Logo

American Heart Association

  2
  0


Final ID: Mo4129

A Machine Learning Approach to Predict Percutaneous Coronary Intervention in Patients with Critical Illness and Signs of Myocardial Injury

Abstract Body (Do not enter title and authors here):
Background
Critically ill patients frequently experience multi-organ dysfunction and unstable vital signs, increasing their risk of myocardial injury and elevated serum troponin levels. Myocardial injury from oxygen supply-demand mismatch, whether due to systemic disease or obstructive coronary artery disease (CAD), often leads to patients undergoing an invasive coronary angiogram (ICA) and potentially percutaneous coronary intervention (PCI) if obstructive CAD is identified. However, this procedure carries significant risks for critically ill patients. To address this challenge, we aimed to develop a machine learning (ML) model to predict the need for PCI in critically ill patients with myocardial injury.

Methods
The study cohort included critically ill patients with elevated serum troponin levels who underwent ICA during their hospitalization at a single institution from 2012 to 2022. We defined critical illness as a life-threatening condition requiring intensive support for vital sign instability. Patients with atherothrombosis identified during ICA were excluded from the study, focusing only on patients with type 2 MI. We utilized Matplotlib to analyze patient data, including demographics, lab results, vital signs, and comorbidities. We then used XGBoost, a supervised ML algorithm, for binary classification tasks. An ensemble of decision trees was employed, with gradient boosting to correct errors from previous trees. XGBoost iteratively adjusted predictions, creating a model to predict which patients required PCI.

Results
Out of 765 patients, 173 (22.6%) underwent PCI. The cohort had a mean age of 60.5 ± 13.5 years, with 41.7% being female. The ML model accurately identified non-PCI candidates at 75% and PCI candidates at 64%, achieving an overall accuracy of 73%. The three most important features identified by the model were a history of diabetes, diastolic heart failure, and whether the patient received a blood transfusion during hospitalization.

Conclusion
A ML approach using XGBoost demonstrated moderate accuracy in predicting the need for PCI in critically ill patients with signs of myocardial injury. This prediction holds clinical significance due to the heightened risks associated with ICA in these patients. Given that PCI occurred in only 22.6% of the cohort, future models should include a higher proportion of patients who underwent PCI to improve the robustness and true positive accuracy of the model.
  • Mueller, Joshua  ( University of Arkansas for Medical Sciences NW , Fayetteville , Arkansas , United States )
  • Stepanova, Daria  ( University of Arkansas , Fayetteville , Arkansas , United States )
  • Chidambaram, Vignesh  ( University of Arkansas for Medical Sciences , Little Rock , Arkansas , United States )
  • Nakarmi, Ukash  ( University of Arkansas , Fayetteville , Arkansas , United States )
  • Al'aref, Subhi  ( University of Arkansas for Medical Sciences , Little Rock , Arkansas , United States )
  • Author Disclosures:
    Joshua Mueller: DO NOT have relevant financial relationships | Daria Stepanova: No Answer | Vignesh Chidambaram: No Answer | Ukash Nakarmi: DO NOT have relevant financial relationships | Subhi Al'Aref: DO have relevant financial relationships ; Advisor:Shockwave Medical:Past (completed) ; Royalties/Patent Beneficiary:Elsevier:Active (exists now)
Meeting Info:

Scientific Sessions 2024

2024

Chicago, Illinois

Session Info:

Critical Care Cardiology Medley Reprise

Monday, 11/18/2024 , 10:30AM - 11:30AM

Abstract Poster Session

More abstracts on this topic:
Association of Estimated Pulse Wave Velocity on Endovascular Treatment Outcome: A Secondary Analysis of the OPTIMAL-BP trial

Han Minho, Joo Haram, Lee Hyungwoo, Heo Joonnyung, Jung Jae Wook, Kim Young Dae, Park Eunjeong, Nam Hyo Suk

An Integrated Germline and Somatic Genomic Model Improves Risk Prediction for Coronary Artery Disease

Kim Min Seo, Zhang Rufan, Liu Zhaoqi, A Yunga, Ellinor Patrick, Natarajan Pradeep, Wang Minxian, Fahed Akl, Yang Xiong, Uddin Md Mesbah, Nakao Tetsushi, Cho So Mi Jemma, Koyama Satoshi, Zhu Xinyu, Xu Tingfeng, Reeskamp Laurens

More abstracts from these authors:
Revascularization of Patients with Low-Density Non-Calcified Plaque was Associated with Lower Occurrence of Acute Coronary Syndrome

Malkasian Shant, Conte Edoardo, Maffei Erica, Plank Fabian, Pontone Gianluca, Feuchtner Gudrun, Samady Habib, Park Hyung-bok, Danad Ibrahim, Cho Iksung, Bax Jeroen, Hubbard Logan, Lee Ji Hyun, Baskaran Lohendran, Hadamitzky Martin, Budoff Matthew, Stone Peter, Heo Ran, Cury Ricardo, Lee Sang-eun, Al'aref Subhi, Stuijfzand Wijnand, Nurmohamed Nick, Lu Yao, Kim Yong-jin, Cademartiri Filippo, Virmani Renu, Narula Jagat, Rizvi Asim, Chinnaiyan Kavitha, Villines Todd, Leipsic Jonathon, Shaw Leslee, Chang Hyuk-jae, Molloi Sabee, Choi Andrew, Marques Hugo, Van Rosendael Alexander, Chow Benjamin, Andreini Daniele

Class 1C Antiarrhythmic Drugs are Safe in Patients with Arrhythmia-induced Cardiomyopathy Following Recovery of Left Ventricular Ejection Fraction

Pohlkamp Ryan, Chidambaram Vignesh, Kumar Amudha, Ghanta Nikhila, Odueke Adetayo, Vallurupalli Srikanth, Paydak Hakan

You have to be authorized to contact abstract author. Please, Login
Not Available