Combining symbolic regression with the Cox proportional hazards model improves prediction of heart failure deaths

Abstract

Heart failure is a clinical syndrome characterised by a reduced ability of the heart to pump blood. Patients with heart failure have a high mortality rate, and physicians need reliable prognostic predictions to make informed decisions about the appropriate application of devices, transplantation, medications, and palliative care. In this study, we demonstrate that combining symbolic regression with the Cox proportional hazards model improves the ability to predict death due to heart failure compared to using the Cox proportional hazards model alone.We used a newly invented symbolic regression method called the QLat-tice to analyse a data set of medical records for 299 Pakistani patients diagnosed with heart failure. The QLattice identified a minimal set of mathematical transformations of the available covariates, which we then used in a Cox model to predict survival.An exponential function of age, the inverse of ejection fraction, and the inverse of serum creatinine were identified as the best risk factors for predicting heart failure deaths. A Cox model fitted on these transformed covariates had improved predictive performance compared with a Cox model on the same covariates without mathematical transformations.Symbolic regression is a way to find transformations of covariates from patients’ medical records which can improve the performance of survival regression models. At the same time, these simple functions are intuitive and easy to apply in clinical settings. The direct interpretability of the simple forms may help researchers gain new insights into the actual causal pathways leading to deaths.

More than a Feeling: Dermatological Changes Impacted by Spaceflight

“…In biological pathway analysis, it is well-known that up-or down-regulation of one gene can have cascading effects such that the function of one gene becomes sensitive to that of another 77 . It has previously been demonstrated that parsimonious machine learning models are able to provide accurate outcome prediction in omics data, while preserving interpretability 78,79 . The interpretability often results from the fact that models might demonstrate otherwise opaque relations which become clear when combined effects are taken into account.…”

Prediction of Survival of Heart Failure Patients: An Application of Classification-Based Machine Learning Algorithms

“…Benzer olarak, bir başka çalışmada atrial fibrilasyon, anemi ve böbrek hastalıklarını da göz önünde bulundurarak 6 fenogrup oluşturuldu (Hedman et al 2020). Pakistan'daki 299 kalp yetmezliği hastası için Qlattice adında sembolik regresyon yöntemine dayanan yeni bir model ile minimal bir matematiksel dönüşüm seti tanımlandı ve bu set Cox modelinde tahmin için kullanıldı (Wilstup and Cave, 2021). Topluluk ağaçları makine öğrenimi tekniklerini kullanarak geliştirilen modelde, Aşırı Gradyan Arttırma (XGBoost, Extreme Gradient Boosting) diğer topluluk ağaçları yöntemlerine göre daha iyi bir performans göstermiştir (Moreno-Sanchez, 2020).…”