Machine Learning Outperforms Existing Clinical Scoring Tools in the Prediction of Postoperative Atrial Fibrillation During Intensive Care Unit Admission After Cardiac Surgery

Karri, Roshan; Kawai, Andrew; Thong, Yoke Jia; Ramson, Dhruvesh M.; Perry, Luke A; Segal, Reny; Smith, Julian A.; Penny-Dimri, Jahan C.

doi:10.1016/j.hlc.2021.05.101

Cited by 20 publications

(29 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Seven studies were unable to identify the confounding factors (5,8–13). Only three studies did not clearly state if exposures were measured in a valid and reliable way (4–6). Finally, all studies used proper statistical analysis, and measured their outcomes in a valid and reliable way.…”

Section: Resultsmentioning

confidence: 99%

“…AI applications also affect clinician outcomes, specifically, clinician decision making, clinician workflow and efficiency, and clinician evaluations and acceptance of AI applications. In this review, thirty-two studies reported clinician outcomes of AI in cardiac surgery (4,5,7,8,10,11,13–38).…”

Section: Resultsmentioning

confidence: 99%

“…Clinicians could potentially be guided by AI applications in making better medical decisions. Eighteen studies reported that AI applications can support clinician decision making (4,7,10,13,16,17,19,21,23,24,26,28,(30)(31)(32)(33)35,38). Machine learning models improve clinician's medical decisions by providing better preoperative risk assessment, stratification and prognostication (10,17,21,24,(30)(31)(32)35,38).…”

Section: B Clinician Outcomesmentioning

confidence: 99%

See 2 more Smart Citations

Artificial Intelligence in Cardiac Surgery: A Systematic Review

Sulague,

Beloy,

Medina

et al. 2023

Preprint

View full text Add to dashboard Cite

BACKGROUND Artificial intelligence has emerged as a tool to potentially increase efficiency and efficacy of healthcare and improve clinical outcomes. The growing body of knowledge of artificial intelligence applications in cardiac surgery necessitates evaluation of past studies to gain insights to the future direction of artificial intelligence applications in cardiac surgery. This study aims to provide a systematic review of the applications of artificial intelligence in cardiac surgery. METHODS A systematic literature search on artificial intelligence applications in cardiac surgery from 2000 to 2022 was conducted in the following databases: PubMed, Embase, Europe PMC, Epistemonikos, CINAHL, Cochrane Central, Google Scholar, Web of Science, Scopus, Cambridge Core, clinicaltrials.gov, and science. Studies on the implementation of artificial intelligence applications in cardiac surgery and the provision of decision support by the application through simulating clinical decision-making processes of healthcare providers were included. Studies not in English, published only as abstracts, review papers, meta-analyses, clinical trials that were still in progress, and published study protocols were excluded. This study was registered on Prospero (CRD42022377530). RESULTS A total of 42 studies were found that reported on artificial intelligence applications in cardiac surgery, all of which are cohort studies. Nine (21.43%) of the studies measured different parameters regarding cardiac surgeries in general. Meanwhile, 6 (14.29%) studies focused on Heart Transplantation (HT), 4 (9.52%) on Transcatheter Aortic Valve Replacement (TAVR), 3 (7.14%) anchored on Aortic Stenosis, and another 3 (7.14%) on Perioperative Complications. Three topics had 2 (4.76%) studies dedicated to them, namely Coronary Artery Bypass Graft (CABG), Postoperative Atrial Fibrillation (POAF), and Acute Kidney Injury (AKI). The remaining eleven studies have their own unique disease topics, procedures or surgeries in focus (n=11, 1 (2.38%), namely Postoperative Major Bleeding, Early Coronary Revascularization, Heart Valve Surgery, Isolated Mitral Valve Replacement (IMVR), Surgical Aortic Valve Replacement (SAVR), Open-Chest Surgery, Infective endocarditis, Post-Operative Deterioration, Red Blood Cell Transfusion, AKI - related Hippocampal Damage, and Open-Heart Surgery. Regarding evaluation outcomes, 26 studies examined the performance, 32 studies examined clinician outcomes, and 2 studies examined patient outcomes. Of the 42 studies, only 13 were conducted in Lower- and Middle-Income Countries. CONCLUSION Artificial intelligence was used to predict mortality, postoperative length of stay, and complications following cardiac surgeries. It can also improve clinicians’ medical decisions by providing better preoperative risk assessment, stratification, and prognostication. While the application of artificial intelligence in cardiac surgery has greatly progressed in the last two decades, more highly powered studies need to be done to assess challenges and to ensure accuracy and safety for use in clinical practice.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: B Clinician Outcomesmentioning

confidence: 99%

See 1 more Smart Citation

Artificial Intelligence in Cardiac Surgery: A Systematic Review

Sulague,

Beloy,

Medina

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently developed explanatory modeling of previously uninterpretable ML models, such as with Shapley additive values, have made it possible to leverage the increased complexity of ML to gain unique insights from the data 17 . Several of the included studies included ML methods to provide unique prediction level risk profiles for patients 18,19 . The ability of ML methods to provide unique explanations for predictions of risk is not a measurable outcome despite being an emerging benefit of these models.…”

Section: Discussionmentioning

confidence: 99%

“… 17 Several of the included studies included ML methods to provide unique prediction level risk profiles for patients. 18 , 19 The ability of ML methods to provide unique explanations for predictions of risk is not a measurable outcome despite being an emerging benefit of these models.…”

Section: Discussionmentioning

confidence: 99%

Machine learning to predict adverse outcomes after cardiac surgery: A systematic review and meta‐analysis

Penny-Dimri

Bergmeir²,

Perry

et al. 2022

Journal of Cardiac Surgery

Self Cite

View full text Add to dashboard Cite

Background Machine learning (ML) models are promising tools for predicting adverse postoperative outcomes in cardiac surgery, yet have not translated to routine clinical use. We conducted a systematic review and meta‐analysis to assess the predictive performance of ML approaches. Methods We conducted an electronic search to find studies assessing ML and traditional statistical models to predict postoperative outcomes. Our primary outcome was the concordance (C‐) index of discriminative performance. Using a Bayesian meta‐analytic approach we pooled the C‐indices with the 95% credible interval (CrI) across multiple outcomes comparing ML methods to logistic regression (LR) and clinical scoring tools. Additionally, we performed critical difference and sensitivity analysis. Results We identified 2792 references from the search of which 51 met inclusion criteria. Two postoperative outcomes were amenable for meta‐analysis: 30‐day mortality and in‐hospital mortality. For 30‐day mortality, the pooled C‐index and 95% CrI were 0.82 (0.79−0.85), 0.80 (0.77−0.84), 0.78 (0.74−0.82) for ML models, LR, and scoring tools respectively. For in‐hospital mortality, the pooled C‐index was 0.81 (0.78−0.84) and 0.79 (0.73−0.84) for ML models and LR, respectively. There were no statistically significant results indicating ML superiority over LR. Conclusion In cardiac surgery patients, for the prediction of mortality, current ML methods do not have greater discriminative power over LR as measured by the C‐index.

show abstract

Artificial intelligence in cardiac surgery: A systematic review

Sulague,

Beloy,

Medina

et al. 2024

World j. surg.

View full text Add to dashboard Cite

BackgroundArtificial intelligence (AI) has emerged as a tool to potentially increase the efficiency and efficacy of cardiovascular care and improve clinical outcomes. This study aims to provide an overview of applications of AI in cardiac surgery.MethodsA systematic literature search on AI applications in cardiac surgery from inception to February 2024 was conducted. Articles were then filtered based on the inclusion and exclusion criteria and the risk of bias was assessed. Key findings were then summarized.ResultsA total of 81 studies were found that reported on AI applications in cardiac surgery. There is a rapid rise in studies since 2020. The most popular machine learning technique was random forest (n = 48), followed by support vector machine (n = 33), logistic regression (n = 32), and eXtreme Gradient Boosting (n = 31). Most of the studies were on adult patients, conducted in China, and involved procedures such as valvular surgery (24.7%), heart transplant (9.4%), coronary revascularization (11.8%), congenital heart disease surgery (3.5%), and aortic dissection repair (2.4%). Regarding evaluation outcomes, 35 studies examined the performance, 26 studies examined clinician outcomes, and 20 studies examined patient outcomes.ConclusionAI was mainly used to predict complications following cardiac surgeries and improve clinicians' decision‐making by providing better preoperative risk assessment, stratification, and prognostication. While the application of AI in cardiac surgery has greatly progressed in the last decade, further studies need to be conducted to verify accuracy and ensure safety before use in clinical practice.

show abstract

Machine Learning Outperforms Existing Clinical Scoring Tools in the Prediction of Postoperative Atrial Fibrillation During Intensive Care Unit Admission After Cardiac Surgery

Cited by 20 publications

References 22 publications

Artificial Intelligence in Cardiac Surgery: A Systematic Review

Artificial Intelligence in Cardiac Surgery: A Systematic Review

Machine learning to predict adverse outcomes after cardiac surgery: A systematic review and meta‐analysis

Artificial intelligence in cardiac surgery: A systematic review

Contact Info

Product

Resources

About