Parallel micro-Raman spectroscopy of multiple cells in a single acquisition using hierarchical sparsity

Area of Science:

• Biophotonics
• Spectroscopy
• Computational Biology

Background:

• Parallel micro-Raman spectroscopy offers high throughput for single-cell analysis.
• Accurate spectral retrieval from superimposed data is a significant challenge.

Purpose of the Study:

• To develop a novel method for accurate spectral reconstruction in parallel micro-Raman spectroscopy.
• To improve the analytical capacity and throughput of single-cell analysis.

Main Methods:

• Developed a hierarchical sparsity method within a compressive sensing framework.
• Implemented a group-selection strategy combined with in-group sparsity for spectral reconstruction.
• Validated the method using both simulated and experimental data.

Main Results:

• Successfully retrieved individual Raman spectra from superimposed data with high accuracy and low noise.
• Demonstrated the effectiveness of the group-selection mechanism in avoiding incorrect eigenspectra selection.
• Achieved orders of magnitude improvement in single-cell analysis throughput.

Conclusions:

• The developed hierarchical sparsity method significantly enhances parallel micro-Raman spectroscopy.
• This technique enables accurate, high-throughput analysis of multiple single cells.
• Expected to find wide applications in long-term monitoring of cellular biological processes.