skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Computational Diagnostic: A Novel Approach to View Medical Data.

Abstract

A transition from traditional paper-based medical records to electronic health record is largely underway. The use of electronic records offers tremendous potential to personalize patient diagnosis and treatment. In this paper, we discuss a computational diagnostic tool that uses digital medical records to help doctors gain better insight about a patient's medical condition. The paper details different interactive features of the tool which offer potential to practice evidence-based medicine and advance patient diagnosis practices. The healthcare industry is a constantly evolving domain. Research from this domain is often translated into better understanding of different medical conditions. This new knowledge often contributes towards improved diagnosis and treatment solutions for patients. But the healthcare industry lags behind to seek immediate benefits of the new knowledge as it still adheres to the traditional paper-based approach to keep track of medical records. However recently we notice a drive that promotes a transition towards electronic health record (EHR). An EHR stores patient medical records in digital format and offers potential to replace the paper health records. Earlier attempts of an EHR replicated the paper layout on the screen, representation of medical history of a patient in a graphical time-series format, interactive visualization with 2D/3D generatedmore » images from an imaging device. But an EHR can be much more than just an 'electronic view' of the paper record or a collection of images from an imaging device. In this paper, we present an EHR called 'Computational Diagnostic Tool', that provides a novel computational approach to look at patient medical data. The developed EHR system is knowledge driven and acts as clinical decision support tool. The EHR tool provides two visual views of the medical data. Dynamic interaction with data is supported to help doctors practice evidence-based decisions and make judicious choices about patient treatment. Further, these two visual views work in unison as multiple coordinated views to provide additional insights about the medical data. In this paper, we will present inbuilt functionality of the computational diagnostic tool that can help doctors identify medical dataset characteristics such as: What patients show worst values for any given variables? What pattern is seen in medical variables of patients undergoing same treatment? From any given dataset, what patient population has severity values for different variables? Do a selected group of patients share similar trend across all different variables of interest? Can we compare two patient groups to see if they share similar trends across different variables? Is it possible to simultaneously compare trends and patterns for selected patients? The paper is organized as follows: Section 2 describes the medical dataset that is used with the computational diagnostic tool. Section 3 introduces the computational diagnostic tool. Further it discusses the two visualization views used to show medical data. The section also covers different data interaction support for both visualizations along with its usage for medical data analysis. Section 4 demonstrates the use of visual views as multiple coordinated views and how they complement each other to show different data characteristics. The paper concludes with discussion of results and future work.« less

Authors:
 [1];  [2]
  1. (Ketan Kirtiraj)
  2. (Katy)
Publication Date:
Research Org.:
Los Alamos National Laboratory
Sponsoring Org.:
USDOE
OSTI Identifier:
978012
Report Number(s):
LA-UR-07-1486
TRN: US201012%%605
Resource Type:
Conference
Resource Relation:
Conference: 5th International Conference on Coordinated&Multiple Views in Exploratory Visualization, July 2, 2007, ETH, Zurich, Switzerland
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; DATA ANALYSIS; DIAGNOSIS; MEDICAL RECORDS; DIAGNOSTIC TECHNIQUES; COMPUTER GRAPHICS

Citation Formats

Mane, K. K., and Börner, K.. Computational Diagnostic: A Novel Approach to View Medical Data.. United States: N. p., 2007. Web.
Mane, K. K., & Börner, K.. Computational Diagnostic: A Novel Approach to View Medical Data.. United States.
Mane, K. K., and Börner, K.. Mon . "Computational Diagnostic: A Novel Approach to View Medical Data.". United States. doi:. https://www.osti.gov/servlets/purl/978012.
@article{osti_978012,
title = {Computational Diagnostic: A Novel Approach to View Medical Data.},
author = {Mane, K. K. and Börner, K.},
abstractNote = {A transition from traditional paper-based medical records to electronic health record is largely underway. The use of electronic records offers tremendous potential to personalize patient diagnosis and treatment. In this paper, we discuss a computational diagnostic tool that uses digital medical records to help doctors gain better insight about a patient's medical condition. The paper details different interactive features of the tool which offer potential to practice evidence-based medicine and advance patient diagnosis practices. The healthcare industry is a constantly evolving domain. Research from this domain is often translated into better understanding of different medical conditions. This new knowledge often contributes towards improved diagnosis and treatment solutions for patients. But the healthcare industry lags behind to seek immediate benefits of the new knowledge as it still adheres to the traditional paper-based approach to keep track of medical records. However recently we notice a drive that promotes a transition towards electronic health record (EHR). An EHR stores patient medical records in digital format and offers potential to replace the paper health records. Earlier attempts of an EHR replicated the paper layout on the screen, representation of medical history of a patient in a graphical time-series format, interactive visualization with 2D/3D generated images from an imaging device. But an EHR can be much more than just an 'electronic view' of the paper record or a collection of images from an imaging device. In this paper, we present an EHR called 'Computational Diagnostic Tool', that provides a novel computational approach to look at patient medical data. The developed EHR system is knowledge driven and acts as clinical decision support tool. The EHR tool provides two visual views of the medical data. Dynamic interaction with data is supported to help doctors practice evidence-based decisions and make judicious choices about patient treatment. Further, these two visual views work in unison as multiple coordinated views to provide additional insights about the medical data. In this paper, we will present inbuilt functionality of the computational diagnostic tool that can help doctors identify medical dataset characteristics such as: What patients show worst values for any given variables? What pattern is seen in medical variables of patients undergoing same treatment? From any given dataset, what patient population has severity values for different variables? Do a selected group of patients share similar trend across all different variables of interest? Can we compare two patient groups to see if they share similar trends across different variables? Is it possible to simultaneously compare trends and patterns for selected patients? The paper is organized as follows: Section 2 describes the medical dataset that is used with the computational diagnostic tool. Section 3 introduces the computational diagnostic tool. Further it discusses the two visualization views used to show medical data. The section also covers different data interaction support for both visualizations along with its usage for medical data analysis. Section 4 demonstrates the use of visual views as multiple coordinated views and how they complement each other to show different data characteristics. The paper concludes with discussion of results and future work.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 01 00:00:00 EST 2007},
month = {Mon Jan 01 00:00:00 EST 2007}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • A test for the statistical significance of observed differences between two measured Receiver Operating Characteristic (ROC) curves has been designed and evaluated. The set of observer response data for each ROC curve is assumed to be independent and to arise from a ROC curve having a form which, in the absence of statistical fluctuations in the response data, graphs as a straight line on double normal-deviate axes. To test the significance of an apparent difference between two measured ROC curves, maximum likelihood estimates of the two parameters of each curve and the associated parameter variances and covariance are calculated frommore » the corresponding set of observer response data. An approximate Chi-square statistic with two degrees of freedom is then constructed from the differences between the parameters estimated for each ROC curve and from the variances and covariances of these estimates. This statistic is known to be truly Chi-square distributed only in the limit of large numbers of trials in the observer performance experiments. Performance of the statistic for data arising from a limited number of experimental trials was evaluated. Independent sets of rating scale data arising from the same underlying ROC curve were paired, and the fraction of differences found (falsely) significant was compared to the significance level, ..cap alpha.., used with the test. Although test performance was found to be somewhat dependent on both the number of trials in the data and the position of the underlying ROC curve in the ROC space, the results for various significance levels showed the test to be reliable under practical experimental conditions.« less
  • Matrix diffusion, advection and chemical reactions are the three most important mechanisms responsible for radionuclide migration in porous media. Last year the authors developed a two-dimensional deterministic heterogeneous matrix diffusion model, which is capable of utilizing porosity information originating from real drill-core samples. The results of numerical infiltration experiment they had performed with the model displayed substantial spatial variations in the penetration depth. Because it is practically impossible to verify experimentally those two-dimensional penetration profiles they had computed, this time they decided to try modelling of measured leaching curves. Unfortunately they have not succeeded in acquiring such curves for themore » exact same samples, which they have used in numerical leaching experiments. Nevertheless it can be seen, that the shape of leaching curves computed with the heterogeneous model is clearly closer to the shape of measured curves, than the shape of curves provided by the standard model. These differences can be utilized as a basis for an approximate numerical method of assessing the geometric factor, which has traditionally been a purely empirical parameter. The results of the new numerical experiments agree with the older results from last year: the heterogeneity of the rock matrix has highly significant impact on the diffusion. However, when interpreting the results, one must not neglect numerous limitations of the model, and hence, one should not attempt to overgeneralize the conclusions.« less