Biochip has been created for screening hundreds of biomarkers in a single drop of blood

The key innovation of the system is the use of nanocavities— structures hundreds of times smaller than the width of a human hair. These structures function as traps for light, amplifying the fluorescent signals that occur when the target microRNA binds to the corresponding probe. This makes it possible to fix even single molecules of the genetic material without the need for their pre-amplification (amplification).

To process the data obtained, the scientists used the Mask R-CNN deep learning model. AI automatically analyzes microscopic images of the chip, classifies the types of microRNA, and separates the target signals from the background noise. During the tests, the system demonstrated an accuracy of target identification above 99%.

In the experiments, the platform successfully identified three types of microRNA (miR-191, miR-25, and miR-130a) associated with non-small cell lung cancer. The analysis was performed directly on human cancer cell extracts without complex sample preparation.

The main advantages of the new platform are:

• Speed: reducing the analysis time from several hours (PCR) to 20 minutes.
• Scalability: the ability to simultaneously screen hundreds and thousands of different biomarkers in a single sample.
• Portability: a prototype of a compact device that works in conjunction with a mobile app has been created.
• Versatility: the technology can be adapted for the diagnosis of a wide range of diseases, including viral infections and metabolic disorders.

The discovery of microRNAs and their role in gene regulation was awarded the Nobel Prize in Physiology or Medicine in 2024, highlighting the significance of NTU’s development. According to independent experts, the implementation of such non-invasive monitoring methods will enable the early detection of dangerous diseases, the rapid stratification of patient risks, and the real-time monitoring of therapeutic effectiveness.