Medical Data Analysis and Mining

Overview

"Medical Data Analysis and Mining" is an applied course that integrates content from multiple disciplines. Centered on the capability requirements in medical data analysis, it is guided by medical problems and links various data analysis and mining methods. Oriented by medical data exploration, it effectively cleans and organizes data. Driven by the multi-angle analysis and presentation of results from traditional Chinese medicine and other perspectives, it visualizes the outcomes of algorithm models. The course objectives are set based on the student's ability to process, analyze, and present medical data. The teaching is conducted through case studies, covering topics such as introduction, data feature analysis, data preprocessing, association rule mining, classification and regression algorithms, clustering algorithms, and algorithm applications. The instructional design adopts a case-based approach, with the fundamental principles connecting the various stages. Through the gradual implementation of classic algorithm case models, students can familiarize themselves with the usage of programming tools while also presenting the results of data analysis and mining, exploring and refining the patterns contained in medical data.

Syllabus

Chapter1 Introduction

1.1 Overview of medical data analysis and mining
1.2 Introduction to common tools

Chapter2 Data Feature Analysis

2.1 Data type
2.2 Descriptive characteristics of the data
2.3 Correlation analysis of data
2.4 Principal component analysis

Chapter3 Data Processing

3.1 Data cleaning
3.2 Transformation of data
3.3 Aggregation of data

Chapter4 Association Rule Mining

4.1 Frequent pattern mining algorithm based on candidate item set generation and test strategy
4.2 Case: TCM syndrome association rules mining
4.3 Frequent pattern mining algorithms that generate post-option sets are not required

Chapter5 Classification and Regression

5.1 Classification and regression overview
5.2 Model Evaluation
5.3 Decision Tree Algorithm
5.4 Support vector machine
5.5 Ensemble learning algorithm
5.6 Case: Screening of growth influence indicators of salvia miltiorrhiza
5.7 Case: Influencing factors of lung cancer metastasis and survival analysis

Chapter6 Clustering

6.1 Overview of clustering algorithms
6.2 Clustering algorithm based on partition
6.3 Case: Credit card risk identification
6.4 Hierarchical clustering algorithm
6.5 Case: Authentic medicinal materials text data mining
6.6 Density based clustering algorithm

Chapter7 Algorithm Application

7.1 Case: Analysis of COVID-19 infection and prevention and control policies in children
7.2 Case: Estimation of COVID-19 case fatality by spline method
7.3 Case: Analysis of diagnosis and treatment efficiency in the early stage of COVID-19
7.4 Case: Text mining of traditional Chinese medicine syndrome of diabetes