CS 410/510 Top: Introduction to Healthcare Data Analytics

Credit Hours:

4/3

Course Coordinator:

Shuwen Wang

Course Description:

Healthcare systems generate vast amounts of data, ranging from electronic health records (EHR) and medical imaging to administrative records and patient-generated data. This course provides an introduction to the fundamental concepts, techniques, and tools used in analyzing healthcare data to derive meaningful insights and improve patient outcomes. Students will learn the basics of data collection, preprocessing, analysis, and interpretation specifically tailored to the healthcare domain.

Prerequisites:

General understanding of machine learning concepts.

Goals:

Upon successful completion of this class, students will be able to:

1. Describe the unique challenges and opportunities in healthcare data analytics.
2. Design preprocessing pipelines for healthcare data.
3. Analyze and uncover the hidden patterns of healthcare data.
4. Interpret and use the statistical models built from healthcare data.
5. Apply data analytics skills to solve real-world healthcare problems.

Textbooks:

Not Required.

Healthcare Data Analytics (Chapman & Hall/CRC Data Mining and Knowledge Discovery Series) by Chandan K. Reddy, Charu C. Aggarwal (Recommended).

References:

None.

Major Topics:

1. Introduction to Healthcare Data:

• Overview of healthcare data sources (HER, medical imaging, etc.)
• Data privacy, security, and ethical considerations in healthcare analytics

2. Exploratory Data Analysis (EDA) in Healthcare:

• Data preprocessing and cleaning
• Data normalization and transformation techniques
• Descriptive statistics and visualization methods for healthcare data
• Identifying patterns and trends in medical data

3. Predictive Modeling in Healthcare:

• Introduction to predictive analytics and machine learning
• Building predictive models for disease diagnosis, prognosis, and patient risk stratification

4. Descriptive Modeling in Healthcare:

• Association rule mining for identifying co-occurrences in medical data
• Clustering methods for patient segmentation and cohort analysis

5. Natural Language Processing (NLP) in Healthcare:

• Extracting insights from clinical notes and medical literature
• Sentiment analysis and topic modeling in healthcare text data

6. Applications of Healthcare Data Analytics:

• Clinical decision support systems
• Population health management
• Healthcare fraud detection and prevention