3D microanatomical rendering and quantitative evaluation of a complete 50 μm-thick colon most cancers tissue slide. Credit score: Nature Communications (2025). DOI: 10.1038/s41467-025-59820-0
Transferring past conventional strategies of observing thinly sliced and stained most cancers tissues, a collaborative worldwide analysis workforce led by KAIST has efficiently developed a brand new expertise. This innovation makes use of superior optical methods mixed with a man-made intelligence-based deep studying algorithm to create practical, just about stained 3D pictures of most cancers tissue with out the necessity for excisional biopsy. This breakthrough is anticipated to pave the way in which for next-generation noninvasive pathological prognosis.
A analysis workforce led by Professor YongKeun Park of the Division of Physics, in collaboration with Professor Su-Jin Shin’s workforce at Yonsei College Gangnam Severance Hospital, Professor Tae Hyun Hwang’s workforce at Mayo Clinic, and Tomocube’s AI analysis workforce, has developed an modern expertise able to vividly displaying the 3D construction of most cancers tissues with out separate staining.
The analysis is revealed within the journal Nature Communications.
For over 200 years, standard pathology has relied on observing most cancers tissues below a microscope, a technique that solely reveals particular cross-sections of the 3D most cancers tissue. This has restricted the flexibility to know the three-dimensional connections and spatial preparations between cells.
To beat this, the analysis workforce utilized holotomography (HT), a sophisticated optical expertise, to measure the 3D refractive index info of tissues. They then built-in an AI-based deep studying algorithm to efficiently generate digital Hematoxylin & Eosin pictures—essentially the most broadly used staining technique for observing pathological tissues. Hematoxylin stains cell nuclei blue, and eosin stains cytoplasm pink.
Comparability of standard 3D tissue pathology process and the 3D digital H&E staining expertise proposed on this research. The normal technique requires getting ready and marking dozens of tissue slides, whereas the proposed expertise can cut back the variety of slides by as much as 10 instances and shortly generate H&E pictures with out the staining course of. Credit score: The Korea Superior Institute of Science and Know-how (KAIST)
The analysis workforce quantitatively demonstrated that the pictures generated by this expertise are extremely just like precise stained tissue pictures. Moreover, the expertise exhibited constant efficiency throughout varied organs and tissues, proving its versatility and reliability as a next-generation pathological evaluation instrument.
Furthermore, by validating the feasibility of this expertise via joint analysis with hospitals and analysis establishments in Korea and the USA, using Tomocube’s holotomography tools, the workforce demonstrated its potential for full-scale adoption in real-world pathological analysis settings.
Professor Park said, “This research is a very significant achievement that expands the unit of pathological analysis from 2D to 3D. It is expected to be widely utilized in various biomedical research and clinical diagnoses, such as analyzing the boundaries of cancer tumors and the spatial distribution of cells in the surrounding areas within the microtumor environment.”
Extra info:
Juyeon Park et al, Revealing 3D microanatomical buildings of unlabeled thick most cancers tissues utilizing holotomography and digital H&E staining, Nature Communications (2025). DOI: 10.1038/s41467-025-59820-0
Supplied by
The Korea Superior Institute of Science and Know-how (KAIST)
Quotation:
3D digital staining expertise allows non-invasive remark of most cancers tissue (2025, Could 26)
retrieved 27 Could 2025
from https://medicalxpress.com/information/2025-05-3d-virtual-technology-enables-invasive.html
This doc is topic to copyright. Other than any honest dealing for the aim of personal research or analysis, no
half could also be reproduced with out the written permission. The content material is offered for info functions solely.
