You are on page 1of 16

Medical-based Decision Support System for

Diabetes and Reverse Diabetes Using ML Algorithms


Introduction:

• Diabetes is a chronic metabolic disorder characterized by elevated blood glucose


levels due to insufficient insulin production by the body.

• It is a significant global health concern with increasing prevalence worldwide.

• Diabetes can be caused by family history, genetics, unhealthy eating, lack of


exercise, and being overweight.

• It can be divided into two categories, type 1 diabetes and type 2 diabetes.

• Many complications occur if diabetes remains untreated and unidentified.

• Early detection and prediction of diabetes play a crucial role in managing the
disease and preventing complications.
Existing System:

• The current system is a manual process where the concerned doctor analyzes the patient’s
reports manually and it requires more time for diabetes identification.
• Home glucose monitoring is also done by the patients using the glucose monitoring device to
detect diabetes.
• Most of the research studies done before mainly focused on one or two particular complex
techniques to test the data, while comprehensive research on many common techniques is
missing.
• Many research works are there on the topic of diabetes disease prediction using machine
learning algorithms, but few papers just presented the idea of diabetes prediction, and no
implementation was done.
• Few papers have done the implementation using Python, R language, Weka tool, and other data
science tools, all these are just machine learning models built using static datasets but no real-
time implementations were done and also there is no reverse diabetes prediction.
Problem Statement:

 It is very important to identify the diabetes disease at early stages to


avoid the disease’s serious complications such as cardiac arrest,
kidney failure, etc...

 As the current process is manual and as there is no system to


predict reverse diabetes patients, we require an efficient system that
can automatically predict diabetes and reverse diabetes using
machine learning algorithms.
Objectives

• To build an automated medical system that will automatically predicts diabetes


disease using efficient supervised learning techniques.

• To develop a real time application with a machine learning model to predict reverse
diabetes patients using machine learning algorithms.

• To develop GUI-based software for fast decision making and providing proper
treatments for patients to cure the disease at the earliest. We apply 2 different
algorithms for prediction and results compared to identify the better algorithm.
Proposed System:
• The proposed system is an automation for diabetes and reverse diabetes identification using the
old diabetes patients’ data.
• System meant for hospitals where doctors and patients can access the application using browsers.
• The system uses many medical-based factors such as Pregnancy, Glucose, Blood Pressure, Skin
Thickness, Insulin, BMI, Blood Glucose Level, Glycated Hemoglobin (HbA1c), Diabetes Pedigree
Function, and Age.
• System uses supervised learning algorithms for diabetes and reverse diabetes prediction.
Algorithms such as naive Bayes, KNN algorithm, decision tree, SVM, etc...
• The system uses 2 different algorithms for prediction and results compared to identify the better
algorithm.
• System developed using Microsoft technologies such as Visual Studio, SQL Server, HTML, CSS,
JS, and jQuery.
Methodology:
Parameters Used:

• Glucose
• Blood Pressure
• Hypertension
• Skin Thickness
• Insulin
• Body Mass Index (BMI)
• Blood Glucose Level
• Glycated Hemoglobin (HbA1c)
• Diabetes Pedigree Function
• Age.
System Requirements:

Hardware Requirements:
Software Requirements:
• RAM: 4 GB +
 IDE: Visual Studio
• Hard Disk: 500 GB +
 Data Base: SQL Server • Processor: Quad Core onwards
 Programming Language: C# • Speed: 2 G HZ

 Front End Technologies: HTML,


CSS, JS and jQuery.

 OS – Windows 10 or 11
Modules of the project:

1. Staff Creation Module (Admin)

2. Patient Registration Module (Receptionist)

3. Parameters Module (Receptionist)

4. Dataset Module (Receptionist)

5. Input Module - New Patient (Physician)

6. Prediction Module - Diabetes and Reverse Diabetes (Physician)

7. Treatment Module (Physician, Patients)

8. Account Module (Admin, Receptionist, Physician, Patient)


System Design:

Architecture Design
THANK YOU

You might also like