Related Experiment Video
Updated: Jul 12, 2026
Mechano-Node-Pore Sensing: A Rapid, Label-Free Platform for Multi-Parameter Single-Cell Viscoelastic Measurements
Published on: December 2, 2022
Light-regulated reconfigurable MXene-hydrogel gas sensing system via machine learning
Xuanjie Xia1, Hao Yuan1, Bin Wang1
1Department of Chemical Engineering, Tsinghua University, Beijing 100084, China; State Key Laboratory of Green Biomanufacturing, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China; Key Laboratory of Industrial Biocatalysis, Ministry of Education, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China.
This study introduces a novel light-regulated gas sensor using MXene-hydrogel composites and machine learning. The system achieves highly accurate gas detection and classification, even distinguishing cancer patients from healthy individuals via breath analysis.
Area of Science:
- Materials Science
- Sensor Technology
- Biomedical Engineering
Background:
- Gas sensing is crucial for medical diagnostics, industrial safety, and environmental monitoring.
- Conventional sensors struggle with selectivity and sensitivity, particularly at low gas concentrations.
- Developing advanced gas sensing systems is essential for improved detection capabilities.
Purpose of the Study:
- To develop a light-regulated gas-sensing system utilizing MXene-hydrogel composites and machine learning.
- To achieve a low detection limit and high-accuracy classification of various gas molecules.
- To explore the system's potential for breath-based medical diagnostics, such as cancer screening.
Main Methods:
- Fabrication of MXene-hydrogel composites with PNIPAM hydrogel for reconfigurable and adsorptive properties.
- Integration of MXene for electrical conductivity and photothermal conversion.
- Application of near-infrared light modulation to enhance selectivity and reduce response/recovery times.
- Implementation of machine learning classification algorithms for data analysis and pattern recognition.
Main Results:
- The developed sensing system demonstrated a low detection limit of 5 ppb for gas molecules.
- High-accuracy classification of ten different gas molecules was achieved with 98.64% accuracy.
- The system successfully distinguished between cancer patients and healthy controls using breath samples with 97.3% accuracy.
Conclusions:
- The combination of light-regulated sensing materials and machine learning offers a powerful approach for gas identification.
- This technology shows significant promise for developing compact, high-performance gas sensors.
- The system's efficacy in breath analysis highlights its potential for non-invasive medical diagnostics and early disease screening.
