Scribd
Upload
14 views19 pages

Heart Disease Prediction System Using Machine Learning

Uploaded by

MD Arsalan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
14 views19 pages

Heart Disease Prediction System Using Machine Learning

Uploaded by

MD Arsalan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 19

HEART DISEASE

PREDICTION SYSTEM USING


MACHINE LEARNING
SUBMITTED BY- UNDER THE GUIDANCE OF-
MD MOHSHIN KHAN & MD ARSALAN MOUMITA CHATERJEE
TABLE OF CONTENTS

 INTRODUCTION
 FEATURE SELECTION & DATASET
 LITERATURE SURVEY
 METHODOLOGY & ALGORITHMS
 EXPERIMENT AND RESULTS
 CONCLUSION
 FUTURE WORK
 REFERENCES
INTRODUCTION

 Heart disease or Cardiovascular disease is one of the biggest reason


for death across the globe.

 It is a cause of 35% of total death worldwide.

 Machine Learning helps in acquiring , classifying and testing of the


data and thus has a great scope in this application

 Processing of the dataset and thus calculating the best accuracy is all
possible due to ML
FEATURES

 Age: Age of the patient


 Sex : Sex of the patient (0-Female 1-Male)
 CP: Chest Pain Type(0-atypical angina,1-typical angina,2-asymptomatic,3-
nonanginal pain)
 TRESTBPS: Resting blood pressure(in mm Hg)
 FBS: Fasting blood sugar>120mg/dl (1=true: 0=false)
 CHOL : Cholestrol level
 RESTECG : Resting electrocardiographic results : assets the heart’s
activity(0=normal, 1=have ST-T , 2=hypertrophy)
 THALACH: Maximum heart rate
 EXANG: Angina caused by exercise is a common complaint of cardiac
patients , particularly when exercising in the cold (0=No, 1=Yes)
 SLOPE: The slope of the peak Exercise ST segment (values 0,1,2)
 CA : Flourosopy coloration of a lot of major vessels(0-3)
 THAL: Number of Defect
 TARGET: Presence or Absence of Heart Disease(0-Absent, 1=Present)
DATASET
LITERATURE SURVEY

AUTHOR TITLE OBJECTIVE


Senthil Kumar Mohan Effective Heart Disease Objective is to find critical
Prediction ML Techniques includes by applying ML
Algorithms , bringing about
improving the exactness in
the prediction.

Sonam Nikhar Prediction of Heart Disease To give a point by point


using ML portrayal of Naïve Bayes and
Decision Tree classifier.
V.V Ramalingam Heart disease prediction The Paper presents a survey
using ML of various models based on
algorithms that found very
popular among the
researchers
LITERATURE SURVEY

AUTHOR TITLE OBJECTIVE

Vijeta Sharma Heart Disease Prediction Find the correlations


using ML Techniques between the different
attributes available in the
dataset with the help of
standard ML Techniques
Dr . M . Kavitha Heart disease prediction In this , The Cleveland Heart
using Hybrid Machine Disease Dataset and
Learning Model Machine Learning
Algorithms are used for
Prediction
 This bar-plot shows that we have
balanced data.
 Number of persons having heart
disease is 526 .
 Number of persons not having
heart disease is 499.
 Percentage of patients having
heart disease is 51.32% and not
having heart disease is 48.68%.
Heart Disease Frequency For Sex

 Red plot indicating having heart


disease.

 Green plot indicating doesn’t any


heart disease.

 From the bar plot the frequency of


female getting heart disease is
higher in this dataset compared to
male.
Heart Disease Frequency For Ages

 This bar-plot shows the


frequency of having heart
disease in different ages.

 Orange plot indicate having


heart disease and Blue plot
indicates not having a heart
disease.

 From the graph we can observe


that ages between 41 to 54 have
a higher frequency of heart
disease.
Maximum Heart Rate VS Age

 It shows the maximum heart rate for


different ages.

 Red dots indicates that a person


having a heart disease.

 Blue dots indicates a person not having


a heart disease.

 From this we can observe that the age


between 40 t0 55 the chances of
having a heart disease is higher.
Heart Disease Frequency to FBS and
Chest Pain
Corelation Matrix

 This correlation matrix shows


the relation between different
attributes.
 Here Chest Pain and Target have
a positive correlation.
 Slope ,Restecg ,Thalach also
have a positive correlation.
 Exercise-induced Angina(exang)
and the target have a negative
correlation.
 Ca , Old Peak also have a
negative correlation with the
target.
ALGORITHMS USED

ALGORITHMS ACCURACY PRECISION RECALL F1-Score

K-Nearest 80% 0.76 0.82 0.79


Neighbour
DECISION TREE 92% 0.94 0.88 0.91

RANDOM 80% 0.77 0.79 0.78


FOREST
CLASSIFIER
SUPPORT 82% 0.78 0.82 0.80
VECTOR
MACHINE
Comparison of Different Algorithm
CONCLUSION

 Patients with heart failure are becoming more prevalent every day .
Early detection and treatment of heart disease can help to prevent
serious complications.

 This work presented a machine learning model that can predict the
early risk of heart disease.

 4 machine learning algorithms has been used to get the best accuracy .

 Decision Tree algorithm achieved the best 91.9% accuracy.


FUTURE WORK

 Developing Web App that can predict with real-time data.

 Implement more deep learning algorithms.

 Implement some new feature engineering approaches.

 Extend this work and will attempt to publish in a journal.


REFERENCES

 Senthil Kumar Mohan (2019).Effective Heart Disease Prediction


Using Hybrid Machine Learning Techniques.
 Sonam Nikhar (2016,June).Prediction of Heart Disease Using
Machine Learning Algorithms . International Journal of
Advanced Engineering , Management and Science(IJAEMS)
 V.V Ramalingam(2018).Heart Disease prediction using machine
learning techniques . International Journal of Engineering &
Technology.
 Vijeta Sharma(2020).Heart Disease Prediction using Machine
LearningTechniques.2nd International Conference on Advances
Computing , Communication Control and Networking(ICACCCN)

You might also like