Related Experiment Video
Updated: Jul 6, 2026
08:46
Electronic Tongue Generating Continuous Recognition Patterns for Protein Analysis
Published on: September 16, 2014
7.8K
C-reactive protein (CRP) evaluation in human urine using optical sensor supported by machine learning
Kacper Cierpiak1, Paweł Wityk2,3,4, Monika Kosowska5
1Department of Metrology and Optoelectronics, Faculty of Informatics, Telecommunications and Informatics, Gdańsk University of Technology, Narutowicza Street 11/12, 80-233, Gdańsk, Poland.
Scientific Reports
|August 14, 2024
Summary
This study presents an optical fiber sensor for detecting C-Reactive Protein (CRP), a key inflammation marker. The sensor achieved high accuracy in distinguishing inflammation states in patient samples, aiding medical diagnostics.
Area of Science:
- Biomedical Engineering
- Optical Sensing
- Medical Diagnostics
Background:
- C-Reactive Protein (CRP) is a crucial biomarker for inflammation.
- Accurate CRP level determination is vital for diagnosing various conditions, including infections and cardiovascular diseases.
- Existing diagnostic methods may require improvement in speed and sensitivity.
Purpose of the Study:
- To develop and validate an interferometric sensor for sensitive C-Reactive Protein (CRP) detection.
- To differentiate between normal and elevated CRP levels indicative of inflammation.
- To assess the sensor's performance in real patient samples.
Main Methods:
- Fabrication of an optical fiber sensor with a microsphere tip for CRP detection.
- Biofunctionalization of the sensor surface for specific CRP binding.
- Testing with standardized CRP solutions and clinical samples from patients with Urinary Tract Infection or Urosepsis.
- Application of 27 machine learning classifiers for data analysis and classification.
Main Results:
- The sensor demonstrated accurate detection of CRP across a wide concentration range (1.9 µg/L to 333 mg/L).
- Machine learning models, particularly ExtraTreesClassifier, achieved 95% accuracy in classifying phantom samples.
- The XGB classifier demonstrated up to 100% accuracy in classifying real patient samples, distinguishing inflammation states effectively.
Conclusions:
- The developed interferometric optical fiber sensor is a sensitive and accurate tool for CRP level determination.
- The sensor shows significant potential for rapid and reliable inflammation diagnostics in clinical settings.
- Integration with machine learning enhances the diagnostic capability for conditions like Urinary Tract Infection and Urosepsis.
