Intelligent Stain-Free Histology on Structural Colorimetric Nanocavities.

Area of Science:

• Biomedical Optics
• Materials Science
• Pathology

Background:

• Hematoxylin and eosin (H&E) staining is the standard in pathology but is costly and time-consuming.
• Nanophotonic chips offer stain-free imaging but are limited by high manufacturing costs.
• There is a need for cost-effective, stain-free histological imaging techniques.

Purpose of the Study:

• To develop and evaluate a novel stain-free histological imaging platform using planar nanocavities-on-silicon (NOS).
• To assess the efficacy of NOS slides in visualizing tissue morphology and aiding in cancer diagnosis.
• To demonstrate the potential of NOS for AI-assisted pathological analysis.

Main Methods:

• Fabrication of NOS slides by depositing a silicon nitride layer on silicon wafers.
• Utilizing optical interference within NOS slides to generate structural colors sensitive to tissue properties.
• Comparative imaging analysis of NOS and H&E-stained slides.
• Training a deep-learning model on NOS images for cancer detection.

Main Results:

• NOS slides provide vibrant structural colors for visualizing histological features without staining.
• Imaging results from NOS slides show good agreement with traditional H&E staining.
• A deep-learning model trained on NOS images achieved >95% accuracy in distinguishing cancerous from healthy colorectal tissues.

Conclusions:

• NOS slides offer a faster, inexpensive, chemical-free, and stain-free alternative for histological imaging.
• This platform simplifies tissue visualization and reduces costs.
• NOS holds promise for future AI-assisted screening and improved pathological diagnostics, particularly for colorectal cancer.