# Enhanced Visualization of Erythrocytes Through Photoluminescence Using NaYbF4:Yb3+,Er3+ Nanoparticles

## TL;DR

Researchers developed a method to visualize red blood cells using glowing nanoparticles, which could improve bioimaging techniques.

## Contribution

A validated protocol for using upconversion nanoparticles to image erythrocytes with minimal toxicity and automated analysis.

## Key findings

- Hexagonal-phase UCNPs with PAA coating localized at erythrocyte membranes without internalization.
- Optimal imaging conditions achieved with 0.01 mg/mL UCNPs and 250–300 mA laser current.
- A Random Forest model classified photoluminescent signals with 80% accuracy and 83% precision.

## Abstract

Rare-earth nanoparticles (RE-NPs), particularly NaYF4:Yb3+,Er3+, have emerged as a promising class of photoluminescent probes for bioimaging and sensing applications. These nanomaterials are characterized by their ability to absorb low-energy photons and emit higher-energy photons through an upconversion luminescence process. This process can be triggered by continuous-wave (CW) light excitation, providing a unique optical feature that is not exhibited by native biomolecules. However, the application of upconversion nanoparticles (UCNPs) in bioimaging requires systematic optimization to maximize the signal and ensure biological compatibility. In this work, we synthesized hexagonal-phase UCNPs (average diameter: 29 ± 3 nm) coated with polyacrylic acid (PAA) and established the optimal conditions for imaging human erythrocytes. The best results were obtained after a 4-h incubation in 100 mM HEPES buffer, using a nanoparticle concentration of 0.01 mg/mL and a laser current intensity of 250–300 mA. Under these conditions, the UCNPs exhibited minimal cytotoxicity and were found to predominantly localize at the erythrocyte membrane periphery, indicating surface adsorption rather than internalization. Additionally, a machine learning model (Random Forest) was implemented that classified the photoluminescent signal with 80% accuracy and 83% precision, with the signal intensity identified as the most relevant feature. This study establishes a quantitative and validated protocol that balances signal strength with cell integrity, enabling robust and automated image analysis.