Australian Research Council Centre of Excellence for Advanced Molecular Imaging, La Trobe University, Melbourne, Victoria, Australia.
Department of Chemistry and Physics, La Trobe Institute for Molecular Science (LIMS), La Trobe University, Melbourne, Victoria, Australia.
Nature. 2021 Oct;598(7879):65-71. doi: 10.1038/s41586-021-03835-2. Epub 2021 Oct 6.
The human eye can distinguish as many as 10,000 different colours but is far less sensitive to variations in intensity, meaning that colour is highly desirable when interpreting images. However, most biological samples are essentially transparent, and nearly invisible when viewed using a standard optical microscope. It is therefore highly desirable to be able to produce coloured images without needing to add any stains or dyes, which can alter the sample properties. Here we demonstrate that colorimetric histology images can be generated using full-sized plasmonically active microscope slides. These slides translate subtle changes in the dielectric constant into striking colour contrast when samples are placed upon them. We demonstrate the biomedical potential of this technique, which we term histoplasmonics, by distinguishing neoplastic cells from normal breast epithelium during the earliest stages of tumorigenesis in the mouse MMTV-PyMT mammary tumour model. We then apply this method to human diagnostic tissue and validate its utility in distinguishing normal epithelium, usual ductal hyperplasia, and early-stage breast cancer (ductal carcinoma in situ). The colorimetric output of the image pixels is compared to conventional histopathology. The results we report here support the hypothesis that histoplasmonics can be used as a novel alternative or adjunct to general staining. The widespread availability of this technique and its incorporation into standard laboratory workflows may prove transformative for applications extending well beyond tissue diagnostics. This work also highlights opportunities for improvements to digital pathology that have yet to be explored.
人类的眼睛可以分辨多达 10000 种不同的颜色,但对强度的变化却远不那么敏感,这意味着在解释图像时,颜色是非常重要的。然而,大多数生物样本本质上是透明的,用标准光学显微镜观察时几乎看不见。因此,非常希望能够在不需要添加任何染色剂或染料的情况下生成彩色图像,因为这些染色剂或染料可能会改变样本的性质。在这里,我们展示了可以使用全尺寸等离子体活动显微镜载玻片生成比色组织学图像。当样本放置在这些载玻片上时,这些载玻片将介电常数的细微变化转化为引人注目的颜色对比。我们通过在 MMTV-PyMT 乳腺肿瘤模型中肿瘤发生的早期阶段区分肿瘤细胞和正常乳腺上皮,证明了这种我们称为组织等离子体学的技术的生物医学潜力。然后,我们将这种方法应用于人类诊断组织,并验证其在区分正常上皮、常见导管增生和早期乳腺癌(导管原位癌)方面的效用。图像像素的比色输出与传统组织病理学进行了比较。我们在这里报告的结果支持这样一种假设,即组织等离子体学可以作为一般染色的一种新的替代或辅助方法。这项技术的广泛应用及其纳入标准实验室工作流程可能会对超出组织诊断应用的领域产生变革性的影响。这项工作还突出了数字病理学的改进机会,这些机会尚未得到探索。