以人工智慧識別腹主動脈瘤手術危險因子及死亡率預測

Risk factor identification and mortality prediction in abdominal aortic surgery using artificial intelligence

詹千慧(Chien-Hui Chan)

指導教授 ： 葛煥昭

淡江大學/工學院/資訊工程學系碩士班/碩士(2010年)

https://doi.org/10.6846/TKU.2010.00193

摘要

本研究提出一集成式腹主動脈瘤手術後併發症預測模型，本模型以1994年至2008年間進行腹主動脈瘤手術之病患資料進行訓練，本研究結果包括一集成式術後併發症預測模型、術後併發症預測記錄及因果關係決策規則，本模型所計算出之併發症機率與實際發生併發症事實比較，並以接收操作特徵曲線(ROC curve) 進行術後併發症預測模型之準確性評估。經過一系列測試，貝式網路(BN)、類神經網路(NN)及支持向量機(SVM)所集成之模型對於腹主動脈瘤修復術術後併發症預測可提供良好的效能。此外，貝式網路之馬可夫覆蓋提供了以粒子計算所產生的基本決策規則而自然形成之因果關係特徵選取。

關鍵字

主動脈瘤修復 ； 手術後併發症 ； 集成式模型 ； 機器學習 ； 馬可夫覆蓋

並列摘要

This study proposes an ensemble model to predict postoperative morbidity after abdominal aortic surgery. The ensemble model was developed using a training set of consecutive patients who underwent abdominal aortic aneurysm (AAA) repair between 1994 and 2008. The research outcomes consisted of an ensemble model to predict postoperative morbidity, the occurrence of postoperative complications prospectively recorded, and the causal-effect decision rules. The probabilities of complication calculated by the model were compared to the actual occurrence of complications and a receiver operating characteristic (ROC) curve was used to evaluate the accuracy of postoperative morbidity prediction. In this series, the ensemble of BN, NN and SVM models offered satisfactory performance in predicting postoperative morbidity after AAA repair. Moreover, the Markov blankets of BN allow a natural form of causal-effect feature selection, which provides a basis for screening decision rules generated by granular computing.

並列關鍵字

Aortic aneurysm repair ； postoperative morbidity ； ensemble model ； machine learning ； Markov blanket

參考文獻

1. Roche-Nagle, G., Lindsay, T.F.: Endovascular ruptured abdominal aortic aneurysm repair - setting up your hospital for an endovascular approach. The surgeon : journal of the Royal Colleges of Surgeons of Edinburgh and Ireland 8 (2010) 39-43

2. Barnes, M., Boult, M., Maddern, G., Fitridge, R.: A model to predict outcomes for Endovascular aneurysm repair using preoperative variables. European Journal of Vascular and Endovascular Surgery 35 (2008) 571-579

3. Bohm, N., Wales, L., Dunckley, M., Morgan, R., Loftus, I., Thompson, M.: Objective risk-scoring systems for repair of abdominal aortic aneurysms: applicability in Endovascular repair. European Journal of Vascular and Endovascular Surgery 36 (2008) 172-177

4. Stijn, C.W., Wouters, L.N., Freek, W.A.V., Rene, M.H.J.B.: Preoperative prediction of early mortality and morbidity in coronary bypass surgery. Cardiovascular Surgery 10 (2002) 500-505

6. Bush, R.L., Johnson, M.L., Hedayati, N., Henderson, W.G., Lin, P.H., Lumsden, A.B.: Performance of endovascular aortic aneurysm repair in high-risk patients: Results from the Veterans Affairs National Surgical Quality Improvement Program. Journal of Vascular Surgery 45 (2007) 227-235