The research team led by senior researchers Yoonhee Lee from the division of biomedical technology and Gyogwon Koo by dividing intelligent robots into DGIST (under President Kunwoo Lee) has developed a technology that distinguishes the mutation of the lung gene with the “migrators” released from cancer in cancer cells in hematological cells of hematology-
Their study allows for a rapid and accurate analysis of individual IVFs and is expected to proceed with a new diagnostic biopsy technique for lung cancer.
Non -microcellular lung cancer (NSCLC) is the most common type of lung cancer, which represents more than 85% of all cases. However, because it rarely shows remarkable symptoms in the early stages, it is often diagnosed at an advanced stage, making treatment difficult.
NSCLC still has a high mortality rate and develops new diagnostic technologies that allow early detection and treatment remains a major challenge in the medical field. Specifically, conventional tissue biopsies give a significant burden on patients and have restrictions on repeated tests. Therefore, non -invasive liquid biopsy technology that uses blood -derived information has recently attracted attention.
The research team led by senior researchers Yoonhee Lee and Gyogwon Koo in isolated DGist extracurricular NSCLC cell lines with separate genetic mutations (A549: KRAS, PC9: EGFR, PC9/GR mutation: EGFR. Using the AFM, the group measured the physical properties of nano-garden of individual IVF in high resolution, including surface stiffness and ray proportions.
They found that extracurricularly derived from A549 cells showed significantly higher stiffness, showing that changes in cell membrane lipids caused by KRAS mutations were also reflected in extracurricular.
In contrast, extracurricularly derived from PC9 and PC9/Gr cells have shown similar properties, indicating a correlation with their common genetic background. These findings show that the physical properties of IVF vary according to the genetic mutations of the cancer cells from which they come from.
To accurately classify these nan -menechanical characteristics of the IVF, the research team used AI technology. The data of the height and stiffness of the IVFs received through the AFM were depicted and used to train a model with a deep learning nerve network (Densnet-121) to classify their cell series of origin.
Extravagant A549 cells were distinguished with notable high accuracy of 96%and the total average AUC reached 0.92. This demonstrates the possibility of a next generation biopsy platform capable of classifying high precision based solely on the physical properties of IVF, without the need to mark fluorescent.
Senior researchers Yoonhee Lee and Gyogwon Koo said: “This study presents a new diagnostic potential for lung cancer distinction with specific genetic mutations using only a small amount of IVF.
Source:
Magazine report:
Park, S., et al. (2025). The classification of extracellular vesicles derived from deep learning using nano-mechanical signatures of AFM. Analytical chemistry. doi.org/10.1021/acs.analchem.5c02009
