Quincy A Hathaway1, Naveena Yanamala2, Matthew J Budoff3, Partho P Sengupta4, Irfan Zeb1. 1. Heart and Vascular Institute, West Virginia University, Morgantown, WV, USA. 2. Heart and Vascular Institute, West Virginia University, Morgantown, WV, USA; Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, USA. 3. Lundquist Institute, Harbor-University of California, Los Angeles, Torrance, CA, USA. 4. Heart and Vascular Institute, West Virginia University, Morgantown, WV, USA; Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, USA. Electronic address: partho.sengupta@rutgers.edu.
Abstract
BACKGROUND: There is growing interest in utilizing machine learning techniques for routine atherosclerotic cardiovascular disease (ASCVD) risk prediction. We investigated whether novel deep learning survival models can augment ASCVD risk prediction over existing statistical and machine learning approaches. METHODS: 6814 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) were followed over 16 years to assess incidence of all-cause mortality (mortality) or a composite of major adverse events (MAE). Features were evaluated within the categories of traditional risk factors, inflammatory biomarkers, and imaging markers. Data was split into an internal training/testing (four centers) and external validation (two centers). Both machine learning (COXPH, RSF, and lSVM) and deep learning (nMTLR and DeepSurv) models were evaluated. RESULTS: In comparison to the COXPH model, DeepSurv significantly improved ASCVD risk prediction for MAE (AUC: 0.82 vs. 0.80, P ≤ 0.001) and mortality (AUC: 0.87 vs. 0.84, P ≤ 0.001) with traditional risk factors alone. Implementing non-categorical NRI, we noted a >40% increase in correct reclassification compared to the COXPH model for both MAE and mortality (P ≤ 0.05). Assessing the relative risk of participants, DeepSurv was the only learning algorithm to develop a significantly improved risk score criteria, which outcompeted COXPH for both MAE (4.22 vs. 3.61, P = 0.043) and mortality (6.81 vs. 5.52, P = 0.044). The addition of inflammatory or imaging biomarkers to traditional risk factors showed minimal/no significant improvement in model prediction. CONCLUSION: DeepSurv can leverage simple office-based clinical features alone to accurately predict ASCVD risk and cardiovascular outcomes, without the need for additional features, such as inflammatory and imaging biomarkers.
BACKGROUND: There is growing interest in utilizing machine learning techniques for routine atherosclerotic cardiovascular disease (ASCVD) risk prediction. We investigated whether novel deep learning survival models can augment ASCVD risk prediction over existing statistical and machine learning approaches. METHODS: 6814 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) were followed over 16 years to assess incidence of all-cause mortality (mortality) or a composite of major adverse events (MAE). Features were evaluated within the categories of traditional risk factors, inflammatory biomarkers, and imaging markers. Data was split into an internal training/testing (four centers) and external validation (two centers). Both machine learning (COXPH, RSF, and lSVM) and deep learning (nMTLR and DeepSurv) models were evaluated. RESULTS: In comparison to the COXPH model, DeepSurv significantly improved ASCVD risk prediction for MAE (AUC: 0.82 vs. 0.80, P ≤ 0.001) and mortality (AUC: 0.87 vs. 0.84, P ≤ 0.001) with traditional risk factors alone. Implementing non-categorical NRI, we noted a >40% increase in correct reclassification compared to the COXPH model for both MAE and mortality (P ≤ 0.05). Assessing the relative risk of participants, DeepSurv was the only learning algorithm to develop a significantly improved risk score criteria, which outcompeted COXPH for both MAE (4.22 vs. 3.61, P = 0.043) and mortality (6.81 vs. 5.52, P = 0.044). The addition of inflammatory or imaging biomarkers to traditional risk factors showed minimal/no significant improvement in model prediction. CONCLUSION: DeepSurv can leverage simple office-based clinical features alone to accurately predict ASCVD risk and cardiovascular outcomes, without the need for additional features, such as inflammatory and imaging biomarkers.