Skip to main content

Table 3 The feature difference analysis for different ICV ranges

From: Machine learning approach for predicting post-intubation hemodynamic instability (PIHI) index values: towards enhanced perioperative anesthesia quality and safety

Features

Levene Statistic

(P value)

Welch Statistic

(P value)

Brown-Forsythe Statistic

(P value)

Sex

5.591 (0.000)

1.277 (0.279)

1.282 (0.276)

Age

3.860 (0.004)

248.949 (0.000)

252.249 (0.000)

Height (cm)

2.624 (0.033)

23.075 (0.000)

24.891 (0.000)

Weight (kg)

18.782 (0.000)

2.146 (0.075)

2.155 (0.073)

BMI

12.310 (0.000)

3.202 (0.014)

2.986 (0.019)

Hypertension

102.908 (0.000)

24.034 (0.000)

22.978 (0.000)

Pre SP (mmHg)

11.700 (0.000)

108.014 (0.000)

95.847 (0.000)

Pre DP (mmHg)

3.729 (0.005)

18.817 (0.000)

16.912 (0.000)

Pre HR (BPM)

2.645 (0.032)

4.003 (0.004)

3.950 (0.004)

Fentanyl (μg/kg)

3.905 (0.004)

0.969 (0.425)

0.977 (0.419)

Lidocaine (mg/kg)

4.112 (0.002)

0.524 (0.718)

0.414 (0.798)

Propofol (mg/kg)

4.828 (0.001)

3.410 (0.010)

2.759 (0.028)

Rocuronium (mg/kg)

8.808 (0.000)

2.470 (0.045)

2.495 (0.042)

  1. The sex and hypertension are discretized to 0–1 value; Welch variance and Brown-Forsythe variance are utilized as all features failed to pass the Levene’s test for equality of variances.
  2. Pre preoperative, SP systolic pressure, DP diastolic pressure, HR heart rate, BMI body mass index, BPM beats per minute.