Comparing Apples and Oranges— Contrast-set Mining: A Survey

Comparing Apples and Oranges— Contrast-set Mining: A Survey

Faculty of Computer Science A Data Warehouse Architecture for Clinical Data Warehousing Tony R. Sahama and Peter R. Croll Amit Satsangi [email protected] CMPUT 605 December 06, 2007February 11, 2008

2006 Department of Computing Science Focus Why are Clinical Data Warehouses (CDW) needed? Issues in their construction Design & design-choices in the construction of a CDW CMPUT 605 2006

Department of Computing Science Why Clinical Data Warehouse? Efficient Storage Uniformity in storage and querying of data Timely analysis Quality of decision making and analytics Decision based on larger sized datasets More accurate information Better strategies and research methods CMPUT 605

2006 Department of Computing Science Why Clinical Data Warehouse? Measurement of the effectiveness of treatment Relationships between causality and treatment protocols Safety Management Breakdown of cost, and charge information Forecasting demand

Better strategies and research methods CMPUT 605 2006 Department of Computing Science Some Facts Large volume of data distributed in a number of small repositoriesislands of information Data has great scientific and medical insight Great potential for people practicing clinical

medicine CMPUT 605 2006 Department of Computing Science Issues Heterogeneitydifferent clinical practices e.g. public vs. private hospitals Data Location Technical platforms & data formats

Organizational behaviors on processing the data Varying cultures amongst data management population CMPUT 605 2006 Department of Computing Science Past efforts Szirbik et al. Medical data Warehouse for elderly patients Six methodological steps to build medical data warehouses for

research. International Journal of Medical Informatics 75 (9): 683691 Used Rational Unified process (RUP) framework Identification of current trends (critical requirements of future) Data Modelling Ontology Building Quality Management and exception handling CMPUT 605 2006 Department of Computing Science

Different DW Architectures (Sen & Sinha 2005) CMPUT 605 2006 Department of Computing Science Design and Planning Business Analytics Approachunderstand the key processes of the business DW architect + Business Analyst + Expected Users Understand Key business processes + the

questions that would be asked of those processes Analysis might be conducted on demographic, diagnosis, severity of illness, length of stay CMPUT 605 2006 Department of Computing Science Approach Integration of data from two Biomedical Knowledge Repositories (BKRs)Oncology & Mental care

Used SAS Data Warehouse Administrator (SAS 2002) Flexibility to integrate external data repositories Hassle-free ETL Analytics with Data Miner Reporting using SAS Enterprise Guide (EG) Operational Data Store Architecture & Distributed Data Warehouse Architecture CMPUT 605 2006 Department of Computing Science

Several data marts to include different administration and management operations Summary reports Monitoring of clinical outcomes by management CMPUT 605 2006 Department of Computing Science Oncology Patient Management

CMPUT 605 2006 Department of Computing Science Mental Health Patient Management CMPUT 605 2006

Department of Computing Science Data Transformation Source systems CDW (ETL ExtractionTransformation-Load) Data preparation & Integration takes 90% of the effort in a given CDW project Excel, SAS External File Interface (EFI) & SAS Enterprise Guide (EG) used to clean the data CMPUT 605 2006

Department of Computing Science Steps in creation of CDW Step 1: Data imported in SAS Standardization into SAS table format Opportunity for data manipulationcreate/delete columns Step 2: Creation of metadata using Operational Data definition Step 3: Creation and loading of Data Tables Different tables for predictive and Database analysis Creation of multi-dimensional cubes

CMPUT 605 2006 Department of Computing Science Discussion Data acquisition step took very longvery little time left for cleaning, transformation Not enough time left to refine the shared environment (no modifications to their interface implementation etc.)

Security issues of federated Data Warehouses anonymization of records CMPUT 605 2006 Department of Computing Science Discussion SAS EM used to interpret relationships between seemingly unconnected data Newer CDW models coming from Case-based, Rolebased & evidence-based data structures need to be

incorporated CMPUT 605 2006 Department of Computing Science Steps in creation of CDW Step 4: Data Mining Tools integrable with or within SAS used EM, EG etc. CMPUT 605

2006 Department of Computing Science Thank You For Your Attention! CMPUT 605 2006

Recently Viewed Presentations

  • MIS 205 LECTURE 1 - Weebly

    MIS 205 LECTURE 1 - Weebly

    MIS 205 LECTURE 3 INFORMATION SYSTEMS APPLICATIONS & CASE STUDIES ... TECHNOLOGICAL VIEW OF INFORMATION SYSTEMS MIS 105 IT infrastructure Computer system hardware Computer system software Communication tools The company's network INTERNET, INTRANET, EXTRANET BUSINESS & INFORMATION SYSTEMS ...
  • Narratives

    Narratives

    An account of a sequence of events, usually in the order that they occurred. Narrative is the general term for telling a story. What is a Narrative? There are two main kinds of narratives: Fictional (made up) Nonfiction (true) Types...
  • Structure activity relationship (SAR) Adrenergic drugs

    Structure activity relationship (SAR) Adrenergic drugs

    SAR. Separation of aromatic and amino group - greatest sympathomimetic activity occurs when two carbon atoms separate the ring from amino group (DA, NA, AD etc) . Substitution on the amino group- increase in the size of alkyl substituents increases...
  • SURROUND THE LUXURY OF STUNNING SOUND PERFECT PITCH

    SURROUND THE LUXURY OF STUNNING SOUND PERFECT PITCH

    Chromecast built-in* and AirPlay 2 (OTA Q4 '19) Up to 24-bit/96kHz HD audio streaming. 4x HDMI inputs with 4K support. 1x HDMI output with ARC. Dolby Audio and DTS Digital Surround. Bluetooth wireless music streaming * Compatible with Harman Kardon...
  • MedEd My AFP Experience Matthew Machin Vascular Surgery

    MedEd My AFP Experience Matthew Machin Vascular Surgery

    Matthew Machin. Vascular Surgery HF at Imperial College London . MedEd. IT IS A VERY GOOD IDEA TO APPLY FOR THE AFP. Med. Ed. APPLY. Research interest. ACF PhD. Autonomy - the SJT . Surgery. 4 months of sociable hours...
  • Penguin Chick By Betty Tatham Illustrated by Helen

    Penguin Chick By Betty Tatham Illustrated by Helen

    The penguin uses his . flippers. to. propel himself through the water. What other animals have . flippers? frozen. Verb: hardened with cold; turned to ice. The ice was . frozen. so the ice skaters . could skate without falling...
  • Interlinking and annotating (parts of) images

    Interlinking and annotating (parts of) images

    Workshop: LitLink: A Cue Card System in a Research Environment of Collaborative Work, Online Publishing and GIS, Heidelberg, February 25, 2010. Peter [email protected] Agenda. Introducethe Cluster. Challengesforthe HRA.
  • NEW COVER SLIDE- qinfo with p & a - U of T Physics

    NEW COVER SLIDE- qinfo with p & a - U of T Physics

    (Eric Cornell paraphrasing Joseph Heller) HOW COLD MUST YOU GET? Need 1 atom per cubic wavelength… wavelength = 1 / momentum… water would need to be near 3 degrees K, but solidifies first (cf He!) Alkalis solidify too; need densities...