Related Experiment Videos
Multivariate computational analysis of biosensor's data for improved CD64 quantification for sepsis diagnosis
1Department of Bioengineering, University of Illinois at Urbana-Champaign, 1270 Digital Computer Laboratory, 1304 W. Springfield Ave, Urbana, IL 61801, USA. uhassan2@illinois.edu rbashir@illinois.edu and Micro and Nanotechnology Lab, University of Illinois at Urbana-Champaign, 208 N. Wright St., Urbana, IL 61801, USA and Stevens Family Biomedical Research Center, Carle Foundation Hospital, Urbana, IL 61801, USA.
Lab on a Chip
|March 23, 2018
Summary
A novel biosensor improves sepsis diagnosis by accurately measuring leukocyte counts and neutrophil CD64 expression. Artificial neural networks enhance quantification, offering a more precise tool for early sepsis detection and patient management.
Area of Science:
- Biomedical Engineering
- Immunology
- Computational Biology
Background:
- Sepsis is a major global health threat, often diagnosed using non-specific Systemic Inflammatory Response Syndrome (SIRS) criteria.
- Leukocyte counts and neutrophil CD64 expression are specific biomarkers for sepsis, but traditional diagnostic methods have limitations.
Purpose of the Study:
- To develop and validate a biosensor for rapid enumeration of leukocyte counts and quantification of neutrophil CD64 expression.
- To investigate the use of artificial neural networks (ANNs) for improved accuracy in quantifying CD64 expression from biosensor data.
- To enhance sepsis diagnosis and prognosis through advanced biosensor analytics.
Main Methods:
- Development of a biosensor chip capable of analyzing a small blood sample for leukocyte counts and CD64 expression.
- Application of multivariate computational models, specifically ANNs, to analyze biosensor data.
- Comparison of ANN-based quantification with traditional univariate regression and flow cytometry.
Main Results:
- The biosensor demonstrated accurate enumeration of leukocyte counts and quantification of neutrophil CD64 expression.
- ANNs significantly improved the accuracy of CD64 expression quantification from biosensor measurements, showing high coefficients of determination and low error.
- The ANN approach outperformed commonly used univariate regression methods in accuracy.
Conclusions:
- The developed biosensor, coupled with ANN-based data analysis, offers a highly accurate and specific method for sepsis biomarker assessment.
- This approach represents a significant advancement in biosensor data analytics, enabling more precise quantification by utilizing multiple data features.
- The technology holds promise for improving early sepsis detection, patient monitoring, and overall clinical outcomes.