Detecting hypovolemia in postoperative patients using a discrete Fourier transform

Area of Science:

• Medical Diagnostics
• Cardiovascular Physiology
• Signal Processing

Background:

• Hypovolemia, a state of decreased blood volume, presents diagnostic challenges.
• Accurate and timely diagnosis of hypovolemia is critical for patient outcomes.

Purpose of the Study:

• To identify waveform-derived variables for precise hypovolemia diagnosis.
• To evaluate the diagnostic utility of the amplitude of the first harmonic (A1) in arterial blood pressure waveforms.

Main Methods:

• Analysis of arterial blood pressure waveforms from hypovolemic postoperative patients using discrete Fourier transform.
• Paired samples t-test to compare A1 between normovolemic and hypovolemic groups.
• Logistic regression modeling using A1 to categorize patients.

Main Results:

• The amplitude of the first harmonic (A1) was significantly higher in normovolemic patients compared to hypovolemic patients (p < 0.001).
• A preliminary study using logistic regression correctly identified hypovolemic patients with 80% accuracy.
• The developed method demonstrated a 20% misclassification rate.

Conclusions:

• The amplitude of the first harmonic (A1) is a potentially valuable parameter for hypovolemia detection.
• Waveform analysis of arterial blood pressure offers a promising non-invasive approach to diagnosing hypovolemia.