DIABETES

PREDICTION
using MACHINE
LEARNING
GOWTHAM KUMAR
CONTENTS
• Introduction
• Literature review 1
• Literature review 2
• Literature review 3
• Comparison
• Proposed System
• Methodology
• Dataset
• Conclusion
Introduction
• Diabetes is the most common disease worldwide and
keeps increasing everyday due to changing lifestyles,
unhealthy food habits and over weight problems.

• Identified when blood glucose is higher than normal level

• 2 types of Diabetes: Type 1 and Type 2

• Type 1 Diabetes: Your immune system mistakenly attacks

and destroys the insulin-making cells in your body.

• Type 2 Diabetes: Your body doesn't use insulin properly,

and it often happens slowly.
CAUSES
• Insulin Resistance
• Lifestyle Factors: Poor diet, lack of physical activity,
obesity, and stress
• Age: Risk increases with age, especially after 45.
• Family History: A family history of diabetes can be a
contributing factor.

SYMPTOMS
• Frequent Urination
• Slow Healing
• Excessive Thirst
• Excessive Hunger
• Fatigue
• Blurred Vision
 LITERATURE REVIEW-1
Title of the paper Diabetes Disease Prediction Using Machine Learning Algorithms

About 2020 IEEE EMBS Conference on Biomedical Engineering and Sciences(IECBES)

Area of work Used in medical field

Dataset Pima Indians Diabetes Database from National Institute of Diabetes and Digestive and Kidney Diseases

Algorithm K-Nearest neighbor, Decision tree, Random Forest , Naive Bayes, Support Vector Machine

Dataset Collection--->Data pre processing--->Setting Classification metrics--->Applying Machine

Methodology Learning
(training set ,testing set )--->K-fold cross validation

Result/Accuracy 76% accuracy (kNN), 73% (SVM), 74% Naive Bayes , 72% Decision tree , 70% Random Forest

The early detection of diabetes disease plays a vital role in making decisions on lifestyle changes in
Advantages
high-risk patients and in turn reduces the complications

In future, we will create a diabetes dataset in collaboration with a hosptial or a medical

Future Proposals
institute and will try to achieve better results
 LITERATURE REVIEW-2
Diabetes Disease Prediction Using Machine Learning Algorithms with Feature Selection and
Title of the paper
Dimensionality Reduction

About 2021 7th International Conference on Advanced Computing and Dimensionality Reduction

Area of work Used in medical field

Pima Indians Diabetes Database from National Institute of Diabetes and Digestive and Kidney Diseases
Dataset

Dataset collection->Data pre-processing->Feature Selection->Dimensionality Reduction->Classifier

Methodology
(Support Vector Machine &Random Forest)->Analysis

Algorithm Random Forest , Support Vector Machine

Result/Accuracy 77.3% accuracy (SVM) and 83% accuracy in Random Forest

Understanding of the dataset and pre-processing it to increase efficiency. Performance of feature

Advantages
selection and Dimensionality reduction to increase overall performance
 LITERATURE REVIEW-3
Title of the paper Prediction of Diabetes Using Machine Learning Algorithms in Healthcare

Proceedings of the 24th International Conference on Automation & Computing, Newcastle

About
University,Newcastle upon Tyne, UK ,2018

Area of work Used in medical field

Pima Indians Diabetes Database from National Institute of Diabetes and Digestive and Kidney Diseases
Dataset

Dataset collection--->Data pre-processing--->Feature Selection--->Split Training data(70%)

Methodology
and Testing Data( 30%)--->Machine learning Algorithm--->train model--->Analysis

K-Nearest neighbor, Decision tree, Random Forest , Naive Bayes, Support Vector Machine, Logistic
Algorithm
Regression

Result/Accuracy 77% accuracy in SVM and KNN, 74% in LR and NB , 71% in DT and RF

Our future work will focus in integrating of other methods into the used model for tuning the
Future Proposals
parameters of models for better accuracy
 COMPARISON
Diabetes Disease Prediction Using
Prediction of Diabetes
Prediction of Diabetes Using Machine Machine Learning Algorithms with
PAPER NAME Using Machine Learning
Learning Algorithms in Healthcare Feature Selection and
Algorithms in Healthcare
Dimensionality Reduction

YEAR 2020 2021 2018

Pima Indians Diabetes Database

from National Institute of Diabetes
DATASET Same dataset Same dataset
and Digestive and Kidney Diseases

ALGORITHM KNN Random Forest SVM,KNN

ACCURACY 76% 83% 77%

 PROPOSED SYSTEM
• Utilizing the Pima Indians Diabetes Database from the National
Institute of Diabetes and Digestive and Kidney Diseases for
diabetes prediction.
• Employing the Support Vector Machine (SVM) algorithm for building
the predictive model
• Developing a user-friendly web application for inputting user data
and displaying prediction results.
 METHODOLOGY
DATASET DATA
DATASET
VISUALISATION PREPROCESSING
COLLECTION

SPLIT
DEPLOY THE APPLY MACHINE
TRAIN(80%)
MODEL INTO LEARNING
TEST(20%)
WEB ALGORITHM
DATASET
APPLICATION (SVM)
 DATASET
Pima Indians Diabetes Database from National Institute of
Diabetes and Digestive and Kidney Disease

https://www.kaggle.com/datasets/uciml/pima-indians-diabetes-da
tabase
 ABOUT THE DATSET
• This dataset is originally from the National Institute of Diabetes and
Digestive and Kidney Diseases.

• The objective of the dataset is to diagnostically predict whether or not a

patient has diabetes, based on certain diagnostic measurements included
in the dataset

• All patients here are females at least 21 years old of Pima Indian heritage.

• Total 768 rows indicating 768 patients

• 9 coloumns containing 8 feature lists and 1 class variable

Feature labels
 CLASS LABEL
• Class label is Outcome
• It contains two values 0 or 1
• 0-->The person is not diabetic
• 1-->The person is diabetic
 Conclusion
• The web application provides an accessible and user-friendly platform for
individuals to assess their diabetes risk.

• Achieve high percentage accuracy in prediction of diabetes

• Timely identification of individuals at risk

• Enabling early intervention through lifestyle changes or medical treatments

• Reducing long term healthcare costs

THANK
YOU