Final Year Project CHP 1& 2 CHENAI MAKOKO.docx

LAYER BIRD VACCINATION MONITORING &
DISEASE DETECTION SYSTEM
Project Proposal
Submitted by
CHENAI MAKOKO
N0199598
Proposed Supervisor
Ms. C. Chivasa
A PROPOSAL SUBMITTED TO
The Department of Informatics at the National University of
Science and Technology in partial fulfillment of the requirements
for the degree of
BACHELOR OF SCIENCE (HONS) INFORMATICS
Faculty of Applied Sciences
[14 November 2021]

1
CHAPTER 1 - Introduction
1.1 Introduction
Layer poultry farming means raising egg-laying poultry birds for commercial
egg production. Layer chickens are a particular species of hens, which need to
be raised from one day old. They start laying eggs commercially from 18-19
weeks of age. They remain to lay eggs continuously until they are 72-78 weeks.
They can produce about one kg of eggs by consuming about 2.25 kg of food
during their egg-laying period. Several highly egg-productive layer breeds are
available worldwide (https://www.facebook.com/growelagrovet, 2018).
Throughout the bird’s lifetime, it should reach all the target weights set. This
will ensure a healthy, well-developed, long-enduring bird. To produce quality
output farmers should follow vaccination and treatments religiously. Layers are
affected by different diseases which may be caused by viruses, bacteria, and
fungi. To avoid losses farmers should know when a bird needs treatment and
how? Should they fail, they can contact a vet.
Farmers need a dependable assistant to conduct their day-to-day work. Small-
scale farmers might not have enough labor at their disposal and are at a
disadvantage when it comes to record-keeping, monitoring, and keeping track
of vaccinations, treatments, and overall bird health. Many of them miss crucial
steps in the process because of work overload. Most available applications cater
to large commercial organizations with automated systems including poultry
machinery that feeds certain data into the system.
Layer birds’ management usually refers to the husbandry practices or
production techniques that help maximize production efficiency. Farmers can
have an application at hand that allows them to track progress and get warnings
or recommendations in this case disease prediction or detection, treatment, and
vaccination reminders. Sound management practices are very essential to
optimize production, and this includes detecting any diseases before things get
out of hand. Developing a System that is exclusively for the small-scale layer

2
poultry farmer helps in not only record-keeping but also has a decision-making
tool or recommender. This software application covers certain aspects of Layer
Farms. A person with basic knowledge can easily use this software.
1.2 Background
Over the years, farming of layer birds has been on the rise in Zimbabwe. This
has especially proven the need to keep track of farming records and automate
systems so that decisions are made easily with a glance at the dashboard. The
need also dictates that recommendations be made to the farmer from the data
they provide. This study will focus on these aspects and how best they help the
layer farmer effectively.
Layer farmers face many challenges, devastating diseases being the major ones.
Poultry is highly infected with coccidiosis, salmonella, and Newcastle diseases
amongst others, which cause a high mortality rate of livestock. Some of the
disorders, for instance, Salmonella are zoonotic, meaning they can even spread
to humans hence affecting the health of the community. In addition to diseases,
farmers lack access to reliable sources of information on poultry due to several
extension officers, distant locations for consultations,s and a lack of awareness
and recommended animal husbandry practices (Lwoga et al., 2010). They
instead rely on word of mouth from friends and their ways and tradition.
Farmers need solutions to the problems to increase productivity; the use of
vaccines is the common countermeasure at hand though does not apply to all
diseases. The recent use of deep learning in disease detection motivates the need
to contribute to robust diagnostics (Albarqouniet al., 2016)
1.3 Problem Description
Coccidiosis, Salmonella, and Newcastle are some of the common poultry
diseases that curtail poultry production if they are not detected early. In most
third-world countries, these diseases are not detected early due to limited access
to agricultural support services by poultry farmers (Machuve et al., 2022).
A system is needed to run a small-scale layer farm adequately with the rising
demand for poultry. To avoid losses, farmers can always prevent the causes i.e.,
diseases. Small-scale farmers may not always have access to a vet at all times.

3
Improved management and disease control can have a substantial impact on
household economies. Reduced losses will ensure that more birds could be
successfully reared and, assuming the extra birds can be properly fed, this will
allow more eggs to be collected and consumed or sold as a regular source of
income. Treatment of the sicknesses after late identification results in a high
mortality rate for the birds (Wong et al., 2017). Therefore, this study aims at
developing a model for the early detection of poultry diseases using deep
learning for early detection of the diseases.
1.4 Aim
To develop a model for early detection of layer bird diseases for layer poultry
farmers.
1.5 Objectives
● To develop a model for early diagnostics of layer bird diseases
● To allow farmers to enter symptoms of layer birds for detecting disease.
● To recommend steps to farmers upon disease identification.
1.6 Project justification
Several cases of layer disease, mainly in Zimbabwe, go undiagnosed due to poor
veterinary support in remote areas. In this context, a centralized system is
needed for eﬀective monitoring and analysis of the layer birds. The farmers are
challenged with inadequate biosecurity measures and limited access to poultry
health services compared to the large-scale commercial poultry system
(Hemalatha et al., 2014). A web-based poultry disease diagnostic system is a
central platform to store the history and predict the possible disease based on
the current symptoms experienced by a bird to ensure a faster and more accurate
diagnosis. Early disease prediction can help the users determine the severity of
the disease and take quick action(Liakos et al. 2018). Improvement in layer
farming practices shall mitigate the effects caused by the diseases, as various
studies indicate that the most efficient way to manage poultry diseases is via
early detection and treatment (Yazdanbakhsh et al., 2017). The target group of

4
this study is small-scale layer farmers who use deep litter or semi-commercial
production systems (Wong et al., 2017). Only a small proportion of the diseases
that affect layers can be controlled with vaccination and studies show that in
very few countries in Africa, 28% use models to solve different problems
(Brooks-Pollock et al., 2015). Therefore, there is a need for tools with efficient
and effective methods for diagnostics of diseases that will lead to better yield
and an increase in production.
1.7 Project Scope
The study will be useful to farmers, as the outcome is a tool for the early
detection of diseases affecting layers, contributing to robust diagnostic
measures. This tool will enable farmers to overcome the loss incurred due to
late diagnosis of the disorders affecting layer birds. In the scientific body of
knowledge, the study will contribute to developing a model based on a deep
learning approach that can be deployed on web platforms that will be easily used
by small-scale farmers, extension officers, and other stakeholders.
Rapid detection and diagnostic technologies allow for responses to be made
sooner when the disease is detected, decreasing further bird transmission and
associated costs. Additionally, systems of rapid disease detection produce data
that can be utilized in decision support systems that can predict when and where
the disease is likely to emerge in poultry. (Astill et al., 2018).
In this research, we will be focusing on tracking vaccinations, monitoring
treatments, and data visualization for easy decision-making. We will not be
focusing on other aspects of the whole layer farming process that includes
weighing or focusing on lighting or egg production. In layer farming processing,
several other aspects will not be our focus in this research. The above-mentioned
aspects of the process will be our focus so that we define how captured data will
be used and how the system will use it to the farmer’s advantage. The goals
specific to this project will be defined.

5
1.8 Project Report overview
● Chapter 1 - Introduction
Introduction of the project and its concepts.
● Chapter 2 - Literature Review
Reviews of previous similar work done by other researchers.
● Chapter 3 - Methodology
An explanation of research and software design methodologies that were used
to come up with the solution is explained.
● Chapter 4 - System Analysis and Design
This chapter gives an analysis of the system. “what?”, “how?”, “who, when?”
and “how?” a detailed design of how the system will be achieved is given in the
form of a selected design technique.
● Chapter 5 - Implementation System
converting the design into a system and putting that system through various tests
before deployment.
● Chapter 6 - Conclusion
It concludes the work done by giving an analysis of the results. It also
recommends how the solution can be improved in the future.

6
Chapter 2 - Literature review
2.1 Introduction
This chapter will focus on previous related work on layer disease diagnosis. The
associated challenges are summarized in this review. Intensified systems of
poultry production will require new technologies for the detection and diagnosis
of diseases. This review sets out to summarize them while providing the
advantages and limitations of different types of technologies being researched.
2.2.1 Disease Overview
Differential diagnoses of diseases in layer birds are based on organ systems.
Diseases of the respiratory and gastrointestinal systems are some of the most
important and common diseases seen in layers and may constitute more than
two-thirds of all the diseases one may encounter(extension.msstate.edu, n.d.).
The proper diagnosis of poultry diseases depends on three critical factors:
● Identification of vital organs and body structure.
● Proper examination of birds' fecal matter.
● Knowledge of disease symptoms and lesions.
● A systematic plan for examining the bird’s body.
Examining a sick bird can help farmers get pictures that can be uploaded to the
system to make a diagnosis.
2.2.2
2.3 Existing systems
The overall goal is to describe the related existing systems and to place methods
and contributions to the field in this context. A clear description of previous
work can better describe the current limitations and the need for a new
methodology.

7
2.3.1 Expert System to Diagnose Disease in Poultry Using Certainty Factor
Method.
The methodology used in developing this expert system uses the Certainty
Factor method. With the Certainty Factors method, it can determine the
certainty of the diagnosis of a fact that is true or not. This research can produce
information in the form of symptoms or even diseases arising from livestock, so
that farmers can easily use this information as a reference in making a decision,
of course, the decision in question is the prevention and management of
symptoms or diseases that arise in livestock for efforts are made in the form of
first aid measures. This system can analyze diseases caused by viruses from the
air entered by the user based on the symptoms that exist in poultry. This system
is capable of representing and storing expert knowledge based on value
(Certainty Factor).
The expert system is designed to solve quite complex problems that only experts
can solve. The making of an expert system is not to replace the expert himself
but can be used as a highly experienced assistant. CF logic mode tends to require
more work on the user's part than is required by the normal Binary logic
mode(www.parlog.com, n.d.).
2.3.2 Poultry diseases diagnostics model using deep learning.
This research presented a deep-learning model for detecting poultry diseases. A
deep Convolutional Neural Network (CNN) model was developed to diagnose
healthy and unhealthy poultry fecal images. Training on different deep
convolutional neural network architectures that included a baseline CNN,
VGG16, InceptionV3, MobileNetV2, and Xception.Trained models with farm-
labeled and laboratory-labeled fecal images and later tested them on farm-
labeled fecal images. After comparison with other models, the MobileNetV2
model showed the highest potential for deployment on smartphones, to be used
as a diagnostic tool that farmers can use to distinguish diseased from healthy
poultry based on fecal images(Machuve et al., 2022).

8
Deep learning methods have been demonstrated to automate the disease
diagnostics procedures for both humans and livestock (Quinn et al., 2016;
Zhuang et al., 2018; Okinda et al., 2019; Wang et al., 2019; Yadav and Jadhav,
2019). This method is ideal but was mainly focused on fecal matter. This was
an issue because some poultry diseases present symptoms physically not just
through feces. Farmers need to cover all bases to prevent losses. It is ideal to
use both fecal images and physical images of the bird, even post-more autopsy.
Deep learning works only with large amounts of data. Training it with large
and complex data models can be expensive. It also needs extensive hardware to
do complex mathematical calculations.
2.3.3 An Approach towards IoT-Based Predictive Service for Early
Detection of Diseases in Poultry Chickens.
According to (Ahmed et al., 2021), The demand for real-time adaptive systems
in the poultry industry motivates to propose of a systematic approach to creating
an IoT-based predictive service framework that observes the poultry chickens’
movement data and more accurately predicts the health of the chickens in real-
time. The study implements deep generative models to extrapolate the IoT-
based sensing data of poultry chickens to address the class imbalance problem.
The study predicted the health of poultry chickens by modeling different
machine learning and deep learning classiﬁcation techniques as an IoT service.
It has been observed that the deep learning tabular data classiﬁcation model
TabNet outperforms with the best-classifying accuracy.
2.3.4 Knowledge-Based System for Predominant Chicken Diseases
Diagnosis, Prevention, and Management
This system provides an explanation, prevention, and management
automatically on the bases of the predominant chicken disease after the disease
is diagnosed. This prototype KBS can provide advice for experts to facilitate the
diagnosis, prevention, and management of the predominant chicken disease.
Knowledge is represented in the form of "if-then" rules generated from the
decision tree. However, to make the system applicable in the domain area for
diagnosis, prevention, and management of predominant chicken diseases, some

9
adjustments like incorporating a well-designed user interface and a mechanism
of NLP (Natural Language Processing) facilities are needed.(Girma, Jimma and
Diriba, 2022).This particular system helps the experts in the field to diagnose
diseases. This helps the vet doctors, and not the farmer directly.
2.3.5 Plant diseases and pests detection based on deep learning
Plant diseases and pest detection is a very important research content in the field
of machine vision. It is a technology that uses machine vision equipment to
acquire images to judge whether there are diseases and pests in the collected
plant images. At present, machine vision-based plant diseases and pest detection
equipment has been initially applied in agriculture and has replaced traditional
naked-eye identification to some extent(Liu and Wang, 2021).
For traditional machine vision-based plant diseases and pest detection methods,
conventional image processing algorithms or manual design of features plus
classifiers are often used. This kind of method usually makes use of the different
properties of plant diseases and pests to design the imaging scheme and chooses
an appropriate light source and shooting angle, which is helpful to obtain images
with uniform illumination. Although carefully constructed imaging schemes
can greatly reduce the difficulty of classical algorithm design, but also increase
the application cost. At the same time, in the natural environment, it is often
unrealistic to expect the classical algorithms designed to eliminate the impact
of scene changes on the recognition results. In the real complex natural
environment, plant diseases and pests detection is faced with many challenges,
such as the small differences between the lesion area and the background, low
contrast, large variations in the scale of the lesion area and various types, and a
lot of noise in the lesion image. Also, there are a lot of disturbances when
collecting plant diseases and pests images under natural light conditions. At this
time, the traditional classical methods often appear helpless, and it is difficult
to achieve better detection results.
2.3.6 Developing an Expert System for Plant Disease Diagnosis
In this research, the design and development of an expert system with two
different methods for diagnosing plant diseases were presented. The first one is

10
using the descriptive method (step by step) and the other one uses the Graphical
representation method. A preliminary evaluation of the system showed that the
expert system with the graphical representation is more favorable than the
descriptive one. This is due to the difficulties in describing the symptoms of the
disease. On the other hand, a graphical picture of the symptoms does not require
much description from the user. Expert Systems are considered one of the most
successful methods used to help and support users in making the right decisions
when they lack knowledge in diagnosing plant diseases. Present expert systems
saved a lot of time and effort in identifying plant disease due to the mechanism
used in receiving the data and providing the decisions. It is clear that CLIPS is
very effective in processing and performing such types of activities in an easy
way and in a short time(Abu-Naser, Kashkash, and Fayyad, 2008).
2.4 Evaluation
In this section, we will evaluate the different methods used to diagnose diseases
as mentioned above. This will give a deeper understanding of what needs to be
rectified.
Machine Learning

11
References
● Anon, (2022). 5 Best Poultry Management Apps for Android & iOS |
Free apps for Android and iOS. [online] Available at:
https://freeappsforme.com/poultry-management-apps/#layer-farm-
manager [Accessed 23 Sep. 2022].
● Astill, J., Dara, R.A., Fraser, E.D.G. and Sharif, S. (2018). Detecting and
Predicting Emerging Disease in Poultry with the Implementation of
New Technologies and Big Data: A Focus on Avian Influenza Virus.
Frontiers in Veterinary Science, 5. doi:10.3389/fvets.2018.00263.
● digit (2019). What Is Scrum Methodology? & Scrum Project
Management. [online] Digital. Available at:
https://www.digite.com/agile/scrum-methodology/
● https://www.facebook.com/growelagrovet (2018). Layer Poultry
Farming Guide for Beginners. – Growel Agrovet. [online] Growel
Agrovet. Available at: https://www.growelagrovet.com/layer-poultry-
farming/.
● Janardan (jana@poultry.care) (n.d.). Poultry Farm Management
System. [online] Poultry Care. Available at:
https://www.poultry.care/blog/poultry-farm-management-system
[Accessed 23 Sep. 2022]
● Lutkevich, B. (n.d.). What is the project scope? [online] SearchCIO.
Available at: https://www.techtarget.com/searchcio/definition/project-
scope
● Machuve, D., Nwankwo, E., Mduma, N. and Mbelwa, J. (2022). Poultry
disease diagnostics models using deep learning. Frontiers in Artificial
Intelligence, [online] five, p.733345. doi:10.3389/frai.2022.733345

12
● Rajora, Harish & Punn, Narinder & Sonbhadra, Sanjay & Agarwal,
Sonali. (2021). Web-based disease prediction and recommender system
● SearchCIO. (n.d.). What is a decision support system (DSS)? [online]
Available at: https://www.techtarget.com/searchcio/definition/decision-
support-system
● Trapani, K. (2018). What is Agile/Scrum? [online] cPrime. Available at:
https://www.cprime.com/resources/what-is-agile-what-is-scrum/
● www.fao.org. (n.d.). Chapter 1 - Egg production. [online] Available at:
https://www.fao.org/3/y4628e/y4628e03.htm
● www.measureevaluation.org. (n.d.). Building a Web-Based Decision
Support System — MEASURE Evaluation. [online] Available at:
https://www.measureevaluation.org/resources/publications/wp-18-
216.html [Accessed 23 Sep. 2022]
● extension.msstate.edu. (n.d.). Poultry Disease Diagnosis | Mississippi
State University Extension Service. [online] Available at:
http://extension.msstate.edu/publications/publications/poultry-disease-
diagnosis
● Journals, B. (n.d.). DISEASE DIAGNOSIS SYSTEM.
www.academia.edu. [online] Available at:
https://www.academia.edu/8648954/DISEASE_DIAGNOSIS_SYSTE
M [Accessed 8 Nov. 2022].
● Ahmed, G., Malick, R.A.S., Akhunzada, A., Zahid, S., Sagri, M.R. and
Gani, A. (2021). An Approach towards IoT-Based Predictive Service for
Early Detection of Diseases in Poultry Chickens. Sustainability, 13(23),
p.13396. doi:10.3390/su132313396.
● Girma, D., Jimma, W. and Diriba, C. (2022). Developing a Knowledge-
Based System for Predominant Chicken Diseases Diagnosis,

13
Prevention, and Management. [online] doi:10.21203/rs.3.rs-
1953185/v1.
● Liu, J. and Wang, X. (2021). Plant diseases and pests detection based on
deep learning: a review. Plant Methods, 17(1). doi:10.1186/s13007-021-
00722-9.
● Abu-Naser, S.S., Kashkash, K.A. and Fayyad, M. (2008). Developing
an Expert System for Plant Disease Diagnosis. Journal of Artificial
Intelligence, 1(2), pp.78–85. doi:10.3923/jai.2008.78.85.
● www.parlog.com. (n.d.). Certainty factors. [online] Available at:
http://www.parlog.com/shared/imhelp/hs40.htm [Accessed 10 Nov.
2022].

Final Year Project CHP 1& 2 CHENAI MAKOKO.docx

More Related Content

Similar to Final Year Project CHP 1& 2 CHENAI MAKOKO.docx (20)

Recently uploaded (20)

Final Year Project CHP 1& 2 CHENAI MAKOKO.docx