MINING YOUR OWN BUSINESS IN HEALTH CARE USING DB2 INTELLIGENT MINER FOR DATA
CONTENTS CHAPTER 1 INTRODUCTION 1 1 WHY YOU SHOULD MINE YOUR OWN BUSINESS 1 2 THE HEALTH CARE BUSINESS ISSUES TO ADDRESS 1 3 HOW THIS BOOK IS STRUCTURED 1 4 WHO SHOULD READ THIS BOOK? CHAPTER 2 BUSINESS INTELLIGENCE ARCHITECTURE OVERVIEW 2 1 BUSINESS INTELLIGENCE 2 2 DATA WAREHOUSE 2 2 1 DATA SOURCES 2 2 2 EXTRACTION/PROPAGATION 2 2 3 TRANSFORMATION/CLEANSING 2 2 4 DATA REFINING 2 2 5 DATAMARTS 2 2 6 METADATA 2 2 7 OPERATIONAL DATA STORE (ODS) 2 3 ANALYTICAL USERS REQUIREMENTS 2 3 1 REPORTING AND QUERY 2 3 2 ON-LINE ANALYTICAL PROCESSING (OLAP) 2 3 4 STATISTICS 2 3 5 DATA MINING 2 4 DATA WAREHOUSE, OLAP AND DATA MINING SUMMARY CHAPTER 3 A GENERIC DATA MINING METHOD 3 1 WHAT IS DATA MINING? 3 2 WHAT IS NEW WITH DATA MINING? 3 3 DATA MINING TECHNIQUES 3 3 1 TYPES OF TECHNIQUES 3 3 2 DIFFERENT APPLICATIONS THAT DATA MINING CAN BE USED FOR 3 4 THE GENERIC DATA MINING METHOD 3 4 1 STEP 1 — DEFINING THE BUSINESS ISSUE 3 4 2 STEP 2 — DEFINING A DATA MODEL TO USE 3 4 3 STEP 3 — SOURCING AND PREPROCESSING THE DATA 3 4 4 STEP 4 — EVALUATING THE DATA MODEL 3 4 5 STEP 5 — CHOOSING THE DATA MINING TECHNIQUE 3 4 6 STEP 6 — INTERPRETING THE RESULTS 3 4 7 STEP 7 — DEPLOYING THE RESULTS 3 4 8 SKILLS REQUIRED 3 4 9 EFFORT REQUIRED CHAPTER 4 HOW TO PERFORM WEIGHT RATING FOR DIAGNOSIS RELATED GROUPS BY USING MEDICAL DIAGNOSES 4 1 THE MEDICAL DOMAIN AND THE BUSINESS ISSUE 4 1 1 WHERE SHOULD WE START? 4 2 THE DATA TO BE USED 4 2 1 DIAGNOSES DATA FROM FIRST QUARTER 1999 4 2 2 INTERNATIONAL CLASSIFICATION OF DISEASES (ICD10) 4 3 SOURCING AND PREPROCESSING THE DATA 4 4 EVALUATING THE DATA 4 4 1 EVALUATING DIAGNOSES DATA 4 4 2 EVALUATING ICD10 CATALOG 4 4 3 LIMITING THE DATAMART 4 5 CHOOSING THE MINING TECHNIQUE 4 5 1 ABOUT THE COMMUNICATION BETWEEN EXPERTS 4 5 2 ABOUT VERIFICATION AND DISCOVERY 4 5 3 LET’S FIND ASSOCIATIVE RULES! 4 6 INTERPRETING THE RESULTS 4 6 1 FINDING APPROPRIATE ASSOCIATION RULES 4 6 2 ASSOCIATION DISCOVERY OVER TIME 4 7 DEPLOYING THE MINING RESULTS 4 7 1 WHAT WE DID SO FAR 4 7 2 PERFORMING WEIGHT RATING FOR DIAGNOSIS RELATED GROUPS CHAPTER 5 HOW TO PERFORM PATIENT PROFILING 5 1 THE MEDICAL DOMAIN AND THE BUSINESS ISSUE 5 1 1 DEEP VEIN THROMBOSIS 5 1 2 WHAT DOES DEEP VEIN THROMBOSIS CAUSE? 5 1 3 USING VENOGRAPHY TO DIAGNOSE DEEP VEIN THROMBOSIS 5 1 4 DEEP VEIN THROMBOSIS AND ICD10 5 1 5 WHERE SHOULD WE START? 5 2 THE DATA TO BE USED 5 3 SOURCING AND PREPROCESSING THE DATA 5 3 1 DEMOGRAPHIC DATA 5 3 2 DATA FROM MEDICAL TESTS 5 3 3 HISTORICAL MEDICAL TESTS 5 4 EVALUATING THE DATA 5 4 1 DEMOGRAPHIC DATA 5 4 2 DATA FROM MEDICAL TESTS 5 4 3 HISTORICAL MEDICAL TESTS 5 4 4 BUILDING A DATAMART 5 5 CHOOSING THE MINING TECHNIQUE 5 5 1 CHOOSING SEGMENTATION TECHNIQUE 5 5 2 USING CLASSIFICATION TREES FOR PREPROCESSING 5 5 3 APPLYING THE MODEL 5 6 INTERPRETING THE RESULTS 5 6 1 UNDERSTANDING CLUSTER 4 5 6 2 UNDERSTANDING CLUSTER 5 5 7 DEPLOYING THE MINING RESULTS 5 7 1 WHAT WE DID SO FAR 5 7 2 WHERE CAN THE METHOD BE DEPLOYED? CHAPTER 6 CAN WE OPTIMIZE MEDICAL PROPHYLAXIS TESTS? 6 1 THE MEDICAL DOMAIN AND THE BUSINESS ISSUE 6 1 1 DIABETES INSIPIDUS AND DIABETES MELLITUS 6 1 2 WHAT CAUSES DIABETES MELLITUS? 6 1 3 TESTS TO DIAGNOSE DIABETES MELLITUS 6 1 4 WHERE SHOULD WE START? 6 2 THE DATA TO BE USED 6 2 1 DIABETES MELLITUS AND ICD10 6 2 2 DATA STRUCTURE 6 2 3 SOME COMMENTS ABOUT THE QUALITY OF THE DATA 6 3 SOURCING AND EVALUATING DATA 6 3 1 STATISTICAL OVERVIEW 6 3 2 DATAMART AGGREGATION FOR ASSOCIATION DISCOVERY 6 4 CHOOSING THE MINING TECHNIQUE 6 5 INTERPRETING THE RESULTS 6 5 1 PREDICTIVE MODELING BY DECISION TREES 6 5 2 PREDICTIVE MODELING BY RADIAL BASIS FUNCTIONS 6 5 3 VERIFICATION OF THE PREDICTIVE MODELS 6 5 4 ASSOCIATION DISCOVERY ON TRANSACTIONAL DATAMART 6 6 DEPLOYING THE MINING RESULTS 6 6 1 WHAT WE DID SO FAR 6 6 2 OPTIMIZATION OF MEDICAL TESTS 6 6 3 BOOMERANG: IMPROVE THE COLLECTION OF DATA CHAPTER 7 CAN WE DETECT PRECAUSES FOR A SPECIAL MEDICAL CONDITION? 7 1 THE MEDICAL DOMAIN AND THE BUSINESS ISSUE 7 1 1 DEEP VEIN THROMBOSIS 7 1 2 WHAT DOES DEEP VEIN THROMBOSIS CAUSE? 7 1 3 CAN DEEP VEIN THROMBOSIS BE PREVENTED? 7 1 4 WHERE SHOULD WE START? 7 2 THE DATA TO BE USED 7 3 SOURCING THE DATA 7 4 EVALUATING THE DATA 7 4 1 THE NONDETERMINISTIC ISSUE 7 4 2 NEED FOR DIFFERENT AGGREGATIONS 7 4 3 ASSOCIATIVE AGGREGATION 7 4 4 TIME SERIES AGGREGATION 7 4 5 INVALID VALUES IN TIME SERIES AGGREGATION 7 5 CHOOSING THE MINING TECHNIQUE 7 5 1 ASSOCIATION DISCOVERY 7 5 2 SEQUENCE ANALYSIS 7 5 3 SIMILAR SEQUENCES 7 6 INTERPRETING THE RESULTS 7 6 1 RESULTS FOR ASSOCIATIVE AGGREGATION 7 6 2 RESULTS FOR TIME SERIES AGGREGATION 7 7 DEPLOYING THE MINING RESULTS 7 7 1 WHAT WE DID SO FAR 7 7 2 HOW CAN THE MODEL BE DEPLOYED? CHAPTER 8 THE VALUE OF DB2 INTELLIGENT MINER FOR DATA 8 1 WHAT BENEFITS DOES IM FOR DATA OFFER? 8 2 OVERVIEW OF IM FOR DATA 8 2 1 DATA PREPARATION FUNCTIONS 8 2 2 STATISTICAL FUNCTIONS 8 2 3 MINING FUNCTIONS 8 2 4 CREATING AND VISUALIZING THE RESULTS 8 3 DB2 INTELLIGENT MINER SCORING RELATED PUBLICATIONS IBM REDBOOKS OTHER RESOURCES REFERENCED WEB SITES HOW TO GET IBM REDBOOKS IBM REDBOOKS COLLECTIONS SPECIAL NOTICES