Diabetes Detection Using Artificial Intelligence

A PROJECT REPORT

Submitted by

Shubham Goswami(21BCS11441)

in partial fulfilment for the award of the degree of

BACHELORS IN ENGINEERING
IN
COMPUTER SCIENCE AND ENGINEERING

CHANDIGARH UNIVERSITY
March 2023

1
 TABLE OF CONTENTS

CHAPTER 1. INTRODUCTION
1.1. Identification of Client/ Need/ Relevant Contemporary issue .....................................4
1.2. Identification of Problem .............................................................................................5-6
1.3. Identification of Tasks .................................................................................................6-7
1.4. Timeline .......................................................................................................................8-9

1.5. Organization of the Report ...........................................................................................9-10

2
 List of Figures

Figure 1.1 …………………………………………………………………. 9

3
 ABSTRACT

Artificial intelligence (AI) is a fast-growing field and its applications to

diabetes, a global pandemic, can reform the approach to diagnosis and
management of this chronic condition. Principles of machine learning have been
used to build algorithms to support predictive models for the risk of developing
diabetes or its consequent complications. Digital therapeutics have proven to be
an established intervention for lifestyle therapy in the management of diabetes.
Patients are increasingly being empowered for self-management of diabetes,
and both patients and health care professionals are benefitting from clinical
decision support.
Diabetes is a metabolic disorder comprising of high glucose level in blood over
a prolonged period in the body as it is not capable of using it properly. The
severe complications associated with diabetes include diabetic ketoacidosis,
nonketotic hyperosmolar coma, cardiovascular disease, stroke, chronic renal
failure, retinal damage and foot ulcers. There is a huge increase in the number
of patients with diabetes globally and it is considered a major health problem
worldwide. Early diagnosis of diabetes is helpful for treatment and reduces the
chance of severe complications associated with it. Machine learning algorithms
(such as ANN, SVM, Naive Bayes, PLS-DA and deep learning) and data
mining techniques are used for detecting interesting patterns for diagnosing and
treatment of disease. Current computational methods for diabetes diagnosis
have some limitations and are not tested on different datasets or peoples from
different countries which limits the practical use of prediction methods. This
paper is an effort to summarize the majority of the literature concerned with
machine learning and data mining techniques applied for the prediction of
diabetes and associated challenges. This report would be helpful for better
prediction of disease and improve in understanding the pattern of diabetes.
Consequently, the report would be helpful for treatment and reduce risk of other
complications of diabetes.

4
 CHAPTER 1.
INTRODUCTION
1.1 Identification of client/need/relevant contemporary issue.
Artificial intelligence (AI) is a scientific discipline based on human reasoning
that allows machines to perform functions related to human beings. AI has been
applied to control, monitoring, and classification tasks to make the decision that
achieves better performance, such as spam filtering, image recognition, and
medical diagnosis. The field of medicine has greatly evolved thanks to the
revolution of AI techniques as they help in the prediction of many diseases like
diabetes. AI solves complex problems by looking for signs of symmetry.
Symmetry-adapted machine-learning paradigm is an emerging artificial
intelligence (AI) technology that relies on the extraction and analysis of data to
identify hidden patterns of data.
Artificial intelligence has the potential to improve diabetes detection and
management by analyzing large amounts of patient data to identify patterns and
risk factors for the disease. AI can also assist healthcare professionals in making
more accurate diagnoses and treatment decisions, leading to better patient
outcomes. Additionally, AI can help identify patients who are at risk for
developing diabetes, allowing for earlier interventions and prevention of the
disease.

1.2 Identification of Problem

Diabetes mellitus refers collectively to a group of diseases resulting from
dysfunction of the glucoregulatory system. Hyperglycemia, the hallmark of
diabetes, is the primary consequence of this dysregulation. Chronic
hyperglycemia in diabetes is associated with long-term complications involving
tissue damage and organ failure, which can decrease life expectancy and even
cause death. The International Diabetes Federation estimates that, by 2017,
diabetes affected 425 million people worldwide, of whom, 4 million died in the
same year. These figures are expected to increase dramatically in the coming
decades, placing a rising burden on health care systems.

5
Most diabetes can be categorized into 3 subgroups: type 1 diabetes (T1D), type
2 diabetes (T2D), and gestational diabetes (GDM). Over the long term, T2D
patients become resistant to the normal effects of insulin and gradually lose
their capacity to produce enough of this hormone. A wide range of therapeutic
options are available for patients with T2D. At the early stages of disease, they
commonly receive medications that improve insulin secretion or insulin
absorption, but eventually they must receive external doses of insulin. On the
other hand, T1D patients have severe impairments in insulin production, and
must use external insulin exclusively to manage their blood glucose (BG).
Treatment of T1D requires consistent doses of insulin through multiple daily
injections (MDIs) or continuous subcutaneous insulin infusion (CSII) using a
pump. GDM is treated similarly to T2D, but only occurs during pregnancy due
to the interaction between insulin and hormones released by the placenta.
In each class of diabetes, timely diagnosis, education of patients in self-
management, and continuous medical care are required to prevent acute
complications and minimize the risk of long-term complications (eg,
nephropathy, retinopathy, diabetic foot, cardiovascular disease, or stroke). In
addition to medication, management of diabetes requires adherence to an array
of self-care behaviors that are often very burdensome for patients: carefully
scheduling meals, counting carbohydrates, exercising, monitoring BG levels,
and adjusting endeavors on a daily basis. The effects of nonadherence to
recommended treatment are not immediately evident and long-term
complications may take years to develop. Accordingly, diabetes therapy is
complex, and therapeutic decisions need to take into account diverse medical
factors and lifestyle-related activities that must be optimized to improve diabetic
patients’ quality of life.

1.3 Identification of Tasks

Before being able to confront a problem its existence needs to be identified.
This might seem an obvious statement but, quite often, problems will have an
impact for some time before they are recognised or brought to the attention of
someone who can do anything about them. Making any project requires
understanding of different tasks that can be involved in identifying, building and
testing the solution for the project, and that is what we will be going through in
this section.

6
Step 1: Data collection and pre-processing
 Collect a large dataset of patient information, including demographic
information, medical history, laboratory test results, and other relevant
factors that may be associated with diabetes.
 Pre-process the data by cleaning and normalizing it, handling missing
values, and ensuring that the data is representative and diverse.
Step 2: Feature selection and extraction
 Select the most relevant features from the dataset that may be associated
with diabetes, such as age, BMI, blood glucose levels, HbA1c levels,
family history, and other factors.
 Extract features from the dataset using appropriate techniques, such as
principal component analysis (PCA) or other feature extraction methods.
Step 3: Algorithm selection and training
 Select an appropriate machine learning algorithm for diabetes detection,
such as logistic regression, support vector machines (SVM), decision
trees, or neural networks.
 Train the algorithm on the pre-processed and feature-selected dataset,
using appropriate techniques such as cross-validation and hyperparameter
tuning.
 Evaluate the performance of the algorithm using appropriate metrics,
such as accuracy, precision, recall, F1 score, or area under the receiver
operating characteristic (ROC) curve.
Step 4: Model interpretation and visualization
 Interpret the trained model to understand which features are most
important for diabetes detection, and how the model is making
predictions.
 Visualize the model output, such as the feature importance plot or the
decision boundary plot, to facilitate model interpretation and
understanding.
Step 5: Testing and validation
 Test the trained model on independent datasets to validate its accuracy
and generalizability.

7
 Fine-tune the model, if necessary, based on the results of the testing and
validation.
1.4. Timeline
A project timeline is a visual reference point that allows teams to understand
the flow of steps necessary to complete a project on time. It offers an
overview of all the required tasks, along with their assignments, priority
levels and milestones.
Having one, no matter how simple, allows for optimal time management
and efficiency. Benefits of a Project Timeline are creating a timeline for
projects of any size can provide significant benefits for every member of
your team. Having a plan to refer to whenever things get murky allows
everyone to stay on task and work as efficiently as possible.
So, we have our project timeline as shown in below picture 1.1

Figure 1.1 timeline of Project

8
1.5Organization of the report
The whole report is divided into five major parts, which is known as
Introduction, Literature Review, design flow/process, Results analysis and
Validation and Conclusion. Basically, each of the content of each chapter is:
Chapter 1: Introduction This chapter contains the background of the problem
statement generally and includes the objectives and the scope of the study.
Overall, in this chapter, it summarized the progress of the whole project
describing the how the whole project is been done.

Chapter 2: Literature Review In this chapter, any information which is related

to the project is studied and summarized it, such as reading all the research
papers related to it, and so on. The source of the information can be from
journals, books, internet, articles and etc Based on the information from past
studies and research, it will guide a correct path for the continuous project.

Chapter 3: design flow/process It describes overview of the research methods

involves data collection and pre-processing, feature selection and extraction,
algorithm selection and training, model interpretation and visualization, testing
and validation, and integration and deployment.

Chapter 4: Results analysis and validation It states all the results such as tables,
figures and graphs when research carrying out. All the important results will be
presented in a comprehensive way. It also analyses the results with experimental
study. The explanation of results will let readers more understand on this
project. Validation of Result It used the result in Chapter 3 to validate the
precision and accuracy of data in achieving optimum value.

Chapter 5: Conclusion It summarizes the main findings and how the scope is
covered fully and brief recommendation for further study. Hence, alternative
ways or suggestions can be recommended for improving the project in
further studies.