Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 1, 1997Medical Decision Making

The neural network demonstrated good discrimination (ROC area 0.81 in validation set) and calibration (Hosmer-Lemeshow p=0.21) for estimating mortality risk.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Can a probabilistic neural network accurately estimate the risk of mortality after cardiac surgery?

Population

1,477 consecutive cardiac surgery patients operated on in a teaching hospital during a four-year period

Design

Cohort, Patient records randomly divided into training and validation sets

Authors

RORichard K. Orr

Discussion

Loading...

Member takes

Overview

Should not yet change practice for cardiac surgery risk estimation; leaves open prospective validation of neural networks.

Structured PICO

Can a probabilistic neural network accurately estimate the risk of mortality after cardiac surgery?

P
Population
1,477 consecutive cardiac surgery patients operated on in a teaching hospital during a four-year period (1991-1994)
I
Intervention
Probabilistic neural network (PNN) model using seven routine clinical variables
O
Outcome
Accuracy, calibration, and discrimination of mortality risk estimation

A probabilistic neural network using seven routine clinical variables can accurately estimate mortality risk following cardiac surgery.

Cite This Study

Richard K. Orr (1997) studied this question.

synapsesocial.com/papers/6a19d59b60e90a7f5feab819https://doi.org/10.1177/0272989x9701700208
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A Comparison of Four Severity-Adjusted Models to Predict Mortality After Coronary Artery Bypass Graft Surgery1995 · 56 citations
  2. 2Prospective validation of artificial neural network trained to identify acute myocardial infarction1996 · 217 citations
  3. 3Assessment of predictive models for binary outcomes: An empirical approach using operative death from cardiac surgery1994 · 50 citations
  4. 4Stratification of Morbidity and Mortality Outcome by Preoperative Risk Factors in Coronary Artery Bypass Patients1992 · 806 citations
  5. 5Artificial neural networks in mammography: application to decision making in the diagnosis of breast cancer.1993 · 425 citations