Improving Accuracy of Noninvasive Hemoglobin Monitors: A Functional Regression Model for Streaming SpHb Data

Area of Science:

• Biomedical Engineering
• Critical Care Medicine
• Medical Devices

Background:

• Noninvasive hemoglobin (SpHb) monitoring is crucial for timely clinical decisions.
• Existing SpHb monitors have limitations in accuracy, impacting patient care.
• Accurate hemoglobin monitoring is vital for managing trauma patients and transfusion protocols.

Purpose of the Study:

• To develop and validate a novel method for improving the accuracy of SpHb monitors.
• To enhance critical care protocols in trauma settings through more precise hemoglobin measurements.
• To provide a more reliable tool for predicting hemoglobin levels in real-time.

Main Methods:

• Fitting smooth spline functions to SpHb measurements over a defined time window.
• Utilizing a functional regression model to predict true hemoglobin (HgB) values.
• Comparing the accuracy of the proposed method against traditional SpHb measurement techniques.

Main Results:

• The proposed method demonstrated improved accuracy in predicting hemoglobin levels.
• Mean absolute error (MAE) was reduced from 1.26 g/dL (traditional methods) to 1.08 g/dL (proposed method).
• The spline fitting approach offers a significant enhancement over raw SpHb readings.

Conclusions:

• Fitting smooth functions to SpHb measurements enhances the accuracy of hemoglobin predictions.
• Improved HgB prediction accuracy can inform sophisticated decision-making for blood product transfusions.
• This method has the potential to optimize transfusion strategies and improve patient outcomes in critical care.