Researchers in South Korea have developed a technology that allows for vivid 3D imaging of cancer tissues without the need for cutting or staining.
A research team led by Professor Park Yong-geun from the Korea Advanced Institute of Science and Technology (KAIST) has noted that, in collaboration with Professor Shin Soo-jin from Yonsei University Gangnam Severance Hospital, Professor Hwang Tae-hyun from the Mayo Clinic, and the artificial intelligence (AI) team from Tomocube, they successfully developed a technology that enables vivid viewing of the 3D structure of cancer tissues without requiring separate staining.
Until now, cancer tissues had to be sliced thinly and stained in order to see detailed structures through a microscope. This method only showed specific cross-sections of the cancer tissue, limiting the ability to understand the three-dimensional consolidation structures and spatial arrangements between cells.
The researchers successfully measured the 3D refractive index information of tissues using advanced optical technology known as holographic tomography (HT) and integrated AI-based deep learning algorithms to generate virtual stained images of cancer tissues. The refractive index refers to the degree to which the speed of light changes based on the material it passes through, and holographic tomography is a technique that measures changes in refractive index by illuminating cells or tissues from various directions.
Using this technology, the researchers demonstrated that the virtual stained images created are very similar to those of actual stained tissues and confirmed consistent performance across various organs and tissues.
Professor Park Yong-geun said, "This research is a very significant achievement in extending the analytical units of pathology from two-dimensional to three-dimensional." He added that it could be widely used in biomedical research and clinical diagnosis, including the analysis of cancer tumor boundaries and the spatial distribution of surrounding immune cells within a micro-tumor environment.
This study was published online in the international journal "Nature Communications" on the 22nd.
References
Nature Communications (2025), DOI: https://doi.org/10.1038/s41467-025-59820-0