Related Experiment Video
Updated: Jul 14, 2026
06:27
Fast and Accurate Exhaled Breath Ammonia Measurement
Published on: June 11, 2014
13.4K
Noninvasive Total Cholesterol Level Measurement Using an E-Nose System and Machine Learning on Exhaled Breath Samples
Anna Paleczek1, Justyna Grochala2, Dominik Grochala1
1AGH University of Krakow, Faculty of Computer Science Electronics and Telecommunications, Institute of Electronics, al. A. Mickiewicza 30, Krakow 30-059, Poland.
ACS Sensors
|November 22, 2024
Summary
This study introduces a novel electronic-nose (e-nose) system using machine learning to noninvasively measure total cholesterol levels from breath samples. This breath analysis offers a promising alternative for monitoring cholesterol. Keywords: electronic-nose, machine learning, total cholesterol, noninvasive measurement, breath analysis.
Area of Science:
- Biomedical Engineering
- Analytical Chemistry
- Machine Learning Applications
Background:
- Current methods for measuring total cholesterol levels are invasive, often requiring blood draws.
- There is a need for noninvasive, convenient, and accurate methods for cholesterol monitoring.
- Exhaled breath contains volatile organic compounds that may correlate with physiological states.
Purpose of the Study:
- To propose and evaluate the first electronic-nose (e-nose) system coupled with a machine learning algorithm for noninvasive total cholesterol measurement.
- To assess the feasibility of using exhaled air samples for predicting cholesterol levels.
Main Methods:
- A cohort of 151 participants provided breath samples.
- Breath samples were analyzed using an e-nose equipped with various gas sensors (TGS1820, TGS2620, TGS2600, MQ3, Semeatech 7e4 NO2/H2S, SGX_NO2/H2S, K33, AL-03P/S).
- The Light Gradient Boosting Machine Regressor (LGBMRegressor) algorithm was employed to predict total cholesterol levels.
Main Results:
- Machine learning models achieved a Mean Absolute Percentage Error (MAPE) of 13.7% for the entire measurement range.
- For the normal cholesterol range (≤200 mg/dL), the MAPE was significantly reduced to 8%.
- The study demonstrates a correlation between exhaled breath composition and total cholesterol levels.
Conclusions:
- It is feasible to develop a noninvasive device for measuring total cholesterol levels using exhaled air.
- The proposed e-nose system combined with machine learning shows potential as a screening tool for cholesterol.
- Further research can refine the system for improved accuracy and clinical application.
