Using neural networks and just nine patient-reportable factors of screen for AMI
The study investigated the effect of different input selections on the performance of artificial neural networks in screening for acute myocardial infarction (AMI) in Malaysian patients complaining of chest pain. We used hospital data to create neural networks with four input selections and used the...
| প্রধান লেখক: | , |
|---|---|
| বিন্যাস: | প্রবন্ধ |
| প্রকাশিত: |
2006
|
| বিষয়গুলি: |
| _version_ | 1825718756121247744 |
|---|---|
| author | Bulgiba, A.M. Fisher, M.H. |
| author_facet | Bulgiba, A.M. Fisher, M.H. |
| author_sort | Bulgiba, A.M. |
| collection | UM |
| description | The study investigated the effect of different input selections on the performance of artificial neural networks in screening for acute myocardial infarction (AMI) in Malaysian patients complaining of chest pain. We used hospital data to create neural networks with four input selections and used these to diagnose AMI. A 10-fold cross-validation and committee approach was used. All the neural networks using various input selections outperformed a multiple logistic regression model, although the difference was not statistically significant. The neural networks achieved an area under the ROC curve of 0.792 using nine inputs, whereas multiple logistic regression achieved 0.739 using 64 inputs. Sensitivity levels of over 90 per cent were achieved using low output threshold levels. Specificity levels of over 90 per cent were achieved using threshold levels of 0.4-0.5. Thus neural networks can perform as well as multiple logistic regression models even when using far fewer inputs. |
| first_indexed | 2024-03-06T05:09:13Z |
| format | Article |
| id | um.eprints-3083 |
| institution | Universiti Malaya |
| last_indexed | 2024-03-06T05:09:13Z |
| publishDate | 2006 |
| record_format | dspace |
| spelling | um.eprints-30832012-05-03T02:56:24Z http://eprints.um.edu.my/3083/ Using neural networks and just nine patient-reportable factors of screen for AMI Bulgiba, A.M. Fisher, M.H. R Medicine The study investigated the effect of different input selections on the performance of artificial neural networks in screening for acute myocardial infarction (AMI) in Malaysian patients complaining of chest pain. We used hospital data to create neural networks with four input selections and used these to diagnose AMI. A 10-fold cross-validation and committee approach was used. All the neural networks using various input selections outperformed a multiple logistic regression model, although the difference was not statistically significant. The neural networks achieved an area under the ROC curve of 0.792 using nine inputs, whereas multiple logistic regression achieved 0.739 using 64 inputs. Sensitivity levels of over 90 per cent were achieved using low output threshold levels. Specificity levels of over 90 per cent were achieved using threshold levels of 0.4-0.5. Thus neural networks can perform as well as multiple logistic regression models even when using far fewer inputs. 2006 Article PeerReviewed Bulgiba, A.M. and Fisher, M.H. (2006) Using neural networks and just nine patient-reportable factors of screen for AMI. Health Informatics Journal, 12 (3). pp. 213-225. ISSN 1460-4582, DOI https://doi.org/10.1177/1460458206066665 <https://doi.org/10.1177/1460458206066665 >. http://www.ncbi.nlm.nih.gov/pubmed/17023409 10.1177/1460458206066665 |
| spellingShingle | R Medicine Bulgiba, A.M. Fisher, M.H. Using neural networks and just nine patient-reportable factors of screen for AMI |
| title | Using neural networks and just nine patient-reportable factors of screen for AMI |
| title_full | Using neural networks and just nine patient-reportable factors of screen for AMI |
| title_fullStr | Using neural networks and just nine patient-reportable factors of screen for AMI |
| title_full_unstemmed | Using neural networks and just nine patient-reportable factors of screen for AMI |
| title_short | Using neural networks and just nine patient-reportable factors of screen for AMI |
| title_sort | using neural networks and just nine patient reportable factors of screen for ami |
| topic | R Medicine |
| work_keys_str_mv | AT bulgibaam usingneuralnetworksandjustninepatientreportablefactorsofscreenforami AT fishermh usingneuralnetworksandjustninepatientreportablefactorsofscreenforami |