Machine Learning Model Predicts Myocardial Ischemia Risk at High Altitudes: Study
A new study has revealed that a machine learning model can accurately predict the risk of myocardial ischemia at high altitudes. The research, conducted by a team of scientists, aimed to address the gap in knowledge regarding this specific health condition and provide valuable insights for the safety of military personnel and others exposed to high altitude environments.
The study employed a prospective cohort design, involving soldiers who underwent health examinations at the 920th Hospital of the Joint Logistic Support Force between January and June 2022. These individuals were scheduled to undergo high-altitude training within six months. In total, 4000 participants took part in the health examination, including 3800 men and 200 women aged 18 to 54.
After careful screening, 1093 individuals were excluded from the study due to various health conditions such as pneumonia, asthma, hypertension, and abnormal electrocardiogram findings. Additionally, 904 participants who did not ultimately participate in high-altitude training were also excluded. The final sample comprised 2855 individuals, including 2810 men and 45 women.
To determine the presence of myocardial ischemia, the researchers evaluated the electrocardiogram (ECG) results of the participants. Specifically, they looked for specific criteria including ST depression and T wave inversion in consecutive leads. The participants were required to abstain from physical training for at least 48 hours before the ECG examination.
A total of 27 variables were used in the machine learning analysis, including age, gender, body mass index, blood pressure, heart rate, oxygen saturation, and various other measures. The data was preprocessed, standardized, and randomly divided into training and test sets.
In order to build an effective predictive model, the researchers utilized algorithms such as logistic regression, random forest, XGBoost, K-nearest neighbors, and support vector machines. The models were tested and evaluated using the area under the receiver operating characteristic curve (AUC) and calibration curve performance.
Furthermore, the Recursive Feature Elimination (RFE) algorithm was applied to identify the most influential variables for an optimal and clinically feasible model. This model was compared to the full dataset model and found to be well-suited for practical use.
The study adhered to the ethical guidelines of the Helsinki Declaration, with the experimental protocol approved by the Human Ethics Committee of the 920th Hospital. Written informed consent was obtained from all participants.
This groundbreaking research provides valuable insights into predicting the risk of myocardial ischemia at high altitudes using machine learning. By identifying key clinical features and developing a reliable predictive model, this study contributes to the safety and well-being of individuals, particularly military personnel, operating in high altitude environments. Further research and validation of these findings could have significant implications in preventing and managing cardiovascular health risks at altitude.
