A NOVEL photonic crystal sensor for cervical cancer screening has demonstrated high sensitivity and more than 91% classification accuracy, offering a non-invasive approach to early detection across a range of temperatures.
Early identification of cervical malignancy remains central to improving outcomes, and optical technologies are increasingly being explored to enhance screening. This new study presents an optical sensor built using photonic crystals, engineered as a simple grid of microscopic silicon rods that guide light along a confined pathway. As light travels through cervical tissue samples, whether healthy, human papillomavirus infected, or cancerous, subtle shifts occur in its optical behaviour depending on tissue properties. These variations enable label free differentiation between tissue types, removing the need for dyes or contrast agents.
Photonic Crystal Sensor Performance in Cervical Cancer Screening
The sensor design allows precise detection of small optical changes generated by differences in tissue structure and composition. Investigators report that the device is highly sensitive and capable of identifying minute variations between normal, infected, and malignant tissue. Its relatively simple architecture may also support scalability and practical implementation in clinical environments.
A central innovation of the study was the evaluation of temperature effects on sensor performance. Previous optical sensing research has largely overlooked this factor, despite temperature influencing light tissue interactions.
Impact Of Temperature on Optical Accuracy
To assess robustness, the sensor was tested at four temperatures: 10 °C; 25 °C; 45 °C; and 60 °C. Performance was optimal at 25 °C, although reliable function was maintained across the full temperature range examined. These findings suggest the device could operate effectively in varied environmental or clinical settings.
Artificial Neural Network Enhances Classification
To strengthen tissue classification, researchers trained an artificial neural network using key optical features generated by the sensor. The model achieved accuracy exceeding 91%, supporting its potential role in automated or decision support systems for cervical cancer screening.
Collectively, the results indicate that this photonic crystal based optical sensor could provide a rapid, accurate, and non-invasive tool for early cervical cancer detection, with practical advantages for real world medical application.
Reference
Karami P et al. Enhanced design of a photonic crystal biosensor for early cervical cancer detection using refractive index variation and neural network classification. Scientific Reports. 2026; https://doi.org/10.1038/s41598-026-41676-z.
Featured image: Tom on Adobe Stock
