Application of machine learning approaches for osteoporosis risk prediction in postmenopausal women

Abstract Introduction: Coronary microvascular disease (CMD) affects small arteries that feed the heart and is more prevalent in postmenopausal women. Since CMD and Coronary artery disease (CAD) have distinct pathologies, but are treated the same way, the majority of the patients with CMD do not receive a proper diagnosis and treatment, which in turn results in higher rates of adverse future events such as heart failure, sudden cardiac death, and acute coronary syndrome (ACS). Previously, we performed full metabolite profiling of plasma samples using GC-MS analysis and tested their classification performance using machine learning approaches. This initial proof-of-concept study showed that plasma metabolite profiles can be used to develop diagnostic signatures for CMD. In the current study, we hypothesize that plasma metabolite and protein composition is different for postmenopausal women with no heart disease, with CAD, or with CMD. Methods: We obtained plasma samples from 70 postmenopausal women who are healthy, women who have CMD, and women who have CAD at the time of blood collection. In addition to GC-MS metabolite profiles, we performed LC-MS metabolomic profiling, and proteomic profiling of a panel of 92 proteins that were implicated in cardiometabolic disease. We identified a combination of metabolites and proteins, and further tested their classification performance using machine learning approaches to identify potential circulating biomarkers for CMD. Results: We identified a comprehensive list of metabolites and proteins that were involved in endothelial cell function, nitric oxide metabolism and inflammation, which significantly different in plasma from women with CMD. We further validated difference in the level of several protein biomarkers, such as RAGE, PTX3, AGRP, CNTN1, and MMP-3, which are statistically significantly higher in postmenopausal women with CMD when compared with healthy women or women with CAD. Conclusion: Our research identified a group of potential molecules that can be used in the design of easy and low-cost blood biomarkers for the clinical diagnosis of CMD.

Download Full-text

Improving Clinical Translation of Machine Learning Approaches Through Clinician-Tailored Visual Displays of Black Box Algorithms: Development and Validation (Preprint)

10.2196/preprints.15791 ◽

2019 ◽

Author(s):

Shannon Wongvibulsin ◽

Katherine C Wu ◽

Scott L Zeger

Keyword(s):

Machine Learning ◽

Random Forest ◽

Predictive Model ◽

Risk Prediction ◽

Clinical Care ◽

Left Ventricular ◽

Black Box ◽

Clinical Translation ◽

Learning Approaches ◽

Visual Displays

BACKGROUND Despite the promise of machine learning (ML) to inform individualized medical care, the clinical utility of ML in medicine has been limited by the minimal interpretability and black box nature of these algorithms. OBJECTIVE The study aimed to demonstrate a general and simple framework for generating clinically relevant and interpretable visualizations of black box predictions to aid in the clinical translation of ML. METHODS To obtain improved transparency of ML, simplified models and visual displays can be generated using common methods from clinical practice such as decision trees and effect plots. We illustrated the approach based on postprocessing of ML predictions, in this case random forest predictions, and applied the method to data from the Left Ventricular (LV) Structural Predictors of Sudden Cardiac Death (SCD) Registry for individualized risk prediction of SCD, a leading cause of death. RESULTS With the LV Structural Predictors of SCD Registry data, SCD risk predictions are obtained from a random forest algorithm that identifies the most important predictors, nonlinearities, and interactions among a large number of variables while naturally accounting for missing data. The black box predictions are postprocessed using classification and regression trees into a clinically relevant and interpretable visualization. The method also quantifies the relative importance of an individual or a combination of predictors. Several risk factors (heart failure hospitalization, cardiac magnetic resonance imaging indices, and serum concentration of systemic inflammation) can be clearly visualized as branch points of a decision tree to discriminate between low-, intermediate-, and high-risk patients. CONCLUSIONS Through a clinically important example, we illustrate a general and simple approach to increase the clinical translation of ML through clinician-tailored visual displays of results from black box algorithms. We illustrate this general model-agnostic framework by applying it to SCD risk prediction. Although we illustrate the methods using SCD prediction with random forest, the methods presented are applicable more broadly to improving the clinical translation of ML, regardless of the specific ML algorithm or clinical application. As any trained predictive model can be summarized in this manner to a prespecified level of precision, we encourage the use of simplified visual displays as an adjunct to the complex predictive model. Overall, this framework can allow clinicians to peek inside the black box and develop a deeper understanding of the most important features from a model to gain trust in the predictions and confidence in applying them to clinical care.

Download Full-text

Machine learning approaches as an alternative to traditional statistical methods in cardiovascular risk prediction

Biomedical Science and Engineering ◽

10.4081/bse.2021.195 ◽

2021 ◽

Vol 4 (s1) ◽

Author(s):

Michela Sperti ◽

Fabrizio D’Ascenzo ◽

Luca Navarini ◽

Giacomo Di Benedetto ◽

Antonella Afeltra ◽

...

Keyword(s):

Machine Learning ◽

Acute Coronary Syndrome ◽

Cardiovascular Risk ◽

Psoriatic Arthritis ◽

Risk Prediction ◽

Inflammatory Arthritis ◽

Learning Approaches ◽

Coronary Syndrome ◽

Cardiovascular Risk Prediction ◽

Risk Predictors

Machine Learning (ML) algorithms have proven promising methodologies in improving Cardiovascular (CV) risk predictors based on traditional statistics. In the present work, two case studies are reported: CV risk prediction in patients affected by Inflammatory Arthritis (IA), with attention to Psoriatic Arthritis (PsA), and patients who experienced Acute Coronary Syndrome (ACS).

Download Full-text

Machine Learning Approaches in COVID-19 Diagnosis, Mortality, and Severity Risk Prediction: A Review

Informatics in Medicine Unlocked ◽

10.1016/j.imu.2021.100564 ◽

2021 ◽

pp. 100564

Author(s):

Norah Alballa ◽

Isra Al-Turaiki

Keyword(s):

Machine Learning ◽

Risk Prediction ◽

Learning Approaches

Download Full-text

Predicting Acute Kidney Injury after Cardiac Surgery by Machine Learning Approaches

Journal of Clinical Medicine ◽

10.3390/jcm9061767 ◽

2020 ◽

Vol 9 (6) ◽

pp. 1767 ◽

Cited By ~ 1

Author(s):

Charat Thongprayoon ◽

Panupong Hansrivijit ◽

Tarun Bathini ◽

Saraschandra Vallabhajosyula ◽

Poemlarp Mekraksakit ◽

...

Keyword(s):

Machine Learning ◽

Cardiac Surgery ◽

Risk Prediction ◽

Prediction Models ◽

Clinical Medicine ◽

Kidney Injury ◽

Risk Scores ◽

Learning Approaches ◽

Risk Prediction Models ◽

General Utility

Cardiac surgery-associated AKI (CSA-AKI) is common after cardiac surgery and has an adverse impact on short- and long-term mortality. Early identification of patients at high risk of CSA-AKI by applying risk prediction models allows clinicians to closely monitor these patients and initiate effective preventive and therapeutic approaches to lessen the incidence of AKI. Several risk prediction models and risk assessment scores have been developed for CSA-AKI. However, the definition of AKI and the variables utilized in these risk scores differ, making general utility complex. Recently, the utility of artificial intelligence coupled with machine learning, has generated much interest and many studies in clinical medicine, including CSA-AKI. In this article, we discussed the evolution of models established by machine learning approaches to predict CSA-AKI.

Download Full-text