1 / 13

Jim Ong Trinka Coster, MD Kevin Leary, MD Ida Sim, MD PhD Stephen Porter, MD

ATA 2007. Goal and Assumptions. .. GoalCurrent state-of-the-artLimitations. Help clinicians review patient data more effectively before, during, and after patient encounters to improve patient care and increase patient safety.EMRs that employ web application servers, relational databases

serge
Download Presentation

Jim Ong Trinka Coster, MD Kevin Leary, MD Ida Sim, MD PhD Stephen Porter, MD

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


    2. ATA 2007 Goal and Assumptions

    3. ATA 2007 Our Approach Patient data views - Each view facilitates rapid data review and support best clinical practices via effective integration and presentation of a clinically-meaningful subset of the patient’s data. For example, a view could present patient data from the perspective of a particular medical problem, body system, or demographic group. Longer term, IPDRA will automatically select relevant views for each encounter, based on current complaint, patient history, clinician specialty and preferences, visit type Information-dense graphics – Multivariate graphical displays enable clinicians to see patterns and trends spanning diverse clinical data (e.g., clinical events, labs, meds). Mouse rollover, popup windows, and web navigation provide quick access to additional details and related information. Our graphical displays are inspired by (but extend) prior work by Plaisant and Schneiderman, Powsner and Tufte, Pharmaceutical Product Development. Extensions include combining time-series and timeline graphs within a single coordinated display, full web support (graphs can be rendered by Java applet in browser or by Java web server, add’l info referenced by web URL can be accessed by clicking on datapoints.) Web access – web browser accesses views generated by IPDRA web application server. Visual view authoring – View logic defines the database retrievals, data transformations, and data presentation needed to generate each view. IPDRA applies and extends SimBionic, a tool for authoring complex logic visually by drawing flow-charts and finite-state machines that can be run within Java and C++ programs. Patient data views - Each view facilitates rapid data review and support best clinical practices via effective integration and presentation of a clinically-meaningful subset of the patient’s data. For example, a view could present patient data from the perspective of a particular medical problem, body system, or demographic group. Longer term, IPDRA will automatically select relevant views for each encounter, based on current complaint, patient history, clinician specialty and preferences, visit type Information-dense graphics – Multivariate graphical displays enable clinicians to see patterns and trends spanning diverse clinical data (e.g., clinical events, labs, meds). Mouse rollover, popup windows, and web navigation provide quick access to additional details and related information. Our graphical displays are inspired by (but extend) prior work by Plaisant and Schneiderman, Powsner and Tufte, Pharmaceutical Product Development. Extensions include combining time-series and timeline graphs within a single coordinated display, full web support (graphs can be rendered by Java applet in browser or by Java web server, add’l info referenced by web URL can be accessed by clicking on datapoints.) Web access – web browser accesses views generated by IPDRA web application server. Visual view authoring – View logic defines the database retrievals, data transformations, and data presentation needed to generate each view. IPDRA applies and extends SimBionic, a tool for authoring complex logic visually by drawing flow-charts and finite-state machines that can be run within Java and C++ programs.

    4. ATA 2007 View Design Method The data retrieval, transformation, and presentation logic of each view is specified by drawing flow charts. Each rectangular nodes specifies calculations and actions. Each oval node specifies conditions that control which path within the flow chart is taken. A node can call a lower-level flow-chart, so complex view logic can be decomposed into multiple flow charts, organized in a hierarchy. Additional optional info – The green node indicates the flow chart’s “entry” node – execution of the flow chart starts at the green node. A red node indicates a “final” node. Execution of the flow chart ends at the red node. The “Catalog” pane on the left side of this window displays objects that comprise the view definitions, such as flow charts, functions that can be called by the flow charts, and global variables. These objects can be grouped within folders. A step-by-step debugger lets users execute views, step-by-step, to verify the correctness of each view.The data retrieval, transformation, and presentation logic of each view is specified by drawing flow charts. Each rectangular nodes specifies calculations and actions. Each oval node specifies conditions that control which path within the flow chart is taken. A node can call a lower-level flow-chart, so complex view logic can be decomposed into multiple flow charts, organized in a hierarchy. Additional optional info – The green node indicates the flow chart’s “entry” node – execution of the flow chart starts at the green node. A red node indicates a “final” node. Execution of the flow chart ends at the red node. The “Catalog” pane on the left side of this window displays objects that comprise the view definitions, such as flow charts, functions that can be called by the flow charts, and global variables. These objects can be grouped within folders. A step-by-step debugger lets users execute views, step-by-step, to verify the correctness of each view.

    5. ATA 2007 In this example, there are five modules labeled Vitals, Events, Studies, Lab Tests, and Medications (medications not shown) that contain vertically-stacked timelines. Symbols show measurements and events associated with points in time. Symbol size, shape, and color helps encode qualitative attributes and highlight significant data points. You can show or hide each module by clicking on the triangle next to the module’s label. This feature is useful when there are many graphs and timelines that cannot be shown all at once. When you click on a graph, a red vertical line is displayed in all graphs to help you compare data points at the time.In this example, there are five modules labeled Vitals, Events, Studies, Lab Tests, and Medications (medications not shown) that contain vertically-stacked timelines. Symbols show measurements and events associated with points in time. Symbol size, shape, and color helps encode qualitative attributes and highlight significant data points. You can show or hide each module by clicking on the triangle next to the module’s label. This feature is useful when there are many graphs and timelines that cannot be shown all at once. When you click on a graph, a red vertical line is displayed in all graphs to help you compare data points at the time.

    6. ATA 2007 In most graphs in this graph set that display a single clinical variable, a circle shows a value in normal range, an upward triangle shows a high value, and a downward triangle shows a low value. Blue lines show target values, and blue, yellow, and red regions show variable ranges for increasing levels of severity. Glucose Tolerance – (this example graph set contains no glucose tolerance data). Circles show normal range values for glucose at 2 hrs. Upward triangles show high values. The yellow region indicates Impaired Glucose Tolerance, and the red region indicates diagnosis for diabetes. Plus signs and crosses show glucose at 1 hr and 3 hrs, respectively. Glycosylated Hemoglobin – the green region shows the normal range. BP – Normal systolic and diastolic BP values are shown as plus signs and crosses, respectively. High values are shown as upward triangles. In most graphs in this graph set that display a single clinical variable, a circle shows a value in normal range, an upward triangle shows a high value, and a downward triangle shows a low value. Blue lines show target values, and blue, yellow, and red regions show variable ranges for increasing levels of severity. Glucose Tolerance – (this example graph set contains no glucose tolerance data). Circles show normal range values for glucose at 2 hrs. Upward triangles show high values. The yellow region indicates Impaired Glucose Tolerance, and the red region indicates diagnosis for diabetes. Plus signs and crosses show glucose at 1 hr and 3 hrs, respectively. Glycosylated Hemoglobin – the green region shows the normal range. BP – Normal systolic and diastolic BP values are shown as plus signs and crosses, respectively. High values are shown as upward triangles.

    7. ATA 2007 These time series graphs show normal ranges for each variable using grid lines that partially span the height of each graph. The timelines at the bottom of the graph container show periods of drug intake. When users click on a note in the right part of the display, the note is highlighted in red, and a vertical highlight line is displayed in all graphs to show the date of the note in relation to the data points and time intervals. This example is inspired by the psychiatric patient summary designed by Powsner and Tufte.These time series graphs show normal ranges for each variable using grid lines that partially span the height of each graph. The timelines at the bottom of the graph container show periods of drug intake. When users click on a note in the right part of the display, the note is highlighted in red, and a vertical highlight line is displayed in all graphs to show the date of the note in relation to the data points and time intervals. This example is inspired by the psychiatric patient summary designed by Powsner and Tufte.

    8. ATA 2007 This mockup of a pediatric weight chart shows reference ranges that change over time. Colored regions represents percentile ranges. The blue reference line indicates median weight vs. age. Reference ranges for boys are shown at left, girls on the right. This mockup of a pediatric weight chart shows reference ranges that change over time. Colored regions represents percentile ranges. The blue reference line indicates median weight vs. age. Reference ranges for boys are shown at left, girls on the right.

    9. ATA 2007 The design of this hypertension view was guided by patient data review recommendations specified by the JNC7 draft guidelines. The initial subview within this view shows blood pressure and heart rate. The blue region indicates Stage 1 hypertension. Yellow and red regions indicate Stage 2 and Stage 3 hypertension, respectively. Filled symbols indicate blood pressure measurements that exceed the threshold for Stage 1 hypertension. These time series graphs are vertically aligned with timelines that show medications over time. Other subviews show physical exam data, risk factors, related problems, and lab data relevant to hypertension.The design of this hypertension view was guided by patient data review recommendations specified by the JNC7 draft guidelines. The initial subview within this view shows blood pressure and heart rate. The blue region indicates Stage 1 hypertension. Yellow and red regions indicate Stage 2 and Stage 3 hypertension, respectively. Filled symbols indicate blood pressure measurements that exceed the threshold for Stage 1 hypertension. These time series graphs are vertically aligned with timelines that show medications over time. Other subviews show physical exam data, risk factors, related problems, and lab data relevant to hypertension.

    10. ATA 2007 Coordinated Cross-Patient and Single-Patient Views

    11. ATA 2007 Coordinated Cross-Patient and Single-Patient Views

    12. ATA 2007 DataMontage Editor The data retrieval, transformation, and presentation logic of each view is specified by drawing flow charts. Each rectangular nodes specifies calculations and actions. Each oval node specifies conditions that control which path within the flow chart is taken. A node can call a lower-level flow-chart, so complex view logic can be decomposed into multiple flow charts, organized in a hierarchy. Additional optional info – The green node indicates the flow chart’s “entry” node – execution of the flow chart starts at the green node. A red node indicates a “final” node. Execution of the flow chart ends at the red node. The “Catalog” pane on the left side of this window displays objects that comprise the view definitions, such as flow charts, functions that can be called by the flow charts, and global variables. These objects can be grouped within folders. A step-by-step debugger lets users execute views, step-by-step, to verify the correctness of each view.The data retrieval, transformation, and presentation logic of each view is specified by drawing flow charts. Each rectangular nodes specifies calculations and actions. Each oval node specifies conditions that control which path within the flow chart is taken. A node can call a lower-level flow-chart, so complex view logic can be decomposed into multiple flow charts, organized in a hierarchy. Additional optional info – The green node indicates the flow chart’s “entry” node – execution of the flow chart starts at the green node. A red node indicates a “final” node. Execution of the flow chart ends at the red node. The “Catalog” pane on the left side of this window displays objects that comprise the view definitions, such as flow charts, functions that can be called by the flow charts, and global variables. These objects can be grouped within folders. A step-by-step debugger lets users execute views, step-by-step, to verify the correctness of each view.

    13. ATA 2007 DataMontage Java API

    14. ATA 2007 DataMontage Summary

More Related