What is it about?
STUDY AIMS: We have examined the correlation between: Hb(Hemoglobin), Serum Creatinine, LDL cholesterol, HDL cholesterol, Triglycerides, ALT, AST, hs-cTnI(High-sensitivity cardiac troponin I), CRP(C-Reactive Protein) and the risk of heart failure over ischemic heart disease in elderly population. We aimed to answer the question: which parameters increase the risk of heart failure (HF) over ischemic heart disease (IHD)? Following our findings: AST, ALT and CRP increase the risk of HF over IHD, while Hb and HDL reduce the risk of HF over IHD. Hb(Hemoglobin) was found as major contributor in the relative risk decrease, reducing the risk of HF over IHD for 21.18 % on average per unit increase. We extended our work to the domain of machine learning and a model that can successfully discriminate between HF and IHD was identified. *Original dataset used in the research: https://ars.els-cdn.com/content/image/1-s2.0-S1018364723000356-mmc2.xlsx *This dataset is integral part of the publication: https://doi.org/10.1016/j.jksus.2023.102573. *The dataset may be used for further scientific investigation, given appropriate citation of our work at: https://doi.org/10.1016/j.jksus.2023.102573.
Featured Image
Photo by National Cancer Institute on Unsplash
Why is it important?
Main findings: - Hb unit-increase reduces the risk of HF over IHD for 21.18 %, p-value < 0.05. - HDL unit-increase reduces the risk of HF over IHD for 3.83 %, p-value < 0.05. - AST unit-increase increases the risk of HF over IHD for 3.43 %, p-value < 0.05. - ALT unit-increase increases the risk of HF over IHD for 2.46 %, p-value < 0.05. - CRP unit-increase increases the risk of HF over IHD for 4.11 %, p-value < 0.05. - Logistic regression machine learning model of parameters: Hb + Serum Creatinine + AST + hs-cTnI + CRP can successfully discriminate between HF and IHD, mean AUROC = 0.805 (results from 20-fold cross-validation).
Perspectives
We used blood tests results from 167 study participants to compute the relative risk of HF over IHD in elderly population. Our findings suggest that unit increase of AST, ALT or CRP increases the risk of HF against CIHD for 3.43 %, 2.46 % and 4.11 % respectively, p-value < 0.05. On the other hand, unit increase of Hb or HDL reduces the risk of HF against IHD for 21.18 % and 3.83 % respectively, p-value < 0.05. Our application to the domain of machine learning, showed that logistic regression ML model over Hb + Serum Creatinine + AST + hs-cTnI + CRP can predict the outcome of HF over IHD with high accuracy.
Ph.D Done Stojanov
Univerzitet Goce Delcev Stip
Read the Original
This page is a summary of: Predicting the outcome of heart failure against chronic-ischemic heart disease in elderly population – Machine learning approach based on logistic regression, case to Villa Scassi hospital Genoa, Italy, Journal of King Saud University - Science, April 2023, Elsevier,
DOI: 10.1016/j.jksus.2023.102573.
You can read the full text:
Contributors
The following have contributed to this page
