Informatics and Information in in Radiation Oncology : OncoSpace - PowerPoint PPT Presentation

Informatics and Information in in Radiation Oncology : OncoSpace John W. Wong, Ph.D. Todd McNutt, Ph.D. Department of Radiation Oncology Supported in Part by Elekta Impac Supported in Part by Elekta-Impac OCI_50 th 2008, JWW

Status of health care information Status of health care information “Today, most business ---- down to the smallest corner y, grocery store have better information about their sales and inventories than even affluent medical practices have about their patients practices have about their patients. …………. ” --- Michael Bloomberg Michael Bloomberg to the Academy National Health Policy Conference, 05/07 OCI_50 th 2008, JWW

NIH Roadmap 3: Re-engineering the Clinical Research Enterprise • R.A. Harrington, M.D., Duke Research Institute (2005) : g , , ( ) – “One of the greatest inefficiencies of the current model of clinical research in our country is the lack of a sustaining infrastructure (which includes shared t i i i f t t ( hi h i l d h d resources, common data standards, and effective use of information technology among researchers), as well as the lack of a convenient forum to share best practices and learn from one another’s mistakes and successes” successes . OCI_50 th 2008, JWW

Cancer Bio informatics Grid (caBIG) Cancer Bio-informatics Grid (caBIG) • NCI: cancer Bioinformatics Grid (caBIG) provides infra- structure support – clinical trials management systems – integrative cancer research integrative cancer research – tissue banks and pathology – Image workspace Image workspace • Not directed to address specific research or clinical questions q OCI_50 th 2008, JWW

Data Loss at the Institutional Level Data “Loss” at the Institutional Level • Data we are capturing – Labs, Images, Treatment Plans • Data we are sending away – Patients in protocols Patients in protocols • Data we are storing – Disparate databases • Data (experience) we are not capturing Data (experience) we are not capturing – Discarded treatment plans (and decision making process) • Information and knowledge are Not captured systematically g p y y • Not utilized efficiently to impact research and patient care OCI_50 th 2008, JWW

Challenges of data longevity and re-use Challenges of data longevity and re use • RTOG – Formed 1968, funded since 1971 – Activated 300 trials • 40 on-going • 60,000 patients enrolled – Established QA, credentialing process for RTP and Q f dosimetry – Centralized date repository; lacks secondary research Centralized date repository; lacks secondary research – No measure of impact on community practice OCI_50 th 2008, JWW

Multiple Informatics Initiatives at JHU Multiple Informatics Initiatives at JHU • Johns Hopkins University Health Systems – Committee for Health Informatics C itt f H lth I f ti – Johns Hopkins Medical Image Archive (JHMIA) – I 4 M: Integration of Imaging, Information and Intervention in g g g, Medicine – Clinical Trial Groups – Industrial collaborations Industrial collaborations • Microsoft (Almaga -- Healthcare Informatics) • IBM (Computational Medicine) • Harris Corporation (Multi-disciplinary clinic) • ………… OCI_50 th 2008, JWW

The JHMIA Program– (Radiology) e J og a ( ad o ogy) A single archive where all medical images and other non textual data (and associated reports etc ) from non-textual data (and associated reports, etc.) from across a Healthcare Enterprise are stored • Clinical (300 TB now --- 700 TB in 2 years) • Research • Waveform • Genomic (planned) • Proteomic (planned) P t i ( l d) • Medical Image Archive � Medical Data Archive? • Medical Image Archive � Medical Data Archive? OCI_50 th 2008, JWW

JHMIA A An Enterprise Image Archive E t i I A hi Current Participants Committed • JHH Radiology • BMC Vascular • JHH Vascular Surgery • BMC Adult Echo Cardio • JHH Peds Cardiology • Endoscopy (GI) • JHH Rad Oncology • JHH Adult Echo Cardio Potential • Surgery g y • OB/GYN • Cardiology • Pathology • BMC Radiology • Howard County General Hospital • Ophthalmology p gy OCI_50 th 2008, JWW

JHMIA and I 4 M JHMIA and I M Analytic Database(s): Analytic Database(s): Query and Security Analytic and Change Tools: Extraction of Information Web-service Decision Support: Data-mining Statistical Modeling • Goal: To improve both medical research and patient care OCI_50 th 2008, JWW

The enterprise model: JHMIA and I 4 M The enterprise model: JHMIA and I M Robotic Radiation Ophthalmology Surgery Oncology Infra-structure: JHMIA Radiology TeraMedica Infra-structure: JHMIA, Radiology, TeraMedica, Analytic Database Shape and Change Tools Data-mining Decision Support • Challenges to implement across multi-disciplines: – Data Standards – Workflow, Procedure, and Management Differences W kfl P d d M t Diff • Different intervention time-scale OCI_50 th 2008, JWW

OncoSpace: p Closed Loop Adaptive Radiation Therapy Patient Data Data – Information – Intervention – Response Information Intervention Response Real Time Image Guided Intervention Treatment Re-optimization; Early Treatment Assessment,… T t t R ti i ti E l T t t A t Population : New protocol, New dose level, New standards OncoSpace Infrastructure OncoSpace Infrastructure OCI_50 th 2008, JWW

OncoSpace: OncoSpace: Radiation Oncology as the I 4 M test-bed Radiation Robotic Ophthalmology Oncology Surgery Infra-structure: JHMIA, Radiology, TeraMedica, Analytic Database Shape and Change Tools Decision Support Decision Support Data-mining OncoSpace OCI_50 th 2008, JWW

Extending the OncoSpace Model: g p Sharing Research and Clinical Care Institute n I4M Infr ra-structu Institute 1 ure Ophthalmology JHU JHU Genomics I4M Infra-structure Radiation Oncology OCI_50 th 2008, JWW

OncoSpace • OncoSpace is a new research infra-structure based on p Radiation Oncology as a “use-case” model • Bioengineering Research Partnership – multi-disciplinary: Radiation Oncology, Radiology, lti di i li R di ti O l R di l Physics and Astronomy, Computer Science and Biostatistics – multi-institutional: Hopkins, clinical partner sites – IMPAC OCI_50 th 2008, JWW

Distributed Research Model Distributed Research Model Current Trial Practice Hypothetical Future Practice Patient Tx Patient Tx Follow up Follow up Treatment Treatment Treatment T t t Protocol Protocol Journal Publications Journal Publication of Publication Data to DB’s STOP Increased potential for data reuse START START OVER OCI_50 th 2008, JWW

Data Delivery in Cooperative Research: Data Delivery in Cooperative Research: Hitting the Wall FTP and GREP are not adequate q • You can FTP 1 MB in 1 sec • You can GREP 1 MB in a second • You can GREP 1 GB in a minute • You can FTP 1 GB / min (~1 $/GB) $/GB) • You can GREP 1 TB in 2 days You can GREP 1 TB in 2 days • You can GREP 1 PB in 3 years • 2 days;1K$ / 3 years and 1M$ • 50 MB l 50 MB local DICOM transfer takes 1 min l DICOM t f t k 1 i • 100 patients x 10 (3D) scans = 5 - 10 TB • A factor of 10 improvement in access speed cannot offset the growth in data and cannot offset the growth in data and complexity • R thi k d t b Rethink databases’ function ’ f ti – following the CS community OCI_50 th 2008, JWW

OncoSpace: Adapting the SkyServer Approach p p g y pp • SDSS is a collaborative • Shared resources effort to map 25% of the – Methodology sky – Software – Expertise • SkyServer publishes data – Experience p from the SDSS from the SDSS • New opportunities • >> 100’s of new – Analysis discoveries in astrophysics – Visualization Visualization • Increased scale and scope – User experience for research • Skyserver.sdss.org Alex Szalay PhD - JHU Jim Gray PhD - Microsoft OCI_50 th 2008, JWW

OncoSpace: Adapting the SkyServer Approach O coSpace dapt g t e S ySe e pp oac • Active Databases • There is too much data to move around, take the analysis to the data! • Do all data manipulations at database – Build custom procedures and functions in the B ild t d d f ti i th database • Established Web-service for broad access Established Web service for broad access – Query across multiple databases • Automatic parallelism guaranteed p g OCI_50 th 2008, JWW

Database Consideration Database Consideration Operational vs Analytical iltering • Workflow management • Decision Support • Patient records and archival • On-line Analytical Processing • Multidimensional analysis • Time variant on & Fi • Non-volatile How might I analyze my data? How do I organize my data? Star Schema T ypically Hierarchical D Design to support analysis… i t t l i DICOM RT Extracti Fast query OO principles Dimension Dimension Dimension FACTS Dimension Dimension Data E OIS, OCIS, EPR OncoSpace TPS, PACS OCI_50 th 2008, JWW

OncoSpace: Work in progress (McNutt) Membership OncoSpace Vocabulary D t Data Preparation Analytical • Technologies Database – SQL Server 2005 – Ruby on Rails Ruby on Rails OCI_50 th 2008, JWW