Adeel Khan

This longitudinal study evaluates the impact of a specific intervention on cardiac development and function in a pediatric cohort from age 1 to age 5. By utilizing linear mixed-effects models, the project tracks the trajectories of systolic performance and structural remodeling, specifically measuring Ejection Fraction, Left Ventricular Mass, and Wall Thickness. The primary objective is to determine if the treatment alters the natural growth patterns of the heart compared to a standard control group.

The results demonstrate that while both groups undergo age-related physiological changes, the treated group exhibits significantly accelerated improvements in heart function and more pronounced structural changes. By accounting for individual variability through random effects, the project provides robust evidence that the intervention drives a unique remodeling process characterized by enhanced pumping efficiency and distinct morphological adaptation.

The analysis of reaction time through scatter plots indicates distinct relationships with intelligence, hormone levels, and training. Specifically, a positive correlation exists between intelligence and reaction time, as well as hormone levels and reaction time, implying that higher intelligence and hormone levels result in increased reaction times. Conversely, training exhibits a negative correlation, meaning that more training leads to faster reaction times. The London lab records the highest median reaction time, followed by the Munich and Boston labs. Classic conditions lead to higher median reaction times compared to modern and experimental conditions. Additionally, individuals with an avoiding personality type show higher median reaction times than those with an approaching personality type. The performance evaluation of two machine learning models, a decision tree and a random forest, demonstrates differing results on the training and test sets. While the random forest model shows superior performance on the training set with lower RMSE, MAE, and higher R square, the decision tree model excels on the test set, exhibiting lower RMSE and MAPE values. This indicates that the decision tree model generalizes better on unseen data, making it more reliable for predicting reaction time in practical applications.