{"id":{"repo_id":"salford","oai_identifier":"oai:salford-repository.worktribe.com:1337442"},"canonical_url":"https://search.dev.ndltd.org/etd/salford/oai:salford-repository.worktribe.com:1337442","repository":{"repo_id":"salford","name":"U. of Salford","base_url":"https://salford-repository.worktribe.com/oaiprovider"},"display":{"title":"Modelling outcome prediction for trauma patients : an artificial intelligence approach","abstract":"Trauma, a term used in medicine to describe a physical injury, is believed to be oneof the major causes of death and disability in modern societies. The development ofthe Trauma and Injury Severity Score (TRISS) method by Boyd et al. (1987) can beconsidered as a high-impact initiative in order to improve the trauma patient's care.This method was used to compare the expected and the observed outcomes inrelation to mortality. Thus, the rate of unexpected deaths or survivals can beexamined and any related problems such as improper trauma patient's care can beidentified. In general, the model with this particular task can be referred to as aprediction or classification model. In our study, the Trauma Audit and ResearchNetwork (TARN) has applied the TRISS method to assist them in a comparativeaudit among the participating hospitals since 1989. Despite the fact that the TRISSmodel is simple and easy to use, there is some limitation in the logistic regressiontechnique which the TRISS model is based upon. In fact, some preliminary resultsfrom other researchers have indicated that prediction accuracy may be improved byusing alternative modelling approaches, such as the artificial intelligence (AI) basedmethods. Therefore, attempts are made in this study with the aim of developing newoutcome prediction models using the AI methods namely; artificial neural networks,support vector machines, A>nearest neighbour and naive Bayesian, and then theresults will be compared to the TRISS-FP model (for comparison purposes, we referto the outcome prediction model based on the TRISS method developed by TARN asthe TRISS-FP model throughout this thesis). The model's predictive performancesare evaluated using performance measures, including sensitivity, specificity, areaunder the receiving operating characteristic curve (AUC) and geometric mean (Gmean).The data for this research is drawn from the TARN database. The empiricalresult has shown that the new Al-based models developed in this study obtainedbetter predictive performances compared to the TRISS-FP model. The ANN modelhas emerged as the best Al-based model. Thus, this model will be recommended tothe TARN for future consideration.","abstract_html":"Trauma, a term used in medicine to describe a physical injury, is believed to be oneof the major causes of death and disability in modern societies. The development ofthe Trauma and Injury Severity Score (TRISS) method by Boyd et al. (1987) can beconsidered as a high-impact initiative in order to improve the trauma patient's care.This method was used to compare the expected and the observed outcomes inrelation to mortality. Thus, the rate of unexpected deaths or survivals can beexamined and any related problems such as improper trauma patient's care can beidentified. In general, the model with this particular task can be referred to as aprediction or classification model. In our study, the Trauma Audit and ResearchNetwork (TARN) has applied the TRISS method to assist them in a comparativeaudit among the participating hospitals since 1989. Despite the fact that the TRISSmodel is simple and easy to use, there is some limitation in the logistic regressiontechnique which the TRISS model is based upon. In fact, some preliminary resultsfrom other researchers have indicated that prediction accuracy may be improved byusing alternative modelling approaches, such as the artificial intelligence (AI) basedmethods. Therefore, attempts are made in this study with the aim of developing newoutcome prediction models using the AI methods namely; artificial neural networks,support vector machines, A>nearest neighbour and naive Bayesian, and then theresults will be compared to the TRISS-FP model (for comparison purposes, we referto the outcome prediction model based on the TRISS method developed by TARN asthe TRISS-FP model throughout this thesis). The model's predictive performancesare evaluated using performance measures, including sensitivity, specificity, areaunder the receiving operating characteristic curve (AUC) and geometric mean (Gmean).The data for this research is drawn from the TARN database. The empiricalresult has shown that the new Al-based models developed in this study obtainedbetter predictive performances compared to the TRISS-FP model. The ANN modelhas emerged as the best Al-based model. Thus, this model will be recommended tothe TARN for future consideration.","abstract_has_math":false,"creators":["Ali, NA"],"institution":null,"degree_name":null,"degree_level":"Doctoral (Level 8)","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011","date_published":"2011","updated_at":"2026-07-24T04:26:09Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:salford-repository.worktribe.com:1337442"],"render_values":[{"text":"oai:salford-repository.worktribe.com:1337442","href":null,"code":true}]}]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.sponsor","label":"Sponsor","values":["#1 FUNDER NOT LISTED"]},{"key":"dc:creator","label":"Author","values":["Ali, NA"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2011-04-01"]},{"key":"dc:date.issued","label":"Date","values":["2011"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://salford-repository.worktribe.com/output/1337442"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral (Level 8)"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:salford-repository.worktribe.com:1337442"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://salford-repository.worktribe.com/file/1337442/1/11387831.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Trauma, a term used in medicine to describe a physical injury, is believed to be oneof the major causes of death and disability in modern societies. The development ofthe Trauma and Injury Severity Score (TRISS) method by Boyd et al. (1987) can beconsidered as a high-impact initiative in order to improve the trauma patient's care.This method was used to compare the expected and the observed outcomes inrelation to mortality. Thus, the rate of unexpected deaths or survivals can beexamined and any related problems such as improper trauma patient's care can beidentified. In general, the model with this particular task can be referred to as aprediction or classification model. In our study, the Trauma Audit and ResearchNetwork (TARN) has applied the TRISS method to assist them in a comparativeaudit among the participating hospitals since 1989. Despite the fact that the TRISSmodel is simple and easy to use, there is some limitation in the logistic regressiontechnique which the TRISS model is based upon. In fact, some preliminary resultsfrom other researchers have indicated that prediction accuracy may be improved byusing alternative modelling approaches, such as the artificial intelligence (AI) basedmethods. Therefore, attempts are made in this study with the aim of developing newoutcome prediction models using the AI methods namely; artificial neural networks,support vector machines, A>nearest neighbour and naive Bayesian, and then theresults will be compared to the TRISS-FP model (for comparison purposes, we referto the outcome prediction model based on the TRISS method developed by TARN asthe TRISS-FP model throughout this thesis). The model's predictive performancesare evaluated using performance measures, including sensitivity, specificity, areaunder the receiving operating characteristic curve (AUC) and geometric mean (Gmean).The data for this research is drawn from the TARN database. The empiricalresult has shown that the new Al-based models developed in this study obtainedbetter predictive performances compared to the TRISS-FP model. The ANN modelhas emerged as the best Al-based model. Thus, this model will be recommended tothe TARN for future consideration."]},{"key":"dc:title","label":"Title","values":["Modelling outcome prediction for trauma patients : an artificial intelligence approach"]}]}],"canonical_facts":{"dc:contributor.sponsor":["#1 FUNDER NOT LISTED"],"dc:creator":["Ali, NA"],"dc:date":["2011-04-01"],"dc:date.issued":["2011"],"dc:description.abstract":["Trauma, a term used in medicine to describe a physical injury, is believed to be oneof the major causes of death and disability in modern societies. The development ofthe Trauma and Injury Severity Score (TRISS) method by Boyd et al. (1987) can beconsidered as a high-impact initiative in order to improve the trauma patient's care.This method was used to compare the expected and the observed outcomes inrelation to mortality. Thus, the rate of unexpected deaths or survivals can beexamined and any related problems such as improper trauma patient's care can beidentified. In general, the model with this particular task can be referred to as aprediction or classification model. In our study, the Trauma Audit and ResearchNetwork (TARN) has applied the TRISS method to assist them in a comparativeaudit among the participating hospitals since 1989. Despite the fact that the TRISSmodel is simple and easy to use, there is some limitation in the logistic regressiontechnique which the TRISS model is based upon. In fact, some preliminary resultsfrom other researchers have indicated that prediction accuracy may be improved byusing alternative modelling approaches, such as the artificial intelligence (AI) basedmethods. Therefore, attempts are made in this study with the aim of developing newoutcome prediction models using the AI methods namely; artificial neural networks,support vector machines, A>nearest neighbour and naive Bayesian, and then theresults will be compared to the TRISS-FP model (for comparison purposes, we referto the outcome prediction model based on the TRISS method developed by TARN asthe TRISS-FP model throughout this thesis). The model's predictive performancesare evaluated using performance measures, including sensitivity, specificity, areaunder the receiving operating characteristic curve (AUC) and geometric mean (Gmean).The data for this research is drawn from the TARN database. The empiricalresult has shown that the new Al-based models developed in this study obtainedbetter predictive performances compared to the TRISS-FP model. The ANN modelhas emerged as the best Al-based model. Thus, this model will be recommended tothe TARN for future consideration."],"dc:identifier":["oai:salford-repository.worktribe.com:1337442"],"dc:identifier.uri":["https://salford-repository.worktribe.com/file/1337442/1/11387831.pdf"],"dc:language":["en"],"dc:relation.isreferencedby":["https://salford-repository.worktribe.com/output/1337442"],"dc:title":["Modelling outcome prediction for trauma patients : an artificial intelligence approach"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral (Level 8)"]},"updated_at":"2026-07-24T04:26:09Z"}