Table 2 Summary of comparative analysis of various health biomarkers between male and female subgroups with respect to variance and mean values

From: A supervised machine learning approach with feature selection for sex-specific biomarker prediction

Biomarker

Levene’s p-value (σ2)

Unpaired Sample t-Test

aU-Statistic

bMann-Whitney U-Test

Waist Circumference (cm)

0.185

< 0.05

2.07e + 05

< 0.05

BMI (kg/m²)

< 0.05

0.121

1.75e + 05

0.406

Albuminuria (mg/L)

0.516

< 0.05

2.43e + 05

< 0.05

UrAlbCr (mg/g)

< 0.05

< 0.05

1.20e + 05

< 0.05

Uric Acid (mmol/L)

0.665

< 0.05

2.85e + 05

< 0.05

Blood Glucose (mmol/L)

0.436

< 0.05

2.15e + 05

< 0.05

HDL (mmol/L)

< 0.05

< 0.05

1.04e + 05

< 0.05

Triglycerides (mmol/L)

0.060

< 0.05

2.05e + 05

< 0.05

Age (years)

0.728

0.596

1.76e + 05

0.579

Systolic Blood Pressure (mmHg)

< 0.05

< 0.05

2.22e + 05

< 0.05

  1. aThe U-Statistic represents the computed value from the test, which is utilised to determine the p-value. In p-values below 0.05, indicating statistically significant differences between sexes for each biomarker.
  2. bThe Mann-Whitney U-Test was used to evaluate if there are significant differences in biomarker values between sexes. As a non-parametric test for independent samples, it does not assume normal distribution.
Back to article page