1. Information Visualization Tools Ketan Mane Ph.D. Candidate Member of Information Visualization Lab Member of Cyberinfrastructure for Network Science School of Library and Information Science (SLIS) Indiana University, Bloomington, IN kmane@indiana.edu

2. This Presentation has Three Parts 1.Information Retrieval Systems 2.Knowledge Management Visualizations 3.Visual Computational Diagnostics of Acute Lymphoblastic Leukemia Patients

3. This Presentation has Three Parts 1.Information Retrieval Systems 2.Knowledge Management Visualizations 3.Visual Computational Diagnostics of Acute Lymphoblastic Leukemia Patients

4. Mane, Ketan & Börner, Katy. (2006). SRS Browser: A Visual Interface to Sequence Retrieval System Visualization and Data Analysis, San Jose, CA, SPIE-IS&T, Jan 15-19, 2006.

5. SRS Browser – Extends SRS Browser Functionality SRS System SRS Browser

6. Features of SRS Browser

7. Filter Network to Show Immediate Neighbors Features of SRS Browser

8. Oncosifter Hierarchical Visualization Interface  Graphical visualization reveal structure in data.  Cancer categories represent hierarchical tree data structure.  Radial tree is used to display cancer categories.  Category classification of cancer is easily available.  Minimum interaction needed to get information.

9. Hierarchical Search Interface and Corresponding Results Page Oncosifter

10. This Presentation has Three Parts 1.Information Retrieval Systems 2.Knowledge Management Visualizations 3.Visual Computational Diagnostics of Acute Lymphoblastic Leukemia Patients

11. Co-word space of the top 50 highly frequent and bursty words used in the top 10% most highly cited PNAS publications in 1982- 2001. Mane & Börner. (2004) PNAS, 101(Suppl. 1): 5287-5290. Mapping Topic Bursts 11

12. 12 Börner & Penumarthy.(2005) PNAS citations received by top U.S. institutions

13. Policy Economics Statistics Math CompSci Physics Biology GeoScience Microbiology BioChem Brain Psychiatry Environment Vision Virology Infectious Diseases Cancer MRI Bio- Materials Law Plant Animal Phys-Chem Chemistry Psychology Education Computer Tech GI Funding Patterns of the National Institutes of Health (NIH) 13 Science map applications: Identifying core competency

14. Policy Economics Statistics Math CompSci Physics Biology GeoScience Microbiology BioChem Brain Psychiatry Environment Vision Virology Infectious Diseases Cancer MRI Bio- Materials Law Plant Animal Phys-Chem Chemistry Psychology Education Computer Tech Funding Patterns of the National Science Foundation (NSF) Science map applications: Identifying core competency 14 Kevin W. Boyack & Richard Klavans, unpublished work.

15. Boyack, Kevin W., Mane, Ketan and Börner, Katy. (2004). Mapping Medline Papers, Genes, and Proteins Related to Melanoma Research. IV2004 Conference, London, UK, pp. 965-971.

17. Top Researched Genes/Proteins in Melanoma Research

18. Association Maps Gene-Paper Network Gene-Gene Network

19. Boyack, Kevin W., Mane, Ketan and Börner, Katy. (2004). Mapping Medline Papers, Genes, and Proteins Related to Melanoma Research. IV2004 Conference, London, UK, pp. 965-971.

20. This Presentation has Three Parts 1.Information Retrieval Systems 2.Knowledge Management Visualizations 3.Visual Computational Diagnostics of Acute Lymphoblastic Leukemia Patients

21. Computational Diagnostics  Visualization Goal: Identify factors that cause relapse in patients  Relapse insight can be gained by –  Global overview of medical condition of all patients in the dataset  Ability to identify worst medical condition in patients  Comparing patient medical condition at diagnostic variable(s) level  Ability to identify and compare patient groups that share similar medical condition across multiple variables Dr. Susanane Raggs Dr. Katy BörnerKetan ManeJulie Haydon Jada Pane

22. Computational Diagnostics – Tool Requested by Client Matrix visualization Phenotype and prognosis Parallel Coordinate Visualization Coupled Windows

23. Computational Diagnostics – Interactive Visualization System Architecture

24. Computational Diagnostics - Dataset Details Diagnostic data variables from medical records for Acute Lymphoblastic Leukemia (ALL) patients are categorized into a. Outcome Patient Variables: relapse, relapse site, alive/death status, and LDKA. b. Biology Patient Variables: immunophenotype, genetic condition, WBC, Hgb, platelets, and CNS. c. Host Patient Variables: diagnostic age (ageDx), gender, and race. d. Treatment Patient Variables: BM 7 and BM 14. e. Social Factors Patient Variables: MFI-class, education level, %single family members, and % family employment. All data was provided by Dr. Susanne Raggs, Julie Haydon and Jada Pane.

25. Matrix Visualization – Phenotype View  Data is shown independent of other variables.  Color codes help to provide a quick insight into patient medical condition.

26. Matrix Visualization – Prognosis View  Color codes indicate event free survival in percent (%EFS).  All variable values are dependent on other variable values.

27. Matrix Visualization – Combined View  Facilitates selection of phenotype/prognosis view for individual diagnostic variables.

28. Parallel Coordinates Visualization  Uses one axis for each data variable.  For each patient, all data values on different parallel axis are connected.  All patient graphs are shown here. Single or multiple patients can be selected and studied in detail.

29. Parallel Coordinates Visualization Tool-tip display to show diagnostic values of selected patient.

30. Parallel Coordinates Visualization – User Interactions Display axes-labels to mark different regions/values along axes  Numerical landmarks along axes showing values for quantitative variables.  Category labels used along axes show values for nominal variables.

31. Parallel Coordinates Visualization – User Interactions Display zones to show severity values for different variables  Triangular zones indicate variables with quantitative values.  Rectangular zones are used for variables with nominal values.

32. Parallel Coordinates Visualization – User Interactions Axis selection to study global variations in patient values  Single axis can be selected to study the trend in patient values.  Red-to-green gradient used to indicate values along the selected axis. [Red = High value, Green = Low value]

33. Parallel Coordinates Visualization A subset of patents can be selected and examined as a group.

34. Parallel Coordinates Visualization Simultaneous display of patient groups to study differences. Patient Group 1 Patient Group 2 Patient Group 1 & 2

35. Parallel Coordinates Visualization Multiple Coordinated Views  Patient can be selected and color coded in matrix view.  Corresponding patient lines are highlighted in parallel coordinate view.

36. Demo of Medical Diagnostics Project