Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm

William Kostis, Cheng Yi, Evangelia Micheli-Tzanakou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We have applied neural network algorithms to estimate future vital status (dead or alive) of patients with acute myocardial infarction. A large database (Myocardial Infarction Data Acquisition System - MIDAS) including all 49,250 myocardial infarctions that occurred in the state of New Jersey in 1986 and 1987 with follow-up as long as five years was used in the development and testing of such a computer algorithm. Since the information included in the database was not sufficient to allow the exact prediction of vital status in all patients with 100% accuracy, we developed a neural network able to categorize patients according to the probability of dying within a given period of time rather than predicting categorically whether a given patient will be dead or alive at a given time in the future. A new algorithm was developed to accommodate cases where identical input vectors were associated with different outputs (vital status). In addition, a method of linear output adjustment was devised to describe the degree of confidence of each prediction. A two hidden layer perceptron, using ALOPEX, a feedback optimization algorithm, for weight updating, with 19 input patient variables of the MIDAS data set was used to predict six month mortality. The neural network was able to learn and was successful in predicting vital status at six months.

Original languageEnglish (US)
Title of host publication1993 IEEE 19th Annual Northeasrt Bioengineering Conference
PublisherPubl by IEEE
Pages99-100
Number of pages2
ISBN (Print)0780309251
StatePublished - Jan 1 1993
EventProceedings of the 1993 IEEE 19th Annual Northeast Bioengineering Conference - Newark, NJ, USA
Duration: Mar 18 1993Mar 19 1993

Publication series

NameBioengineering, Proceedings of the Northeast Conference

Other

OtherProceedings of the 1993 IEEE 19th Annual Northeast Bioengineering Conference
CityNewark, NJ, USA
Period3/18/933/19/93

Fingerprint

Neural networks
Data acquisition
Feedback
Testing

All Science Journal Classification (ASJC) codes

  • Bioengineering

Cite this

Kostis, W., Yi, C., & Micheli-Tzanakou, E. (1993). Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm. In 1993 IEEE 19th Annual Northeasrt Bioengineering Conference (pp. 99-100). (Bioengineering, Proceedings of the Northeast Conference). Publ by IEEE.
Kostis, William ; Yi, Cheng ; Micheli-Tzanakou, Evangelia. / Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm. 1993 IEEE 19th Annual Northeasrt Bioengineering Conference. Publ by IEEE, 1993. pp. 99-100 (Bioengineering, Proceedings of the Northeast Conference).
@inproceedings{23cf9663288e4773ad5d416fccf9e9e9,
title = "Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm",
abstract = "We have applied neural network algorithms to estimate future vital status (dead or alive) of patients with acute myocardial infarction. A large database (Myocardial Infarction Data Acquisition System - MIDAS) including all 49,250 myocardial infarctions that occurred in the state of New Jersey in 1986 and 1987 with follow-up as long as five years was used in the development and testing of such a computer algorithm. Since the information included in the database was not sufficient to allow the exact prediction of vital status in all patients with 100{\%} accuracy, we developed a neural network able to categorize patients according to the probability of dying within a given period of time rather than predicting categorically whether a given patient will be dead or alive at a given time in the future. A new algorithm was developed to accommodate cases where identical input vectors were associated with different outputs (vital status). In addition, a method of linear output adjustment was devised to describe the degree of confidence of each prediction. A two hidden layer perceptron, using ALOPEX, a feedback optimization algorithm, for weight updating, with 19 input patient variables of the MIDAS data set was used to predict six month mortality. The neural network was able to learn and was successful in predicting vital status at six months.",
author = "William Kostis and Cheng Yi and Evangelia Micheli-Tzanakou",
year = "1993",
month = "1",
day = "1",
language = "English (US)",
isbn = "0780309251",
series = "Bioengineering, Proceedings of the Northeast Conference",
publisher = "Publ by IEEE",
pages = "99--100",
booktitle = "1993 IEEE 19th Annual Northeasrt Bioengineering Conference",

}

Kostis, W, Yi, C & Micheli-Tzanakou, E 1993, Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm. in 1993 IEEE 19th Annual Northeasrt Bioengineering Conference. Bioengineering, Proceedings of the Northeast Conference, Publ by IEEE, pp. 99-100, Proceedings of the 1993 IEEE 19th Annual Northeast Bioengineering Conference, Newark, NJ, USA, 3/18/93.

Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm. / Kostis, William; Yi, Cheng; Micheli-Tzanakou, Evangelia.

1993 IEEE 19th Annual Northeasrt Bioengineering Conference. Publ by IEEE, 1993. p. 99-100 (Bioengineering, Proceedings of the Northeast Conference).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm

AU - Kostis, William

AU - Yi, Cheng

AU - Micheli-Tzanakou, Evangelia

PY - 1993/1/1

Y1 - 1993/1/1

N2 - We have applied neural network algorithms to estimate future vital status (dead or alive) of patients with acute myocardial infarction. A large database (Myocardial Infarction Data Acquisition System - MIDAS) including all 49,250 myocardial infarctions that occurred in the state of New Jersey in 1986 and 1987 with follow-up as long as five years was used in the development and testing of such a computer algorithm. Since the information included in the database was not sufficient to allow the exact prediction of vital status in all patients with 100% accuracy, we developed a neural network able to categorize patients according to the probability of dying within a given period of time rather than predicting categorically whether a given patient will be dead or alive at a given time in the future. A new algorithm was developed to accommodate cases where identical input vectors were associated with different outputs (vital status). In addition, a method of linear output adjustment was devised to describe the degree of confidence of each prediction. A two hidden layer perceptron, using ALOPEX, a feedback optimization algorithm, for weight updating, with 19 input patient variables of the MIDAS data set was used to predict six month mortality. The neural network was able to learn and was successful in predicting vital status at six months.

AB - We have applied neural network algorithms to estimate future vital status (dead or alive) of patients with acute myocardial infarction. A large database (Myocardial Infarction Data Acquisition System - MIDAS) including all 49,250 myocardial infarctions that occurred in the state of New Jersey in 1986 and 1987 with follow-up as long as five years was used in the development and testing of such a computer algorithm. Since the information included in the database was not sufficient to allow the exact prediction of vital status in all patients with 100% accuracy, we developed a neural network able to categorize patients according to the probability of dying within a given period of time rather than predicting categorically whether a given patient will be dead or alive at a given time in the future. A new algorithm was developed to accommodate cases where identical input vectors were associated with different outputs (vital status). In addition, a method of linear output adjustment was devised to describe the degree of confidence of each prediction. A two hidden layer perceptron, using ALOPEX, a feedback optimization algorithm, for weight updating, with 19 input patient variables of the MIDAS data set was used to predict six month mortality. The neural network was able to learn and was successful in predicting vital status at six months.

UR - http://www.scopus.com/inward/record.url?scp=0027226432&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027226432&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0027226432

SN - 0780309251

T3 - Bioengineering, Proceedings of the Northeast Conference

SP - 99

EP - 100

BT - 1993 IEEE 19th Annual Northeasrt Bioengineering Conference

PB - Publ by IEEE

ER -

Kostis W, Yi C, Micheli-Tzanakou E. Estimation of long-term vital status of patients after myocardial infarction using a neural network based on the ALOPEX algorithm. In 1993 IEEE 19th Annual Northeasrt Bioengineering Conference. Publ by IEEE. 1993. p. 99-100. (Bioengineering, Proceedings of the Northeast Conference).