CHEST PAIN DIAGNOSIS ALGORITHM

Decision Support Program's for Diagnosis of Acute Chest Pain

This diagnosis program is based on a logistic regression model developed from a study on chest pain in Edinburgh Royal Infirmary during autumn 1993, and published in the European Heart Journal, August 1996 (abstract). It applies to patients presenting to the emergency room with non traumatic chest pain and uses the published logistic regression coefficients to calculate the estimated probability of Acute Myocardial Infarction. The model achieved an ROC curve area of 89.1% +- 2.18 on a separate Edinburgh test data set.

It is intended as a demonstration of how such models may be applied in clinical settings and must not be used for actual diagnosis in its present form.

The algorithm is implemented as a JAVA Applet included in the Web page. This Applet provides the check boxes to enter the data then calculate the probability using the standard logistic function:

Edinburgh 6 Input Model

See also:

The TIPI Model Developed in New England Medical Center