• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Advances in optics for biotechnology, medicine and surgery XV >A BIOPHYSICAL RAMAN SPECTROSCOPIC MODEL FOR NONINVASIVE SCREENING OF SKIN CANCER
【24h】

A BIOPHYSICAL RAMAN SPECTROSCOPIC MODEL FOR NONINVASIVE SCREENING OF SKIN CANCER

机译：皮肤癌无创筛查的生物物理拉曼光谱模型

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Raman spectroscopy (RS) is sensitive to the molecular composition of biological tissues. Raman optical fiber-based probes have demonstrated efficacy in noninvasive cancer screening of the skin, breast, stomach, cervical, lung and brain. Currently, statistical algorithms such as principle component analysis (PCA) are the standard approaches for describing the spectral variance of the RS data and providing tissue classification. However, a PCA-based analysis does not allow for an examination of the biophysical basis of disease, such as microstructural organization of proteins and lipids. Understanding those biophysical parameters is essential to interpret the diagnostic result similar to that a pathologist is familiar reading, and develop diagnostic algorithms for fast and accurate cancer screening. Here, we proposed the first biophysical Raman model for in vivo human skin cancer screening. The basic principle is to describe the in vivo human skin spectra as a linear combination of the basis spectra of the most relevant skin constituents. The fit parameters of the skin constituents provided insights into the biochemical and structural makeup of the skin tissue and were then used to develop diagnostic models to discriminate skin cancers. We expanded upon a previous skin cancer model [1] by using in situ skin constituents as the building blocks instead of synthetic chemicals to better represent the skin microenvironment. To achieve this goal, we performed Raman imaging on human skin sections using a custom built 830nm confocal Raman microscope (Figure 1) and collected a library of basis spectra for individual skin constituents. We then validated our model using a previous clinical screening study [2] that covered a wide range of nonmelanoma and melanoma skin disease states. Our model revealed the most relevant building blocks to represent the human skin RS data: collagen, elastin, keratin, triolein, ceramide, cell nucleus, melanin and water. More importantly, we found collagen and triolein were the most significant biomarkers in discriminating malignant melanoma from benign pigmented lesions, and discriminating nonmelanoma skin cancer from normal skin. In the long term, our biophysical Raman model has the potential to impact the screening and margin assessment in other cancer types, such as breast, colorectal and prostate.
机译：拉曼光谱法（RS）对生物组织的分子组成敏感。基于拉曼光纤的探针已经证明可以有效地筛查皮肤，乳腺癌，胃癌，宫颈癌，肺癌和脑癌。当前，诸如主成分分析（PCA）之类的统计算法是描述RS数据的光谱方差并提供组织分类的标准方法。但是，基于PCA的分析无法检查疾病的生物物理基础，例如蛋白质和脂质的微结构组织。理解那些生物物理参数对于解释类似于病理学家熟悉的诊断结果并开发诊断算法以进行快速准确的癌症筛查至关重要。在这里，我们提出了第一个用于体内人类皮肤癌筛查的生物物理拉曼模型。基本原理是将体内人类皮肤光谱描述为最相关的皮肤成分的基础光谱的线性组合。皮肤成分的拟合参数提供了对皮肤组织的生化和结构组成的了解，然后被用于开发诊断模型以区分皮肤癌。我们扩展了以前的皮肤癌模型[1]，使用原位皮肤成分代替合成化学物质作为构建基块，以更好地代表皮肤微环境。为了实现此目标，我们使用定制的830nm共焦拉曼显微镜（图1）在人体皮肤切片上进行了拉曼成像，并收集了单个皮肤成分的基础光谱库。然后，我们使用先前的临床筛选研究[2]验证了我们的模型，该研究涵盖了广泛的非黑色素瘤和黑色素瘤皮肤疾病状态。我们的模型揭示了代表人类皮肤RS数据的最相关的构建块：胶原蛋白，弹性蛋白，角蛋白，三油精，神经酰胺，细胞核，黑色素和水。更重要的是，我们发现胶原蛋白和三油精是区分恶性黑色素瘤与良性色素病变和区分非黑色素瘤皮肤癌与正常皮肤的最重要生物标志物。从长远来看，我们的生物物理拉曼模型有可能影响其他癌症类型（如乳腺癌，结直肠癌和前列腺癌）的筛查和切缘评估。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号