|
Hokkaido University Collection of Scholarly and Academic Papers >
Theses >
博士 （総合化学） >
Study on Single Cell Raman Analysis to Enhance Differentiability of Cell Types in Non-homogeneous Environments
|
Files in This Item:
|
|
|
Related Items in HUSCAP:
|
| Title: | Study on Single Cell Raman Analysis to Enhance Differentiability of Cell Types in Non-homogeneous Environments |
| Other Titles: | 不均一環境下における細胞識別性向上に関する1細胞ラマン解析研究 |
| Authors: | Abdul, Halim Bhuiyan Browse this author |
| Issue Date: | 25-Sep-2023 |
| Publisher: | Hokkaido University |
| Abstract: | Raman imaging is a powerful technique used in biological sample measurement. It gives both spatial and spectral representation of the sample, that can be integrated with machine learning systems to develop new medical diagnosis tool. The Raman measurements were performed with an high-speed Raman microscope, the slit-scanning Raman microscope. It extracts the underlying spatial and spectral information of a sample typically two orders of magnitude faster than raster scanning. In this study, thyroid cell lines, FTC-133(cancerous) and Nthy-ori 3-1(normal) were used as a model to investigate the pertinence of Raman spectroscopy in the diagnosis of thyroid cancer. Line illumination Raman microscope ex tracts the underlying spatial and spectral information of a sample, typically, a few hundred times faster than raster scanning. This makes it possible to measure a wide range of bio logical samples such as cells and tissues – that only allow modest intensity illumination to prevent potential damage – within feasible time frame. However, a non-uniform intensity distribution of the laser line illumination may induce some artifacts in the data and lower the accuracy of machine learning models trained to predict sample class membership. Here, using cancerous, and normal human thyroid follicular epithelial cell lines, FTC-133 and Nthy-ori 3-1 lines, whose Raman spectral difference is not so large, I showed the standard preprocessing of spectral analyses widely used for raster scanning microscope introduced some artifacts. To address this issue, I proposed a detrending scheme based on random forest regression, a nonparametric model-free machine learning algorithm, combined with position-dependent wavenumber calibration scheme along illumination line. It was shown that the detrending scheme minimizes the artificial biases arising from non-uniform laser source and significantly enhances the differentiability of the sample states, i.e., cancerous or normal epithelial cells, compared to the standard preprocessing scheme. |
| Conffering University: | 北海道大学 |
| Degree Report Number: | 甲第15633号 |
| Degree Level: | 博士 |
| Degree Discipline: | 総合化学 |
| Examination Committee Members: | (主査) 教授 武次 徹也, 教授 小松﨑 民樹, 教授 伊藤 肇, 教授 髙橋 啓介 |
| Degree Affiliation: | 総合化学院（総合化学専攻） |
| Type: | theses (doctoral) |
| URI: | http://hdl.handle.net/2115/90792 |
| Appears in Collections: | 課程博士 (Doctorate by way of Advanced Course) > 総合化学院(Graduate School of Chemical Sciences and Engineering)
学位論文 (Theses) > 博士 （総合化学）
|
|
