Scribd
Upload
321 views26 pages

Data Mining in Healthcare

This document discusses data mining in healthcare and provides examples of how it can be used. It begins with an introduction to data mining and describes some of its main tasks like classification, clustering, and prediction. It then discusses how data mining can be applied in healthcare to improve healthcare quality by reducing errors, discover side effects of drugs, and determine disease patterns. Data mining can also help support health insurance through fraud detection and understanding customer needs. It concludes by mentioning some challenges of data mining in healthcare like data accessibility and privacy issues.

Uploaded by

Thuy Tien
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
321 views26 pages

Data Mining in Healthcare

This document discusses data mining in healthcare and provides examples of how it can be used. It begins with an introduction to data mining and describes some of its main tasks like classification, clustering, and prediction. It then discusses how data mining can be applied in healthcare to improve healthcare quality by reducing errors, discover side effects of drugs, and determine disease patterns. Data mining can also help support health insurance through fraud detection and understanding customer needs. It concludes by mentioning some challenges of data mining in healthcare like data accessibility and privacy issues.

Uploaded by

Thuy Tien
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 26

Data Mining in Healthcare

Nguyen Dao Thuy Tien - 9517


Phung Nguyen Anh Khoa - 9588
AGENDA
• Data mining in general
• Data mining in Healthcare
• Classification in data mining
• Naïve Bayes algorithm
• Demo
INTRODUCTION TO DATA MINING

• Data mining is the process of identifying hidden patterns and


establishing relationships in large databases to solve many problems.

• Both businesses and sciences benefit from Data Mining.


MAIN TASKS OF DATA MINING

• Anomaly detection (change/deviation detection)


• Dependency modelling
• Clustering
• Classification
• Regression
• Summarization
DATA MINING TECHNIQUES

• Association Rule
• Classification
• Clustering
• Prediction
• Sequential patterns
• Decision Tree
• Neural Networks
DATA MINING PROCESS
DATA MINING IN HEALTHCARE
• Healthcare covers a detailed processes of the diagnosis, treatment
and prevention of disease, injury and other physical and mental
impairments in humans.
• The number of electronic health records or EHRs has increased
quickly
Data mining can help in healthcare sector.
• Benefits of applying data mining in healthcare:
• Improving healthcare quality
• Supporting insurances
• Helping healthcare administration
IMPROVING HEALTHCARE QUALITY

• Reducing many errors and issues that lead to unnecessary deaths.

• Finding out patterns and anomalies better.

• Discovering the unreported side effects of common drugs.

• Determining disease by considering the pattern of the lifestyle.


SUPPORTING HEALTH INSURANCE

• Detecting fraudulent and abusive behavior.

• Attracting new consumers as well as keeping current ones.

• Helping insurance companies understand the health insurance plans


better.
HELPING HEALTHCARE ADMINISTRATION

• Making a profit.

• Managing resources.

• Minimizing costs.

• Maintaining good customer services.


PROBLEMS OF DATA MINING IN HEALTHCARE

• Healthcare data mining can be limited by the accessibility of data.

• Data in healthcare has quality problems.

• Legal and social issues, such as data ownership and privacy issues
related to healthcare data.

• Advanced knowledge is required to gain optimal results.


TRENDS IN THE FUTURE

• Develop more specific applications with reasonable price for


healthcare industry.

• Electronic medical records will replace traditional paper-based


records.

• The healthcare system will be real-time and deployed on the clouds.


Data Mining
Classification
• K-Nearest neighbor(kNN) • Neutral networks(NN)

• Decision trees • Support vector machines(SVM)

• Naïve Bayes • Linear regression(LR)

• Logistic regression
Naïve Bayes
ADVANTAGES DISADVANTAGES
• Fast • Strong assumption

• Scalable • Scarcity

• Big data require


• Variety
• Independent predictors
• Intelligent
CHARACTERISTICS
• Formatted
• Adequation
REAL WORLD DATA PREPARATION
• Discretization
• Cleaning
• Integration
• Transformation
• Reduction
APPLIED IN PROJECT
Implementation
LIBRARIES ALGORITHMS

• Naivebayes • Naïve Bayes

• Dplyr

• Equal-width Binning
• Ggplot

• Psych
DATA PROCESSING
• Defining
• Shrinking
• Calculating
SKETCHING
APPLY NAÏVE BAYES
THANK YOU FOR YOUR LISTENING!

You might also like