You are on page 1of 16

EAST WEST INSTITUTE OF TECHNOLOGY

DEPARTMENT OF INFORMATION SCIENCE AND


ENGINEERING
INTERNSHIP PRESENTATION ON

DEMENTIA DISEASE PREDICTION

Presented By
EXTERNAL GUIDE Dinesh N INTERNAL GUIDE
Mr. Mahesh Deginal (1EW20IS026) Ms. Prathyaksha
Trainer, Karunadu Asst. Prof
Technologies Dept. of ISE
TABLE OF CONTENTS

Abstract
Introduction
Company Profile
Proposed System
Implementation
Results
Applications
Conclusion
References
ABSTRACT
The proposed system aims to predict whether the person is demented or non-demented. The dataset
includes information about Subject_ ID, MRI_ ID, Visit, Delay, M/F, Hand, Age, Education[EDUC],
Mini-Mental State Examination[MMSE], Clinical Dementia Rating[CDR], Atlas Scale Factor[ASF],
Estimated Total Intracranial Volume[eTIV], Socio-Economic Status[SES] and Whole Body
Vibration[nWBV]. Data preprocessing techniques such as data cleaning, normalization, and feature
engineering are used to prepare data for the machine learning model. The model is trained on the
preprocessed data, and various algorithm such as Decision Tree, K-Nearest Neighbors[KNN],
Random Forest and Naive Bayes are implemented to check the model performance and accuracy.
The result of the proposed system can be useful to determine whether the person has dementia
disease or not.
INTRODUCTION
 Dementia is a debilitating and progressive neurological disorder that primarily affects cognitive
functions, including memory, thinking and reasoning. It can significantly impact a person's ability
to perform daily activities and lead to a decline in overall quality of life.
 The dementia disease prediction system is one such which can be used to predict whether the
person is demented or not based on the person’s medical information and personal data.
 Here, we use machine learning techniques to build the prediction model.
 Machine Learning is a branch of artificial intelligence and computer science which focuses on the
use of data and algorithms to imitate the way that humans learn, gradually improving it’s
accuracy.
 In other words it is broadly defined as the capability of a machine to imitate intelligent human
behavior.
 In summary, the results revealed that the Random Forest algorithm model achieved a
commendable accuracy in predicting dementia disease with an overall accuracy of 96%.
COMPANY PROFILE
 Karunadu Technologies Private Limited is a leading IT software solutions and services industry
incorporated on 19 July 2018. It is classified as Non-govt company and mainly focuses on
quality standards and customer values.
 The company offers broad range of customized software applications powered by concrete
technology and industry expertise. It also offers end to end embedded solutions and services.
 They deal with broad range of product development along with customized features ensuring at
most customer satisfaction and also empower individual with knowledge, skills and
competencies that assist them to escalate as integrated individuals with a sense of commitment
and dedication.
 Provide cost effective and reliable solutions to customers across various latest technologies.
 Create, design and deliver business solutions with high value and innovation by leveraging
technology expertise and innovative business models to address long-term business objectives.
PROPOSED SYSTEM
 Proposed system for dementia disease prediction involves integrating various components and
technologies to identify individuals at risk of developing dementia.
 First data frame contains all the values of 15 variables of the dataset and the second data frame
contains the filtered rows of the first data frame where the ‘Group’ column is equal to
‘Converted’ for making the model generalized towards classification of the disease as ‘Demented’
and ‘NonDemented.
 The dementia disease analysis system builds graphs using seaborn module to understand the data
frame.
 The dementia dataset contains different information about the details of person such as Subject_
ID, MRI_ ID, Visit, Delay, M/F, Hand, Age, EDUC, MMSE, CDR, ASF, eTIV, SES and nWBV. It
has 373 instances with 15 variables each.
 The system uses supervised machine learning algorithm such as Random Forest Classifier for
predicting whether a person has dementia disease or not.
IMPLEMENTATION
Dementia Disease Prediction:
 The dataset is read using the csv file and it is analyzed using bar graphs.
 The data preprocessing technique is applied to the dataset to split the data into input data and
output data.
 The machine learning model is trained with the preprocessed data using the machine learning
algorithm such as Random Forest Classifier to predict whether a person is demented or non-
demented.
 Other various machine learning algorithms are also applied to predict the results based on the
input data and to compare each model accuracy.
 Along with the predicted result , the model accuracy is also calculated to determine how accurate
the model is predicting the result based on the given input data .
RESULTS

DISTRIBUTION OF ‘GROUP’ BY DISTRIBUTION OF EDUCATION FOR


‘GENDER’ EACH ‘GENDER’ AND ‘GROUP’
HEATMAP TO VISUALIZE DISTRIBUTION OF AGE FOR EACH
CORRELATION OF ATTRIBUTES GROUP
CLASSIFICATION OF EDUCATION CLASSIFICATION OF AGE
CLASSIFICATION OF CLINICAL CLASSIFICATION OF MINI-MENTAL
DEMENTIA RATING STATE EXAMINATION
APPLICATIONS
• A healthcare industry can make use of this proposed system to determine whether the person is
demented or not.
• HealthCare industries can make use of such model to predict the disease type upon the inputs
and x-ray images of the patient and provide the result with acceptable accuracy.
• MindMate, MEDSCAN are the ultimate platform for people with dementia, thier families and
caregivers which helps them to stay independent.
• Random Forest algorithm works well if we have trained the model using real-time dataset.
After training the evaluation process is performed and we come across the commendable
accuracy.
CONCLUSION

In conclusion, this proposed systemt aimed to develop a predictive model for dementia disease
based on a comprehensive analysis of relevant data. We successfully gathered a diverse dataset
containing various personal and clinical features of individuals. Extensive data preprocessing and
cleaning were performed to ensure the quality and reliability of the dataset. We have used Random
Forest Classifier prediction model to determine whether a person is affected by dementia disease or
not . The ability to do so provides a strong predictive tool for online healthcare platform to have a
better understanding of their patients and to improve their services accordingly. We hope that this
proposed system will be helpful for AI and ML researchers and medical practitioners who are
working in the domain of automated diagnostic systems for dementia prediction.
REFERENCES

 www.karunadutechnologies.com
 Ayodele, Taiwo Oladipupo, “Types of machine learning algorithms”. In New advances in machine
learning. Intech Open, 2010
 www.kaggle.com
THANK YOU

You might also like