Related Experiment Video
Updated: Jul 15, 2025
08:58
Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning
Published on: November 19, 2018
12.5K
Classification of cancer cells at the sub-cellular level by phonon microscopy using deep learning
Fernando Pérez-Cota1, Giovanna Martínez-Arellano2, Salvatore La Cavera3
1Optics and Photonics Group, Faculty of Engineering, University of Nottingham, Nottingham, UK. fernando.perez-cota@nottingham.ac.uk.
Scientific Reports
|September 27, 2023
Summary
This study uses phonon acoustics and deep learning to accurately measure cell elasticity, differentiating cancerous from normal breast cells with 93% accuracy for potential clinical tools.
Area of Science:
- Biophysics
- Cell Mechanics
- Acoustic Microscopy
Background:
- Cell elasticity correlates with normal, dysplastic, and cancerous states.
- Clinical applications of cell mechanics lack significant progress.
- Phonon acoustics offer a novel approach to measure cell elasticity.
Purpose of the Study:
- Explore phonon acoustics for measuring cell elasticity.
- Differentiate between cancerous (MDA-MB-231) and normal (MCF10a) breast cell lines.
- Develop a compact sensor for clinical integration.
Main Methods:
- Utilized phonon microscopy to measure elastic properties of breast cells.
- Applied deep learning to time-resolved phonon-derived data.
- Investigated classification using a physical model.
Main Results:
- Achieved 93% accuracy in differentiating cell lines using single phonon measurements.
- Demonstrated classification based on physical models suggesting new mechanical markers.
- Developed a proof-of-principle compact sensor design.
Conclusions:
- Phonon acoustics combined with deep learning can accurately assess cell elasticity.
- This technology shows promise for early cancer detection and diagnosis.
- The compact sensor design is compatible with needles and endoscopes for in vivo applications.
