You are on page 1of 15

Disease Prediction using

Machine Learning
Microproject Co-ordinator : Mrs S.M. Dhongade
 Team Member
Sr No Name Roll No.

1. Madhur Madane 21

2. Kalpesh Mahale 22
Introduction to
Disease Prediction

Types of Diseases
Predicted Using ML

Data Collection
and Preprocessing

CONTENTS ML Algorithms
for Disease Prediction
Introduction to
Disease Prediction
Disease Prediction

Disease Prediction
Machine learning for disease
prediction utilizes data analysis to
anticipate health risks based on
individual characteristics and medical
history. By recognizing patterns in
vast datasets, these models enable
early intervention and personalized
preventive measures, enhancing
healthcare outcomes and reducing
burdens on healthcare systems
Use Of Disease Prediction
.
risk assessment targeted prevention
Predictive models evaluate individual Strategies focus on high-risk individuals to
susceptibility to specific diseases. mitigate disease onset.

Early identification personalized interventions proactive healthcare.


Disease prediction enables early detection of Tailored preventive measures are Timely interventions and preventive
health risks. implemented based on risk assessment. strategies improve overall health outcomes.
Types of Diseases Predicted Using ML
Types of Diseases

Diabetes Neurological disorders


Predictive models utilize patient Machine learning techniques process
demographics, lifestyle factors, and medical neurological data to forecast the onset of
history to assess diabetes risk disorders such as Alzheimer's and Parkinson's
disease.

Cancer Infectious diseases


Predictive models assess the likelihood of
ML algorithms analyze various data sources to
infectious disease outbreaks based on factors
predict the likelihood of cancer development.
like climate, population density, and travel
patterns.
Data Collection and Preprocessing
Steps of data
Data Sources: Identification of diverse sources such Data Integration: Integration of multiple datasets
as electronic health records, medical imaging, from different sources to create a comprehensive
wearable devices, and genetic databases. dataset for analysis.

Data Cleaning: Removal of inconsistencies, missing Data Transformation: Normalization or


values, and outliers from the collected datasets to standardization of data to ensure uniformity and
ensure data quality. comparability across variables.

Handling Imbalance: Addressing class imbalance


Feature Selection: Identification of relevant features
issues by oversampling, undersampling, or using
or variables that contribute most to disease
techniques like SMOTE (Synthetic Minority Over-
prediction while reducing dimensionality.
sampling Technique).

Validation: Performing cross-validation or


Splitting Data: Partitioning the dataset into
training, validation, and testing sets to train and holdout validation to assess model performance
evaluate machine learning models effectively. and generalize predictions on unseen data.
ML Algorithms
for Disease Prediction
ML Algorithms

1 2 3 4 5

Logistic Regression Decision Trees Random Forests Support Vector Naive Bayes
Machines (SVM)
Algorithms

Logistic Regression: Suitable for binary classification


1 tasks, such as predicting the presence or absence of Support Vector Machines (SVM): Effective for both
a disease based on input features.
4 linear and nonlinear classification tasks, SVMs aim to
find the optimal hyperplane that separates classes in
feature space.

Decision Trees: Easily interpretable models that split


2 the data based on features to predict disease
outcomes

Naive Bayes: Probabilistic classifiers based on Bayes'


5 theorem, suitable for handling high-dimensional data
and making predictions quickly.

Random Forests: Ensemble learning method that


combines multiple decision trees to improve
3 prediction accuracy and handle complex datasets.
Python Code

https://colab.research.google.com/?u
tm_source=scs-index
Thank you
Microproject Co-ordinator : Mrs S.M. Dhongade

You might also like