Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study

• Published in: PloS one 2024-04, Vol.19 (4), p.e0300796-e0300796
• Main Authors: Del Gaizo, John, Sherard, Curry, Shorbaji, Khaled, Welch, Brett, Mathi, Roshan, Kilic, Arman
• Format: Article
• Language: English
• Subjects: Aged; Artificial Intelligence; Biology and Life Sciences; Computational linguistics; Computer and Information Sciences; Coronary artery bypass; Coronary Artery Bypass - methods; Female; Health aspects; Humans; Language processing; Male; Medical research; Medicine and Health Sciences; Medicine, Experimental; Methods; Middle Aged; Mortality; Natural language interfaces; Natural Language Processing; Patient outcomes; People and Places; Physical Sciences; Practice; Surgeons; Treatment Outcome
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0300796

Healthcare providers currently calculate risk of the composite outcome of morbidity or mortality associated with a coronary artery bypass grafting (CABG) surgery through manual input of variables into a logistic regression-based risk calculator. This study indicates that automated artificial intelligence (AI)-based techniques can instead calculate risk. Specifically, we present novel numerical embedding techniques that enable NLP (natural language processing) models to achieve higher performance than the risk calculator using a single preoperative surgical note. The most recent preoperative surgical consult notes of 1,738 patients who received an isolated CABG from July 1, 2014 to November 1, 2022 at a single institution were analyzed. The primary outcome was the Society of Thoracic Surgeons defined composite outcome of morbidity or mortality (MM). We tested three numerical-embedding techniques on the widely used TextCNN classification model: 1a) Basic embedding, treat numbers as word tokens; 1b) Basic embedding with a dataloader that Replaces out-of-context (ROOC) numbers with a tag, where context is defined as within a number of tokens of specified keywords; 2) ScaleNum, an embedding technique that scales in-context numbers via a learned sigmoid-linear-log function; and 3) AttnToNum, a ScaleNum-derivative that updates the ScaleNum embeddings via multi-headed attention applied to local context. Predictive performance was measured via area under the receiver operating characteristic curve (AUC) on holdout sets from 10 random-split experiments. For eXplainable-AI (X-AI), we calculate SHapley Additive exPlanation (SHAP) values at an ngram resolution (SHAP-N). While the analyses focus on TextCNN, we execute an analogous performance pipeline with a long short-term memory (LSTM) model to test if the numerical embedding advantage is robust to model architecture. A total of 567 (32.6%) patients had MM following CABG. The embedding performances are as follows with the TextCNN architecture: 1a) Basic, mean AUC 0.788 [95% CI (confidence interval): 0.768-0.809]; 1b) ROOC, 0.801 [CI: 0.788-0.815]; 2) ScaleNum, 0.808 [CI: 0.785-0.821]; and 3) AttnToNum, 0.821 [CI: 0.806-0.834]. The LSTM architecture produced a similar trend. Permutation tests indicate that AttnToNum outperforms the other embedding techniques, though not statistically significant verse ScaleNum (p-value of .07). SHAP-N analyses indicate that the model learns to associate low blood urine nitrate (BUN) an