Isaac S. Kohane, Ira J. Haimowitz, Hypothesis-Driven Data Abstraction with Trend Templates, Symposium on Computer Applications in Medical Care 1993. Home/Search Document Details and Download
Summary Related Articles |
This paper is cited in the following contexts:
Deriving Trends in Historical and Real-Time Continuously.. - Salatian, Hunter (1998) (1 citation) (Correct)
....In many domains we are confronted with large sets of continuously sampled data. Reasoning about the relationships between the consecutive individual measurements of one variable is computationally expensive and gets worse when several variables are interpreted together. We agree with (Kohane and Haimowitz, 1993) when they say that: The abstraction of primary data into intervals over which a specified predicate holds is a central task in process monitoring. It relieves the monitoring program of the complexity of having to repeatedly reason about the relationships between each datum in potentially vast ....
....domain specific knowledge. Intervals which are increasing, decreasing or steady are derived using quantity space conversion tables which are based on the slope of change. A series of data points can be classified into a temporal interval by using a set of constraints. For example Kohane s TrenDx (Kohane and Haimowitz, 1993) is a system that performs abstraction as it hypothesises medical process disorders represented by trend templates (an archetypal pattern of data variation in a process disorder) Within each trend template, the component intervals constitute candidate abstraction intervals for data which match ....
Isaac S. Kohane, Ira J. Haimowitz, Hypothesis-Driven Data Abstraction with Trend Templates, Symposium on Computer Applications in Medical Care 1993.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC