The broad, tong-term objective of this Program Project Grant is to develop an effective imaging-based information and health care delivery system to support clinical practice, research, and education.
The specific aims of the grant are to: (1) evolve PACS into an effective infrastructure that promotes the objectification of subjective patient clinical symptoms, (2) develop methods for improving the characterization of medical data through structured data collection, natural language processing of medical reports (NLP) and parametric summarization for medical images, (3) provide flexible, patient -specific presentation methods of medical images, timelines, and structured medical data. The objectification, intelligent access, and flexible presentation of medical data provide better information, which will facilitate the evidence-based practice of medicine and enhance research and evaluation. Five integrated projects employ novel techniques to address specific elements of the system. Intelligently selected imaging protocols are used to objectify patient symptoms. Well-defined information units capture and structure diverse forms of data, whether directly or indirectly through NP for text of parametric summarization for images. Patient medical records are correlated with medical literature by content. Timelines organize the data into a format that allow medical events, their dependencies, and conditional trends to be easily visualized (Project 3). Scenario-based proxies provide up-to-date access to relevant medical information. Relaxation broadens queries to medical information when exact matches are not found. Software toolkits and user models enable user-, and domain-, and task-specific customizations. The hardware independent architecture will facilitate access to the system across different platforms and software subsystems. Together, they form a unique infrastructure that provides broad and intelligently customized access to well-defined structured data and up-to-date literature. This, in addition to patient-specific relevant data, expert opinion, and similar cases with known outcome, will promote the evidence-based practice of medicine. Five integrated projects employ novel techniques to address specific elements of the system. Intelligently secreted imaging protocols are used to objectify patient symptoms. Well-defined information units capture and structure diverse forms of data, whether directly or indirectly through NLP for text or parametric summarization for images. Patient medical records are correlated with medical literature by content. Timelines organize the data into a format that allow medical events, their dependencies, and conditional trends to be easily visualized (Project 3). Scenario-based proxies provide up-to-date access to relevant medical information. Relaxation broadens queries to medical information when exact matches are not found. Software toolkits and user models enable user-, and domain-, and task-specific customizations. The hardware independent architecture will facilitate access to the system across different platforms and software subsystems. Together, they form a unique infrastructure that provides broad and intelligently customized access to well-defined structured data and up-to-date literature. This, in addition to patient-specific relevant data, expert opinion, and similar cases with known outcome, will promote the evidence-based practice of medicine. Evaluation of the impact of the proposed system will focus on technical measures; process of care; and patient and physician satisfaction. The evaluation will also explore the relationship between process changes and specific outcomes, particularly short-term health related quality of life. Although a formal cost-effectiveness study is not proposed, the foundation is laid for these measurements when these PACs technologies mature. These measurements will be facilitated by recording resource utilization, determining of imaging-based episodes of care, and counter- specific information related to a chief complaint.
| Bashyam, Vijayaraghavan; Hsu, William; Watt, Emily et al. (2009) Problem-centric organization and visualization of patient imaging and clinical data. Radiographics 29:331-43 |
| Taira, Ricky K; Taira, Ricky; Bui, Alex et al. (2008) A tool for improving the longitudinal imaging characterization for neuro-oncology cases. AMIA Annu Symp Proc :712-6 |
| Son, Roderick Y; Taira, Ricky K; Kangarloo, Hooshang et al. (2008) Context-sensitive correlation of implicitly related data: an episode creation methodology. IEEE Trans Inf Technol Biomed 12:549-60 |
| Bui, Alex A T; Morioka, Craig; Dionisio, John David N et al. (2007) OpenSourcePACS: an extensible infrastructure for medical image management. IEEE Trans Inf Technol Biomed 11:94-109 |
| Taira, Ricky K; Bashyam, Vijayaraghavan; Kangarloo, Hooshang (2007) A field theoretical approach to medical natural language processing. IEEE Trans Inf Technol Biomed 11:364-75 |
| Bui, Alex A T; Aberle, Denise R; Kangarloo, Hooshang (2007) TimeLine: visualizing integrated patient records. IEEE Trans Inf Technol Biomed 11:462-73 |
| Ardekani, Siamak; Kumar, Anand; Bartzokis, George et al. (2007) Exploratory voxel-based analysis of diffusion indices and hemispheric asymmetry in normal aging. Magn Reson Imaging 25:154-67 |
| Ardekani, Siamak; Selva, Luis; Sayre, James et al. (2006) Quantitative metrics for evaluating parallel acquisition techniques in diffusion tensor imaging at 3 Tesla. Invest Radiol 41:806-14 |
| Ardekani, Siamak; Sinha, Usha (2006) Statistical representation of mean diffusivity and fractional anisotropy brain maps of normal subjects. J Magn Reson Imaging 24:1243-51 |
| Ardekani, Siamak; Sinha, Usha (2005) Geometric distortion correction of high-resolution 3 T diffusion tensor brain images. Magn Reson Med 54:1163-71 |
Showing the most recent 10 out of 21 publications
