Multispectral Imaging of Skin Oxygenation

This research focuses on the investigation of a multispectral imaging technique for quantitative measurement of skin tissue oxygen saturation (StO2). Oxygen saturation is the percentage of oxygen carried in the blood over the total oxygen carrying capacity of blood. Skin StO2 is an important indicator of skin tissue clinical status. Measurement of skin tissue oxygenation has been in many clinical applications such as cosmetic product development, chronic wound care and plastic surgery tissue recovery monitoring. For decades, researchers have been investigating optical spectroscopy and imaging techniques for measuring tissue oxygen saturation. However, many existing techniques are subjective and qualitative due to background bias, tissue heterogeneity, and inter-patient variation. A technique for quantitative and reliable measurement of skin tissue oxygen saturation is necessary. Multispectral / hyperspectral imaging is a promising technique which has been proposed for quantitative measurement of skin StO2 by capturing and analyzing skin images at multiple wavelengths.This dissertation reports a multispectral imaging method which is capable of (1) quantitative measurement of skin oxygen saturation with minimal bias from different skin types, and (2) imaging skin oxygen saturation dynamics. There’re mainly three parts of this research.First, theoretical study and algorithm development. A numerical model was established to simulate skin tissue reflectance of different skin conditions including oxygen saturation, blood concentration, tissue scattering and melanin concentration. From the simulation result, a parameter SDR, which can be calculated from skin reflectance images of 544, 552, 568, 576, 592, 600nm, was found to be mainly determined by oxygenation regardless of different tissue conditions. Thus, skin oxygen saturation can be measured by taking multispectral images of six wavelengths. Second, imaging system development. A multispectral / hyperspectral imaging system was built by integrating camera, filter and light source. A graphical user interface was developed for equipment control, synchronization and image acquisition. Imaging processing algorithm was developed for generation of skin oxygen saturation map from the reflectance image of six wavelengths. Third, benchtop experiments and human subject test. The imaging method was first verified using a phantom which was a liquid mixture of different concentrations of blood, intralipid and ink to simulate different skin conditions. StO2 measured by our method was not affected by the different concentration of phantom materials. Then, multispectral imaging of skin oxygenation was demonstrated by a human subject test in which 10 subjects with different skin types were recruited. The variation of skin StO2(%) measured by multispectral imaging is over two times smaller than the measurements from other two oximetry techniques, which indicates that skin StO2 measurement by our method is not affected by different skin types. After that, skin StO2 dynamics was measured using a postocclusive reactive hyperemia (PORH) protocol. The change of StO2 measured by our system was compatible with the oxygen measurements by other oximetry techniques. Finally, in order to test if our StO2 imaging method is affected by additional background absorption, measurement was taken on the skin on which a layer of ink was painted. The averaged skin StO2 measured by our system did not show a significant difference with the additional ink layer. The major innovations of this research include (1) the method of multispectral imaging of skin StO2 using the above six wavelengths is firstly reported; (2) it’s capable of imaging StO2 dynamics ; and (3) it has the potential for quantitative imaging of skin StO2 with minimal bias from different skin conditions. This research is a comprehensive investigation on a multispectral imaging method for quantitative measurement of skin oxygen saturation. It provides not only a new approach for spectroscopic imaging algorithm development but also a clinical tool for many applications.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Biophysics

Grantor dc:publisher

The Ohio State University

Year dc:date

2012

Author dc:creator

• Huang, Jiwei

Contributors dc:contributor

• Xu, Ronald

• Biomedical Engineering
• Biomedical Research
• Medical Imaging
• Optics

Statement dc:rights

• unrestricted
• This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.

Language dc:language

English

Repository record dc:identifier

http://rave.ohiolink.edu/etdc/view?acc_num=osu1356637098

OAI identifier oai:identifier

oai:etd.ohiolink.edu:osu1356637098