Related Experiment Video
Updated: Aug 14, 2026
Mechano-Node-Pore Sensing: A Rapid, Label-Free Platform for Multi-Parameter Single-Cell Viscoelastic Measurements
Published on: December 2, 2022
Machine learning across label-free optical measurement platforms for cellular dynamics and biomechanics
Balint Beres1, Robert Horvath2
1Nanobiosensorics Laboratory, Institute of Engineering Physics and Materials Science of the HUN-REN Energy Research Centre, Konkoly-Thege Miklós út 29-33, Budapest H-1121, Hungary; Department of Automation and Applied Informatics, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Műegyetem Rkp. 3., 1111 Budapest, Hungary.
Label-free optical biosensing and machine learning offer advanced live-cell analysis. This review details single- and multimodal approaches for improved cell classification and understanding cellular behavior.
Area of Science:
- Biophysics
- Biotechnology
- Data Science
Background:
- Label-free optical biosensing provides high-resolution live-cell analysis.
- Machine learning enhances the interpretation of complex biological data.
Purpose of the Study:
- To review and organize workflows in label-free optical biosensing combined with machine learning.
- To examine single-modal and multimodal approaches for live-cell analysis.
- To highlight applications in cell-state classification and biomechanical variable estimation.
Main Methods:
- Categorization of works into single-modal and multimodal workflows.
- Analysis of representation-oriented and inference-oriented approaches in single-modal analysis.
- Examination of reference-based calibration and joint multimodal inference in multimodal workflows.
- Review of techniques including surface-enhanced Raman spectroscopy (SERS), surface plasmon resonance/resonant waveguide grating (SPR/RWG), and digital holographic microscopy (DHM).
Main Results:
- Single-modal approaches focus on signal enhancement or direct mapping of optical data to biological variables.
- Multimodal workflows fuse complementary data for robust cell state estimation.
- Specific techniques like SERS, SPR/RWG, and DHM are evaluated within these frameworks.
Conclusions:
- The combination of label-free optical biosensing and machine learning significantly advances live-cell analysis.
- Organized workflows, both single- and multimodal, are crucial for extracting meaningful biological insights.
- Future research can leverage these integrated approaches for deeper understanding of cellular dynamics.
