What question did this study set out to answer?

To quantitatively determine which clinical signs, symptoms, and laboratory data serve as optimal descriptors and discriminators in the differential diagnosis of chest pain.

April 1, 1968Methods of Information in Medicine

Computer Evaluation of Statistical Properties of Clinical Information in the Differential Diagnosis of Chest Pain

View Full Paper

Ask AI

Bookmark

To quantitatively determine which clinical signs, symptoms, and laboratory data serve as optimal descriptors and discriminators in the differential diagnosis of chest pain.
Analyzed 498 clinical items (429 binary questions and 69 numerical parameters) collected from 1,238 patients presenting with chest pain.
Assessed discriminative power using contingency table analysis and chi-square tests, followed by discriminant function analysis to reduce the feature set.
Only 46 of 498 clinical items achieved a chi-square value of 40 or higher, showing that many high-incidence signs and symptoms offered minimal diagnostic discrimination.
Discriminant function analysis reduced the predictor set to fewer than 10 variables, which correctly classified more than 95% of 1,000 patients with coronary artery disease and pneumonia.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Pilot study demonstrates that under ten clinical variables accurately classify over 95 percent of chest pain cases, indicating the efficiency of statistical data reduction in computer diagnosis.

Cite This Study

Klingemanand et al. (1968) studied this question.

synapsesocial.com/papers/6a9889dc8b5843205fcb9c55 https://doi.org/10.1055/s-0038-1636347

View Full Paper

Ask AI

Bookmark

Implication

Pilot study demonstrates that under ten clinical variables accurately classify over 95 percent of chest pain cases, indicating the efficiency of statistical data reduction in computer diagnosis.

Computer Evaluation of Statistical Properties of Clinical Information in the Differential Diagnosis of Chest Pain

Key Points

Discussion

Implication

Cite This Study

Computer Evaluation of Statistical Properties of Clinical Information in the Differential Diagnosis of Chest Pain

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider