiency and
mitigating losses (Peng et al. 2022).
GPS monitoring, a cornerstone feature of these
wearable devices, revolutionizes the way animals are
monitored and managed. Through the utilization of
GPS technology, these devices furnish realtime data
on the movement patterns of animals, furnishing
invaluable insights into their grazing behaviors and
territorial tendencies. The integration of GPS trackers
into livestock management practices facilitates the
implementation of precision farming methodologies,
enabling farmers to make informed decisions
regarding herd management and resource allocation.
Additionally, complementary technologies such as
satellite imagery and drones further augment the
capabilities of GPSbased monitoring systems,
enriching the depth and scope of animal surveillance
efforts (Achour et al. 2022; Okokpujie et al. 2023).
Furthermore, RFID tags constitute another
indispensable component of wearable collars,
facilitating seamless identification and tracking of
individual animals. These radio frequency
identification tags serve as the linchpin of automated
feeding and health monitoring systems, enabling
farmers to efficiently manage large herds while
ensuring optimal care and nutrition for each animal. By
leveraging the combined power of GPS tracking and
RFID technology, wearable collars emerge as
indispensable tools in modern livestock management
practices, empowering farmers to enhance
productivity, optimize resource utilization, and
safeguard animal welfare (Peng et al. 2022).
Wearable sensors, including portable sensors and
threedimensional accelerometers, play a pivotal role in
recognizing animal behaviors and activity patterns.
These sensors enable realtime monitoring, facilitating
early disease detection and comprehensive movement
tracking of animals. By harnessing the data provided
by these sensors, farmers can make informed decisions
regarding feed replenishment and enhance operational
planning in animal feed manufacturing (Chung et al.
2020; Zhang et al. 2020).
In dairy cattle management, health monitoring
systems are increasingly employing implantable
biosensors and portable wireless transmitters to
measure subdermal temperature and other
physiological parameters. Additionally, inertial motion
capture systems capture movement trajectories to
facilitate robotic motor skill training. Biometric sensors
further expand monitoring capabilities by tracking
physiological parameters such as heart rate,
temperature, and stress levels, interfacing with the
Internet of Things (IoT) for continuous health
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
206
surveillance (Dutta et al. 2022; Jebari et al. 2023;
Morchid et al. 2024; Saravanan y Saraniya 2018).
The integration of health monitoring sensors, inclu
ding heart rate monitors and blood glucose sensors,
enables early detection of health issues, disease
prevention, and treatment monitoring in livestock. The
adoption of sensor technologies in animal monitoring
signifies a paradigm shift towards precision livestock
farming and sustainable agriculture. Through the
utilization of wearable sensors, remote monitoring
systems, and advanced technologies like ZigBee for
environmental monitoring, farmers can optimize
production outcomes, enhance animal welfare, and
ensure overall herd health (Alonso et al. 2020).
Moreover, the application of Bluetooth Low Energy
(BLE) facilitates the collection of locomotor and
physiological activity signals in circadian rhythm
studies, while LoRa technology enables realtime
monitoring of livestock health, movements, and
environmental conditions. Additionally, lowcost
sensors deployed in silos infer feed consumption and
estimate anticipated animal growth, further enhancing
operational efficiency and resource management in
animal husbandry practices. The ongoing research and
innovation in sensor technologies promise continual
advancements in animal monitoring practices,
delivering tangible benefits to farmers and the
agricultural sector as a whole (Alonso et al. 2020; Hua
Luo Shuting Cheng y Wang 2021).
4.1.2. Mobile applications
Mobile applications have become integral tools in
modern agriculture, particularly in the realm of animal
monitoring. These applications leverage the power of
smartphones to provide farmers with convenient
access to realtime data on their livestock, enabling
them to make informed decisions and optimize their
management practices. One key application of
cellphone technology in animal monitoring is health
tracking. By utilizing sensors and data analytics, these
applications can monitor vital health metrics of
livestock, such as heart rate, temperature, and activity
levels, allowing farmers to proactively address any
health issues that may arise (Okokpujie et al. 2023).
Another important aspect of mobile applications in
animal monitoring is behavior tracking. These
applications are equipped with sensors that can
monitor various behavior patterns of animals,
including feeding habits, movement, and social
interactions. By analyzing this data, farmers can gain
valuable insights into the wellbeing and behavior of
their livestock, enabling them to detect any deviations
from normal patterns and take appropriate action [1].
Furthermore, mobile applications would play a
crucial role in disease detection and management in
livestock. These applications could utilize advanced
algorithms and machine learning techniques to analyze
health data and identify potential signs of diseases
early on. By providing alerts and notifications for
disease outbreaks, farmers can implement timely
interventions to prevent the spread of illnesses and
safeguard the health of their animals (Tekın et al. 2021).
Additionally, cellphone applications offer remote
surveillance capabilities, allowing farmers to monitor
their livestock and premises from anywhere at any
time. Through live streaming features, farmers can
keep an eye on their animals, ensuring their safety and
wellbeing even when they are not physically present
on the farm. This remote monitoring capability
enhances farm security and enables quick responses to
any emergencies or unusual events (Gao et al. 2023).
4.1.3. Video and image processsing
Video analysis and image processing technologies
are revolutionizing animal management, enabling real
time monitoring and control with a high level of
precision. Computer vision and object detection
techniques, such as YOLO models, allow for the
individual identification of cows and sheep in images.
Technologies like SheepInst, based on Mask RCNN
and ConvNeXtE, segment sheep instances in images
for more accurate analysis. Cameras and video analysis
systems enable the assessment of individual animal
behavior, anomaly detection, and evaluation of their
health status (Li et al. 2021; Zhao et al. 2023).
Convolutional Neural Networks (CNNs) and object
detection algorithms are utilized to identify patterns in
behavior and detect signs of disease or stress [2].
Computer vision techniques, RGBD sensors, and
convolutional neural networks are employed for
applications such as weight estimation, animal
identification, and realtime tracking. Machine learning
algorithms are also used to classify realtime cow
behavior patterns from accelerometer and location data
(Cotticelli et al. 2023).
Underwater cameras and ultrasound detectors play
a crucial role in locating fish schools and facilitating
precise feeding in aquaculture. These technologies
enhance the efficiency and effectiveness of aquaculture
operations by providing insights into fish behavior and
optimizing feeding practices. Additionally, they
contribute to the overall sustainability of aquaculture
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
207
y minimizing feed waste and environmental impact.
Through the integration of underwater cameras and
ultrasound detectors, aquaculturists can monitor fish
populations, assess their health, and ensure optimal
growth and development. Moreover, these
technologies aid in the early detection of potential
issues, allowing for timely intervention and mitigation
of risks. Overall, underwater cameras and ultrasound
detectors represent invaluable tools for enhancing
productivity and sustainability in aquaculture
operations (Gong et al. 2023).
Further advancements in animal monitoring include
the utilization of drones equipped with thermal cameras
to detect heat stress in livestock. This technology enables
farmers to identify potential health issues early on,
allowing for prompt intervention and prevention of
heatrelated illnesses. Additionally, satellite data is
employed to assess pasture quality and predict forage
availability. By leveraging satellite imagery, farmers can
optimize grazing management practices, ensuring
sufficient food supply for their animals while preserving
pasture health. These advanced monitoring techniques
contribute to the overall welfare and productivity of
livestock, enhancing sustainable agricultural practices
(Xiong et al. 2019).
4.1.4. Artificial intelligence, big data and machine
learning
Artificial intelligence (AI), big data, and machine
learning (ML): technologies are increasingly deployed
across various sectors, including the meat industry,
aquaculture, and agriculture, to enhance animal
management, health, welfare, and productivity.
Big data analytics in the meat industry: Big data
analytics play a crucial role in the meat industry,
facilitating the creation of significant data resources for
predictive modeling and rigorous validation testing. By
harnessing large datasets, predictive models can be
developed to improve meat quality, ensuring consumer
satisfaction and operational efficiency (Gao et al. 2023).
Deep learning applications: Deep learning, a subset
of machine learning, is extensively applied across
various domains, including automatic species
identification in fish and the enhancement of intelligent
aquaculture practices. Through the utilization of deep
learning algorithms, automated species recognition
systems can accurately identify fish species, thereby
optimizing aquaculture operations and promoting
sustainable fish farming practices. Furthermore, deep
learning algorithms are proficient in analyzing images
of poultry to identify visual cues associated with
different diseases. By training models on extensive
datasets comprising annotated images, these algorithms
can effectively discern specific symptoms or
abnormalities indicative of potential diseases. Moreover,
deep learning models possess the capability to process
vast amounts of data, encompassing sensor data from
Internet of Things (IoT) devices, to detect patterns or
anomalies signaling the onset of a disease. Through
continuous monitoring and data analysis, these models
furnish early warnings to farmers, facilitating prompt
intervention (Echegaray et al. 2022; Gong et al. 2023).
Deep learning models can be trained to analyze non
invasive data sources, such as vocalization traits, dermal
temperature variations, and fecal characteristics, without
inducing stress in poultry. This noninvasive approach
allows for uninterrupted monitoring of poultry health
without disrupting their natural behavior patterns.
Additionally, deep learning technology enables realtime
monitoring of poultry health parameters, including
temperature variations, vocalization patterns, and
behavioral changes. By continuously analyzing these
parameters, the system can identify deviations from
normal patterns and promptly alert farmers to potential
health issues (Jebari et al. 2023).
Deep learning algorithms exhibit exceptional
performance in classification tasks, enabling them to
categorize data into distinct classes based on learned
features. In the realm of poultry health, these
algorithms excel in classifying health indicators or
symptoms into categories such as healthy, diseased, or
at risk, thereby facilitating early detection and
intervention measures.
Advanced machine learning techniques: Advanced
machine learning techniques are employed to predict
livestock body weight, quantify growth rates, and enhance
the accuracy of broiler weight prediction. By leveraging
machine learning models, farmers can optimize feeding
regimes, monitor animal growth trajectories, and
maximize production efficiency (Tekın et al. 2021).
Image processing and anomaly detection:
Algorithms for object detection, optical flow, and
image processing are utilized to identify anomalous
behaviors and health conditions in animals. By
analyzing video feeds and sensor data, these
algorithms can detect deviations from normal behavior
patterns, enabling timely intervention and disease
prevention (Yang y Qiu 2024).
AI for health management: Artificial intelligence,
particularly deep learning algorithms, is leveraged to
analyze vast amounts of data, predict potential health
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
208
issues, and improve animal health management
practices. Through AIpowered systems, farmers can
proactively address health concerns, minimize disease
outbreaks, and optimize veterinary care protocols
(Alonso et al. 2020; Jebari et al. 2023; Li et al. 2021).
Machine learning for stress and productivity
prediction: Machine learning algorithms are utilized to
predict stress levels, illness occurrences, and
productivity levels in livestock. By analyzing
physiological data and environmental factors, these
algorithms provide valuable insights into animal well
being, enabling farmers to implement targeted
interventions and improve overall productivity
(Hansen et al. 2022).
Casebased reasoning and aiassisted support
systems: Casebased reasoning mechanisms are
employed to identify anomalies in animal behavior and
vital signs. Additionally, AIassisted support systems
are developed to optimize farm procedures and
enhance animal welfare. By integrating AI technologies
into farm management practices, farmers can make
datadriven decisions, streamline operations, and
ensure the wellbeing of their animals (Hua Luo
Shuting Cheng y Wang 2021; Jebari et al. 2023).
IOT and Cloud Computing
IoT: IoT sensors are increasingly deployed to moni
tor both physiological and environmental parameters
crucial for animal wellbeing, including body
temperature, physical gestures, and relative humidity.
These sensors leverage technologies such as the
Internet of Things to track the health and location of
livestock, employing wireless devices equipped with
sensors for realtime tracking and detection of changes
in animal behavior during grazing activities. Moreover,
IoT sensors gather data from micro and
macroenvironments, as well as individual information
from each animal, facilitating comprehensive
monitoring and analysis (Morchid et al. 2024).
Cloud computing: Cloudbased applications serve
as pivotal platforms for gathering and analyzing data
from various sensors, offering valuable insights into
animal health and productivity. Smart farming
platforms, exemplified by systems like ThingSpeak,
enable farmers to aggregate, visualize, and analyze live
data from sensors, empowering them to make
informed decisions in real time. Furthermore, the
deployment of edge devices for processing data closer
to the source enables realtime analytics for animal
monitoring. Artificial intelligence (AI) algorithms
deployed at the edge play a crucial role in detecting
irregular patterns in animal behavior and identifying
health conditions promptly. Additionally, edgebased
decision support systems facilitate datadriven decision
making in livestock management, enhancing operational
efficiency and animal welfare (Dineva y Atanasova 2023;
MateoFornés et al. 2021; Saravanan y Saraniya 2018).
Integration of iot sensors and cloud computing::
The integration of IoT sensors and cloud computing
enables seamless communication among devices and
components within the farm, monitoring
environmental variables such as temperature,
humidity, and air quality. This integration fosters real
time connections between microservices and web
applications through a clouddeployed microservices
architecture, thereby enhancing system responsiveness
and enabling timely interventions. By leveraging IoT
sensors and cloud computing technologies, farmers can
gain comprehensive insights into animal health and
behavior, fostering improved management practices
and enhancing overall productivity in livestock
operations (Saravanan y Saraniya 2018).
Emerging Technologies
Augmented Reality (AR): Augmented Reality (AR)
is under investigation for remote postmortem
veterinary inspection in pigs, facilitating bidirectional
communication between veterinarians and technicians
at a distance. Additionally, AR holds promise for
enhancing training and education in animal health and
welfare (Cotticelli et al. 2023; Peng et al. 2022).
Robotics: Robotics is increasingly employed in
mechanized agricultural operations, leading to reduced
reliance on manual labor and enhanced efficiency in
tasks such as feeding, milking, and monitoring. Robotic
systems contribute to improved animal welfare and
productivity by automating routine tasks and
providing realtime data on animal behavior and
health (Liu et al. 2023).
Edge computing: Edge Computing is utilized to
process data at the location where it is generated,
enabling rapid response to events in real time. By
deploying computing resources closer to the data
source, Edge Computing reduces latency, enhances
data security, and improves overall system
performance in animal monitoring applications
(Alonso et al. 2020).
Smart animal feeding systems: Smart feeding
systems are developed to ensure adequate nutrition
and efficient development of animals. Innovations like
eSynch combine electronically controlled hormone
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
209
delivery devices with sensing capabilities for remote
programming and monitoring of hormone delivery,
thereby optimizing animal growth and reproduction
processes (Ren et al. 2023).
Blockchain: Blockchain technologies offer
unprecedented opportunities for enhancing
traceability, security, and consumer trust in the dairy
industry. Through immutable and transparent
transaction records, Blockchain ensures the traceability
of dairy products, certifying their origin and
production processes. Additionally, Blockchain
enhances data security and integrity, mitigating the risks
of fraud and ensuring the safety and quality of dairy
products. By enabling consumers to track the journey of
dairy products through the value chain, Blockchain
fosters trust and confidence in the authenticity and
quality of dairy produce (Zheng et al. 2021).
Table 5. Synthesis of technologies applied to guinea pigs production. Source: Created by the authors.
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
210
Table 5. Synthesis of technologies applied to guinea pigs production. Source: Created by the authors. (Continuation)
To assess the real applicability of these technologies
to guinea pigs (See Table 5), it is essential to consider
their specific physiological, behavioral, and productive
traits. Guinea pigs are small, highly social, and
sensitive to thermal and environmental stress, with
rapid growth, short production cycles, and carcass
quality strongly linked to feed intake and housing
conditions. In addition, they are typically raised in
large population groups, which makes it more effective
to deploy technologies designed to manage and
monitor groups at scale, such as video capture
combined with AI and machine learning for automatic
detection of collective behavior patterns and
anomalies. Therefore, AI, big data, IoT, and emerging
technologies should be oriented toward noninvasive
monitoring of body weight, growth curves, feed and
water consumption, locomotion patterns,
vocalizations, grouping behavior, and microclimatic
conditions in cages or pens. By training machine
learning models on these guineapigsspecific
variables, it becomes possible to detect early signs of
stress or disease, optimize feeding strategies, and
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
211
improve productive performance, ensuring that the digital
tools described in this section are realistically adapted to
the biology and management practices of this species.
Essential characteristics of a monitoring system for
livestock production
An effective monitoring system in guinea pig or
livestock production must possess several key
characteristics to ensure precise weight and temperature
measurement. These attributes are crucial for optimizing
management practices and safeguarding animal health.
The system's most fundamental requirement is
accuracy and reliability in its measurements. It must
provide consistent and precise readings of both weight
and temperature to enable informed decisionmaking.
This precision is paramount for effective management.
Realtime monitoring capabilities are also essential,
allowing for immediate detection of anomalies and
swift intervention to prevent potential health crises
(Okokpujie et al. 2023). Automated data collection and
analysis further streamline operations, reducing the
labor intensity of manual monitoring and minimizing
the risk of human error.
Beyond measurement, the system's design must be
noninvasive, minimizing stress and discomfort for the
animals to promote their welfare and ensure data
integrity (Tekın et al. 2021). Durability and robustness
are necessary for the system to withstand the rigorous
livestock environment, ensuring longterm functionality
and minimizing maintenance. Wireless connectivity
facilitates remote monitoring and data access, enhancing
flexibility and enabling timely responses to emerging
issues. A userfriendly interface is critical for ease of
operation and data interpretation, empowering farmers
and veterinarians to make informed decisions effectively
(Okokpujie et al. 2023). Lastly, integration with existing
farm management systems optimizes data flow and
overall operational efficiency, while robust data security
measures protect sensitive information and maintain the
monitoring process's integrity.
Impact on management practices and animal health
These collective characteristics significantly contri
bute to improved management practices and the
preservation of animal health.
By providing precise, realtime data, the system
enables datadriven decisionmaking, leading to
optimized feeding strategies and early disease
detection. Continuous monitoring of individual animal
weights, a key health and performance indicator,
becomes far more efficient and accurate than
traditional, laborintensive method. Automated weight
monitoring, in particular, reduces labor costs, enhances
data accuracy, and allows for realtime tracking of
growth rates and feed conversion ratios. The
integration of sensor technologies and artificial
intelligence (AI) provides invaluable insights into
animal health, productivity, and overall wellbeing
(Neethirajan 2023).
Precision livestock farming leverages advanced
software and systems to integrate diverse data,
improving health, welfare, and production while
minimizing environmental impact through tele
surveillance and other sophisticated monitoring
mechanisms. By establishing normal parameter ranges
for farm infrastructure and animal behavior, these
systems can detect deviations signaling potential
health issues, supporting reliable information for AI's
routine learning processes (Tekın et al. 2021).
Sensor technologies, including wearable devices and
environmentbased sensors, are crucial for collecting real
time animal data (Neethirajan 2023). Wearable sensors
track physiological parameters and behavioral traits,
while environmental sensors like video cameras and
thermal imaging monitor surrounding conditions. This
continuous health surveillance is vital, especially with
growing global concerns about zoonotic disease
transmission and increasing meat demand. AI, combined
with sensors, offers innovative solutions for challenges
like identifying and managing "shy feeders"—animals
that consume less food in group settings and are prone to
health and productivity issues (Neethirajan 2023).
Automated systems that monitor animal behavior
have become increasingly prevalent, using image
analysis to estimate weight and body condition,
monitor feed and water intake, assess gait, and detect
estrus. Smart animal feeding systems can measure
individual feed consumption, ensuring each animal
receives the appropriate ration for optimal weight
gain. AI can identify shy feeder cows and adjust
feeding strategies by analyzing behavioral patterns via
video and motion sensors. Monitoring water volume,
drinking frequency, and total water intake also helps
detect heat stress and reduce morbidity rates.
Modern approaches employ electronic wearables
like smart ear tags, neck collars, and leg collars to
collect data on temperature, activity, rumination, and
ingestion patterns. These devices are often combined
with robotic milkers, automatic feeders, and inline milk
sensors to enhance data collection and analysis. The
data can predict feed intake, weight gain, and overall
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
212
animal health, enabling early disease detection and
promoting optimal management (Dos Reis et al. 2021).
Realtime monitoring of animal behavior and biological
changes allows for early problem diagnosis and
immediate assessment of housing practices (Cotticelli et
al. 2023). This ability to continuously assess and
promptly respond to animal needs enhances well
being, increases productivity, and promotes
sustainable livestock farming (Dos Reis et al. 2021;
Neethirajan 2023).
A LoRa sensor network can provide a costeffective
solution for extensive pasture monitoring by tracking
livestock location and activity. While challenges exist
with wearable sensors, including the need for long
range networking, reliable power supplies, and low
costs, opensource technologies and software can
reduce implementation complexity. LoRa technology
enables remote grazing systems and animal health
monitoring when integrated with GPS (Pagano et al.
2023), and its adaptability extends to both indoor and
outdoor environments for monitoring barn parameters.
Realtime communication is vital for maximizing the
benefits of livestock monitoring, with WiFi or
Bluetooth often used in confinement operations
requiring highdensity, lowdistance data
transmissions (Dos Reis et al. 2021).
Continuous livestock monitoring and tracking
through LoRaWAN networks enable realtime data
collection, processing, and management, enhancing
operational efficiency and animal welfare. The
combination of inmemory data capture and mobile
connectivity has proven valuable, reducing animal
stress and improving data quality. Lowcost CPUs
compatible with Arduino, motion sensors, GPS
receivers, and LoRa radios can be used to modify CPU
sampling frequency via opensource Arduino IDE
software (Okokpujie et al. 2023). Employing a mobile
gateway alongside static gateways can optimize data
extraction rates and energy consumption, particularly
in extensive livestock areas (Pagano et al. 2023). As
technology advances, the integration of these systems will
continue to drive innovation in livestock farming, leading
to more efficient, sustainable, and ethical practices.
Challenges in implementing noninvasive monitoring
systems
The development and deployment of noninvasive
monitoring systems for temperature and weight in
livestock and guinea pig production face several
significant hurdles that can impact their effectiveness
and broader adoption.
Data management, security, and ethical considerations
One of the most pressing challenges involves data
management, encompassing security, privacy, and the
sheer volume of data generated. Sensor technologies
and AI collect vast amounts of sensitive information,
necessitating sophisticated data analytics and storage
capabilities for proper management and interpretation.
Ensuring robust data security and privacy is
paramount to protect this sensitive information.
Beyond the technical aspects, ethical concerns are
crucial. There's a valid argument that increased
reliance on technology can lead to the objectification of
animals, reducing them to mere data points. It's vital to
address perceptions of the systems being intrusive or
causing stress to the animals (Neethirajan, 2023), and to
ensure strict adherence to animal welfare standards.
Furthermore, there's a risk of overreliance on AI
without proper understanding, which could lead to
erroneous decisions if staff aren't adequately trained.
Integration, adoption barriers, and regulatory compliance
Interoperability issues present a significant hurdle
when integrating diverse sensor technologies and AI
applications. Achieving seamless operation requires
careful planning and the establishment of standardized
protocols for data collection, storage, and analysis. On
the human side, resistance from farmers or workers
can hinder adoption, stemming from concerns about
cost, complexity, or potential job displacement.
Addressing this requires adequate training for staff on
how to install and maintain these technologies,
interpret the data, and act upon insights provided by
AI. Since noninvasive monitoring systems are still in
early stages, many businesses are hesitant to make
significant investments. Overcoming this may require
incentives through government or industry programs
that encourage businesses to implement these
technologies (Gong et al., 2023). Finally, the success of
these technologies is intertwined with regulatory
compliance, demanding strict adherence to safety,
environmental, and animal welfare regulations.
Strategic placement of sensors in hightraffic areas like
watering and supplementary feeding points can also
significantly improve data consistency and reliability
(Tedeschi et al., 2021).
The adoption of these technologies faces significant
economic, social, and practical barriers for smallscale
guinea pig producers, who represent the majority of
the sector in Latin America. Many systems (wearable
sensors, robotics, cloud platforms, blockchain,
advanced AI solutions) were originally designed for
largescale livestock operations and require substantial
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
213
animal health, enabling early disease detection and
promoting optimal management (Dos Reis et al. 2021).
Realtime monitoring of animal behavior and biological
changes allows for early problem diagnosis and
immediate assessment of housing practices (Cotticelli et
al. 2023). This ability to continuously assess and
promptly respond to animal needs enhances well
being, increases productivity, and promotes
sustainable livestock farming (Dos Reis et al. 2021;
Neethirajan 2023).
A LoRa sensor network can provide a costeffective
solution for extensive pasture monitoring by tracking
livestock location and activity. While challenges exist
with wearable sensors, including the need for long
range networking, reliable power supplies, and low
costs, opensource technologies and software can
reduce implementation complexity. LoRa technology
enables remote grazing systems and animal health
monitoring when integrated with GPS (Pagano et al.
2023), and its adaptability extends to both indoor and
outdoor environments for monitoring barn parameters.
Realtime communication is vital for maximizing the
benefits of livestock monitoring, with WiFi or
Bluetooth often used in confinement operations
requiring highdensity, lowdistance data
transmissions (Dos Reis et al. 2021).
Continuous livestock monitoring and tracking
through LoRaWAN networks enable realtime data
collection, processing, and management, enhancing
operational efficiency and animal welfare. The
combination of inmemory data capture and mobile
connectivity has proven valuable, reducing animal
stress and improving data quality. Lowcost CPUs
compatible with Arduino, motion sensors, GPS
receivers, and LoRa radios can be used to modify CPU
sampling frequency via opensource Arduino IDE
software (Okokpujie et al. 2023). Employing a mobile
gateway alongside static gateways can optimize data
extraction rates and energy consumption, particularly
in extensive livestock areas (Pagano et al. 2023). As
technology advances, the integration of these systems will
continue to drive innovation in livestock farming, leading
to more efficient, sustainable, and ethical practices.
Challenges in implementing noninvasive monitoring
systems
The development and deployment of noninvasive
monitoring systems for temperature and weight in
livestock and guinea pig production face several
significant hurdles that can impact their effectiveness
and broader adoption.
Data management, security, and ethical considerations
One of the most pressing challenges involves data
management, encompassing security, privacy, and the
sheer volume of data generated. Sensor technologies
and AI collect vast amounts of sensitive information,
necessitating sophisticated data analytics and storage
capabilities for proper management and interpretation.
Ensuring robust data security and privacy is
paramount to protect this sensitive information.
Beyond the technical aspects, ethical concerns are
crucial. There's a valid argument that increased
reliance on technology can lead to the objectification of
animals, reducing them to mere data points. It's vital to
address perceptions of the systems being intrusive or
causing stress to the animals (Neethirajan, 2023), and to
ensure strict adherence to animal welfare standards.
Furthermore, there's a risk of overreliance on AI
without proper understanding, which could lead to
erroneous decisions if staff aren't adequately trained.
Integration, adoption barriers, and regulatory compliance
Interoperability issues present a significant hurdle
when integrating diverse sensor technologies and AI
applications. Achieving seamless operation requires
careful planning and the establishment of standardized
protocols for data collection, storage, and analysis. On
the human side, resistance from farmers or workers
can hinder adoption, stemming from concerns about
cost, complexity, or potential job displacement.
Addressing this requires adequate training for staff on
how to install and maintain these technologies,
interpret the data, and act upon insights provided by
AI. Since noninvasive monitoring systems are still in
early stages, many businesses are hesitant to make
significant investments. Overcoming this may require
incentives through government or industry programs
that encourage businesses to implement these
technologies (Gong et al., 2023). Finally, the success of
these technologies is intertwined with regulatory
compliance, demanding strict adherence to safety,
environmental, and animal welfare regulations.
Strategic placement of sensors in hightraffic areas like
watering and supplementary feeding points can also
significantly improve data consistency and reliability
(Tedeschi et al., 2021).
The adoption of these technologies faces significant
economic, social, and practical barriers for smallscale
guinea pig producers, who represent the majority of
the sector in Latin America. Many systems (wearable
sensors, robotics, cloud platforms, blockchain,
advanced AI solutions) were originally designed for
largescale livestock operations and require substantial
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
Conflicts of Interest: The authors have no conflicts of interest to declare.
Author Contributions: John BarcoJiménez conceived the study, performed the analyses and wrote the
manuscript. William Arévalo contributed to the research idea and study design. Héctor Mora, Miller Ruales and
José Camilo Eraso critically reviewed the manuscript and contributed to the interpretation of results..
Funding: This work was supported by Universidad CESMAG through its Vicerrectoría de Investigaciones.
Edited by: Victor VelezMarroquin
214
upfront investment, subscription fees, and ongoing
technical support that are difficult to afford in small
family farms with low profit margins, high market
uncertainty, and frequent informality. Limited internet
connectivity in rural areas, dependence on stable
electricity, and scarce local technical services for
installation, calibration, and maintenance further
increase the perceived risk of investing in “fragile”
technologies that may fail and be difficult to repair.
From a social and practical perspective, adoption is
also constrained by digital literacy, daily workload,
and the need for solutions that provide clear,
immediate benefits. Smallholders tend to prioritize
lowcost, robust tools that can be easily integrated into
existing routines over complex systems focused on
individualanimal monitoring. This reinforces the
relevance of grouporiented technologies (e.g. video
and AI for pens or rooms, environmental sensors at
room level, simple mobile apps for aggregated records)
rather than expensive individuallevel devices. It also
highlights the importance of technical assistance,
training, and cooperative schemes (producer
associations, cooperatives, or community projects) to
share costs, reduce perceived complexity, and foster
genuine technological uptake in guinea pig production
systems over the medium term.
Conclusions
The adoption of noninvasive monitoring systems in
guinea pig production represents a promising step
forward in animal health management. Integrating
sensorbased measurements of temperature and weight
with artificial intelligence and IoT technologies can
support earlier detection of health problems, more precise
decisionmaking, and improvements in welfare and
productivity. The literature reviewed shows that
automated weighing systems, environmental monitoring,
and AIbased analytics—although mostly developed for
other species—offer a valuable framework that can be
adapted to the specific physiological and management
characteristics of guinea pigs.
For practical implementation in guinea pig
production, the most immediate opportunities lie in
grouplevel monitoring rather than individual, device
based solutions. Lowcost environmental sensors
(temperature, humidity, CO2) combined with simple
automated weighing platforms at pen or cage level,
and, where feasible, videobased monitoring with basic
computer vision, can provide actionable information
while keeping costs manageable for smallholders. It is
advisable to prioritize incremental adoption through
pilot projects in cooperatives or producer associations,
accompanied by training in data interpretation and
basic maintenance, to reduce perceived technological
complexity and support longterm uptake.
Future research should focus on generating species
specific evidence for guinea pigs. Priority lines include:
(i) defining threshold values and dynamic patterns for
temperature and weight associated with early disease
or welfare impairment; (ii) developing and validating
machine learning models trained on data from guinea
pig production systems, especially under Latin
American conditions; (iii) conducting cost–benefit
analyses of different sensor and AI configurations for
small and mediumscale producers; and (iv) evaluating
the comparative performance of noninvasive
monitoring versus traditional methods in terms of
diagnostic accuracy, animal stress, and labor
requirements. Addressing these gaps will be essential
to move from conceptual potential to robust, field
tested solutions that are economically viable and
socially acceptable for the guinea pig sector.
Acknowledgments
We gratefully acknowledge the invaluable support
provided by Universidad Nacional Abierta y a
Distancia UNAD, specifically its Escuela de Ciencias
Básicas, Tecnología e Ingeniería in CCAV Pasto and the
Vicerrectoría de Investigación y Emprendimiento
(VIEM). Our gratitude also extends to Universidad
CESMAG and its Vicerrectoría de Investigaciones for
their contributions to this work.
BarcoJiménez et al.
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
215
Literature Cited
Achour, B., Belkadi, M., Aoudjit, R., Laghrouche, M.,
Lalam, M., & Daoui, M. (2022). Classification of dairy
cows’ behavior by energyefficient sensor. Journal of
Reliable Intelligent Environments, 8(2), 165–182.
https://doi.org/10.1007/s40860021001443
Gao, R., Liu, Q., Li, Q., Ji, J., Bai, Q., Zhao, K., & Yang,
L. (2023). Multitarget rumination behavior analysis
method of cows based on target detection and optical
flow algorithm. Sustainability (Switzerland), 15(18).
https://doi.org/10.3390/su151814015
Alonso, R. S., SittónCandanedo, I., García, Ó., Prieto,
J., & RodríguezGonzález, S. (2020). An intelligent
edgeIoT platform for monitoring livestock and crops
in a dairy farming scenario. Ad Hoc Networks, 98,
102047. https://doi.org/10.1016/j.adhoc.2019.102047
BarcoJiménez, J., Martínez, M., & Solarte, A. L. (2021).
Sistema de pesaje automatizado que facilita el
manejo de cuyes (Cavia porcellus). Archivos de
Zootecnia, 70(269), 112–116. https://doi.org/
10.21071/az.v70i269.5425
Castro, H. (2002). Sistemas de crianza de cuyes a nivel
familiarcomercial en el sector rural. Benson
Agriculture and Food Institute.
Chung, H., Li, J., Kim, Y., Van Os, J. M. C., Brounts, S.
H., & Choi, C. Y. (2020). Using implantable biosensors
and wearable scanners to monitor dairy cattle’s core
body temperature in real time. Computers and
Electronics in Agriculture, 174, 105453. https://
doi.org/10.1016/j.compag.2020.105453
Cotticelli, A., Verde, M. T., Liccardo, A., De Alteriis, G.,
Lamonaca, F., Matera, R., Neglia, G., Peric, T., Prandi,
A., & Bonavolontà, F. (2023). On the use of 3D camera
to accurately measure volume and weight of dairy
cow feed. Acta IMEKO, 12(4), 1–6. https://doi.org/
10.21014/actaimeko.v12i4.1633
Dineva, K., & Atanasova, T. (2023). Health status
classification for cows using machine learning and
data management on AWS Cloud. Animals, 13(20),
3254. https://doi.org/10.3390/ani13203254
Dos Reis, B. R., Easton, Z., White, R. R., & Fuka, D.
(2021). A LoRa sensor network for monitoring
pastured livestock location and activity.
Translational Animal Science, 5(2), 1–9. https://
doi.org/10.1093/tas/txab010
Dutta, D., Natta, D., Mandal, S., & Ghosh, N. (2022).
MOOnitor: An IoTbased multisensory intelligent
device for cattle activity monitoring. Sensors and
Actuators A: Physical, 333, 113271. https://doi.org/
10.1016/j.sna.2021.113271
Echegaray, N., Hassoun, A., Jagtap, S., TettehCaesar,
M., Kumar, M., Tomasevic, I., Goksen, G., &
Lorenzo, J. M. (2022). Meat 4.0: Principles and
applications of Industry 4.0 technologies in the
meat industry. Applied Sciences, 12(14), 6986.
https://doi.org/10.3390/app12146986
Gong, B., Dai, K, Shao, J., Jing, L., & Chen, Y. (2023).
FishTViT: A novel fish species classification method
in multiwater areas based on transfer learning and
vision transformer. Heliyon, 9(6), e16761. https://
doi.org/10.1016/j.heliyon.2023.e16761
Guerra León, C. (2009). Manual técnico de crianza de
cuyes. CEDEPAS Norte.
Hansen, M. F., Oparaeke, A., Gallagher, R., Karimi, A.,
Tariq, F., & Smith, M. L. (2022). Towards machine
vision for insect welfare monitoring and behavioural
insights. Frontiers in Veterinary Science, 9, 1–7.
https://doi.org/10.3389/fvets.2022.835529
Luo, H., Cheng, S., Jiang, Z., Hou, W., & Wang, Z.
(2021). Economical Bluetooth Low Energybased
telemetry system with combined data processing
method for longterm laboratory animal monitoring
for biological rhythm research. Chronobiology
International, 38(3), 451–465. https://doi.org/
10.1080/07420528.2020.1868489
Ramos, I. (2014). Crianza, producción y
comercialización de cuyes (1st ed.). MACRO.
Jebari, H., Hayani Mechkouri, M., Rekiek, S., &
Reklaoui, K. (2023). PoultryEdgeAIIoT system for
realtime monitoring and predicting by using
artificial intelligence. International Journal of
Interactive Mobile Technologies, 17(12), 149–170.
https://doi.org/10.3991/ijim.v17i12.38095
Kitchenham, B., Brereton, O. P., Budgen, D., Turner,
M., Bailey, J., & Linkman, S. (2009). Systematic
literature reviews in software engineering: A
systematic literature review. Information and
Software Technology, 51(1), 7–15. https://doi.org/
10.1016/j.infsof.2008.09.009
Li, S., Fu, L., Sun, Y., Mu, Y., Chen, L., Li, J., & Gong,
H. (2021). Individual dairy cow identification based
on lightweight convolutional neural network.
PLOS ONE, 16(11), 1–14. https://doi.org/10.1371/
journal.pone.0260510
Liu, X., Wang, Z., Li, J., Cangelosi, A., & Yang, C.
(2023). Demonstration learning and generalization
of robotic motor skills based on wearable motion
tracking sensors. IEEE Transactions on
Instrumentation and Measurement, 72, 1–15.
https://doi.org/10.1109/TIM.2023.3288240
MateoFornés, J., PagèsBernaus, A., PlàAragonés, L.
M., CastellsGasia, J. P., & BabotGaspa, D. (2021).
An Internet of Things platform based on
Noninvasive monitoring in guinea pig production
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
216
Tekın, K., YurdakökDıkmen, B., Kanca, H., & Guatteo,
R. (2021). Precision livestock farming technologies:
Novel direction of information flow. Ankara
Üniversitesi Veteriner Fakültesi Dergisi, 68(2), 193–212.
https://doi.org/10.33988/auvfd.837485
Morchid, A., El Alami, R., Raezah, A. A., & Sabbar, Y.
(2024). Applications of Internet of Things (IoT) and
sensors technology to increase food security and
agricultural sustainability: Benefits and challenges.
Ain Shams Engineering Journal, 15(3), 102509.
https://doi.org/10.1016/j.asej.2023.102509
ISSNL 10221301. Archivos Latinoamericanos de Producción Animal. 2025. 33 (4): 199  216
microservices and cloud paradigms for livestock.
Sensors, 21(17), 1–22. https://doi.org/10.3390/
s21175949
Neethirajan, S. (2023). Artificial intelligence and sensor
technologies in dairy livestock export: Charting a
digital transformation. Sensors, 23(16). https://
doi.org/10.3390/s23167045
Okokpujie, K., Okokpujie, I. P., Ogundipe, A. T.,
Anike, C. D., Asaboro, O. B., & Vincent, A. A.
(2023). Development of a sustainable Internet of
Thingsbased system for monitoring cattle health
and location with web and mobile application
feedback. Mathematical Modelling of Engineering
Problems, 10(3), 740–748. https://doi.org/10.18280/
mmep.100302
Pagano, A., Croce, D., Tinnirello, I., & Vitale, G. (2023).
A survey on LoRa for smart agriculture: Current
trends and future perspectives. IEEE Internet of
Things Journal, 10(4), 3664–3679. https://doi.org/
10.1109/JIOT.2022.3230505
Peng, Y., Zeng, Z., Lv, E., He, X., Zeng, B., Wu, F., Guo,
J., & Li, Z. (2022). A realtime automated system for
monitoring individual feed intake and body weight
of grouphoused young chickens. Applied Sciences,
12(23). https://doi.org/10.3390/app122312339
Ren, Y., Duhatschek, D., Bartolomeu, C. C., Erickson,
D., & Giordano, J. O. (2023). An automated system
for cattle reproductive management under the IoT
framework. Part I: The eSynch system and cow
responses. Frontiers in Animal Science, 4, 1–11.
https://doi.org/10.3389/fanim.2023.1093851
ReveloSánchez, O., CollazosOrdóñez, C. A., &
JiménezToledo, J. A. (2018). El trabajo colaborativo
como estrategia didáctica para la enseñanza/
aprendizaje de la programación: Una revisión
sistemática de literatura. TecnoLógicas, 21(41), 115–
134. https://doi.org/10.22430/22565337.731
RSPCA. (2015). Care for your guinea pigs. Harper
Collins.
Saravanan, K., & Saraniya, S. (2018). Cloud IoTbased
novel livestock monitoring and identification
system using UID. Sensor Review, 38(1), 21–33.
https://doi.org/10.1108/SR0820170152
Xiong, X., Lu, M., Yang, W., Duan, G., Yuan, Q., Shen,
M., Norton, T., & Berckmans, D. (2019). An
automatic head surface temperature extraction
individual broiler. Sensors, 19(5286).
Yang, H., & Qiu, S. (2024). A novel dynamic contextual
feature fusion model for small object detection in
satellite remotesensing images. Information, 15(4),
230. https://doi.org/10.3390/info15040230
Yin, X., Wu, D., Shang, Y., Jiang, B., & Song, H. (2020).
feature fusion model for small object detection in
satellite remotesensing images. Information, 15(4),
230. https://doi.org/10.3390/info15040230
Zhang, M., Feng, H., Luo, H., Li, Z., & Zhang, X.
(2020). Comfort and health evaluation of live
mutton sheep during transportation based on
wearable multisensor system. Computers and
Electronics in Agriculture, 176, 105632. https://
doi.org/10.1016/j.compag.2020.105632
Zhao, H., Mao, R., Li, M., Li, B., & Wang, M. (2023).
SheepInst: A highperformance instance
segmentation of sheep images based on deep
learning. Animals, 13(8), 1–19. https://doi.org/
10.3390/ani13081338
Zhao, X., Tanaka, R., Mandour, A. S., Shimada, K., &
Hamabe, L. (2025). Remote vital sensing in clinical
veterinary medicine: A comprehensive review of
recent advances, accomplishments, challenges, and
future perspectives. Animals, 15(7), 1033. https://
doi.org/10.3390/ani15071033
Zheng, H., Zhang, T., Fang, C., Zeng, J., & Yang, X.
(2021). Design and implementation of poultry
farming information management system based on
cloud database. Animals, 11(3), 1–15. https://
doi.org/10.3390/ani11030900
BarcoJiménez et al.