ANALYSIS OF THE CLINICAL-VARIABLES DRIVING DECISION IN AN ARTIFICIAL NEURAL NETWORK TRAINED TO IDENTIFY THE PRESENCE OF MYOCARDIAL-INFARCTION

被引：55

作者：

BAXT, WG

机构：

[1] Departments of Emergency Medicine and Medicine, University of California, San Diego Medical Center

来源：

ANNALS OF EMERGENCY MEDICINE | 1992年 / 21卷 / 12期

关键词：

D O I：

10.1016/S0196-0644(05)80056-3

中图分类号：

R4 [临床医学];

学科分类号：

1002 ; 100602 ;

摘要：

Study objective: To determine which clinical variables drive the output of an artificial neural network trained to identify the presence of myocardial infarction. Design: Partial output analysis. Setting: Tertiary university teaching center. Participants: Seven hundred six patients more than 1 8 years old presenting with anterior chest pain. Measurements: Differential network output analysis. Main results: A methodology was developed as the first step in measuring the impact input clinical variables have on the output (diagnosis) of an artificial neural network trained to identify the presence of acute myocardial infarction. The methodology revealed that the network used the presence of ECG findings, as well as the presence of rales, syncope, jugular venous distension, response to trinitroglycerin, and nausea and vomiting, as major predictive sources. Although this first-step analysis studied individual variables, it must be stated that the network comes to clinical closure based on the settings of all variables in a pattern and that the impact of a single variable cannot be taken out of the context of a pattern. Conclusion: An artificial neural network trained to recognize the presence of myocardial infarction appears to place diagnostic importance on clinical variables that have not been shown previously to be highly predictive for infarction.

引用

页码：1439 / 1444

页数：6

共 21 条

[1] Rumelhart, Hinton, Williams, Learning internal representations by error propagation, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, pp. 318-364, (1986)
[2] McClelland, Rumelhart, Training hidden units, Explorations in Parallel Distributed Processing, pp. 121-160, (1988)
[3] Weigend, Huberman, Rumelhart, Predicting the Future: A Connectionist Approach, Stanford PDP Research Group, (1990)
[4] Hudson, Cohen, Anderson, Determination of testing efficacy in carcinoma of the lung using a neural network model, Symposium on Computer Applications in Medical Care 1988 Proceedings: 12th Annual Symposium, Washington, DC, 12, pp. 251-255, (1988)
[5] Smith, Everhart, Dickson, Et al., Using the ADAP learning algorithm to forecast the onset of diabetes mellitus, Symposium on Computer Applications in Medical Care 1988 Proceedings: 12th Annual Symposium, Washington, DC, 12, pp. 261-265, (1988)
[6] Saito, Nakano, Medical diagnostic expert system based on PDP model, Proceedings of the International Joint Conference on Neural Networks, San Diego, 2, pp. 255-262, (1988)
[7] Kaufman, Chiabera, Hatem, Et al., A neural network approach for bone fracture healing assessment, IEEE Eng Med Biol, 9, pp. 23-30, (1990)
[8] Hiraiwa, Shimohara, Tokunaga, EEG topography recognition by neural networks, IEEE Eng Med Biol, 9, pp. 39-42, (1990)
[9] Cios, Chen, Langenderfer, Use of neural networks in detecting cardiac diseases from echocardiographic images, IEEE Eng Med Biol, 9, pp. 58-60, (1990)
[10] Marconi, Scalia, Ridella, Et al., An application of back propagation to medical diagnosis, Proceedings of the International Joint Conference on Neural Networks, Washington, DC, 2, (1989)

← 1 2 3 →