SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:

SPIE Photonics West 2019 | Register Today

SPIE Defense + Commercial Sensing 2019 | Call for Papers



Print PageEmail PageView PDF

Biomedical Optics & Medical Imaging

Using multiphoton microscopy to analyze aging skin

Applying advanced imaging and analytical techniques to measure patterns in the makeup of dermal tissue could help our understanding of how skin changes over time.
4 February 2011, SPIE Newsroom. DOI: 10.1117/2.1201012.003318

Skin undergoes structural degradation in response to both the passing of years and exposure to the sun.1–4 The degeneration of collagen, a component in the dermal extracellular matrix, is a major factor in skin alteration with aging. But the mechanism differs depending on the cause. Researchers have not yet establishing a quantitative relationship between collagen changes and different types of aging. Here, we describe work aimed at quantitatively linking collagen alteration with aging progression by multiphoton microscopy (MPM).

We have applied MPM5–7 to assess intrinsic-age-related and photo-age-related differences in skin in living organisms. This technique combines an Axiovert 200 microscope with a Zeiss LSM510 META (a 32-channel detector module) laser-scanning microscope and a Coherent Mira 900-F mode-locked femtosecond titanium:sapphire laser to detect a second-harmonic-generation (SHG) signal from collagen fibrils, which represents intrinsic phase matching in the medium. We used an excitation wavelength of 850nm and limited the average power to P≤10mW.8 As an animal model, we used 50 mice of different ages and 10 mice exhibiting photoaging and looked at the structural changes in their skin.

Figure 1. Second-harmonic-generation (SHG) intensity versus depth of skin. Error bars represents calculated standard deviations (n=40) in each mouse. n: Number of measurements. a.u.: Arbitrary units.

We began with a statistical analysis of collagen obtained by an oil objective with a Plan-Apochromat ×63 objective, numerical aperture (NA)=1.4. We found that the size of the collagen changes slightly with age and that the collagen in younger skin is denser than that in older skin. We also found that SHG intensity decreases with age. To obtain changes in collagen content, we quantitatively analyzed SHG intensity in different types of aging skin (see Figure 1). The depth of strongest SHG intensity in younger skin is deeper than that in older skin, as is total detectable depth.

Figure 2. Collagen structure from multiphoton microscopy (MPM) and Fourier analysis in skin from eight-week-old mice. (a) Visualization of collagen structure (SHG signals) at a depth of 54μm below the surface, the area of greatest SHG intensity. (b) Collagen-orientation-index map corresponding to (a). The collagen orientation index was calculated using an equation reported elsewhere.9(c) 3D representation of (b) in lengthwise orientation. The collagen bundle packing x was calculated by the distance between two peaks of the first-order maxima.

Next, we obtained a collagen orientation index in different types of aging skin structure using MPM with a Plan-Neofluar ×10 objective (NA=0.3) and analyzed it by fast Fourier transform (FFT) to calculate the orientation index of collagen bundles. Images provided by FFT generally indicate higher values along the direction orthogonal to that of the collagen fibers. A plot of the images describes orientation behavior in the form of ellipses. In our study, we used the short/long axis ratio of the power plot of the image—see Figure 2(b)—to estimate the collagen orientation index. We then calculated the actual index.9According to FFT analysis, in chronologically aged skin the shape of the ellipse progresses from short to long with age. We also found that collagen is aligned irregularly in younger skin and in a more parallel fashion in older skin. In photoaged skin, in contrast, collagen orientation shows a shorter ellipse than in younger skin.

Figure 3. Contrast with pixel distance for skin of different ages in vivo. (a) Normalized value of contrast with pixel distance. (b) Normalized contrast in a distance of 20 pixels.

Finally, we obtained collagen structures in different types of aging skin obtained by MPM with a Plan-Neofluar ×10 objective and analyzed them using the gray level co-occurrence matrix (GLCM).10 Spatial gray-level co-occurrence estimates image properties related to second-order statistics. We selected and analyzed three textural features, for example, contrast, correlation, and entropy. The pixel distance calculated ranges from 0 to 60 pixels (0–152.4μm) horizontally in each image. The textural features were analyzed using the GLCM module of ImageJ software.11 Contrast analysis (see Figure 3) showed that the value of younger skin increases sharply. Moreover, the value is higher than that of older skin for certain pixel distances, which suggests that the collagen texture of younger skin is distinct. For older skin, contrast changes only slightly as distance increases, implying that the collagen texture of older skin is obscure and the pixel matrix homogeneous, especially in the case of photoaged skin. Analysis of the collagen fibril correlation feature (see Figure 4) shows that the value falls off sharply with distance in chronologically aged skin, implying obscure texture and a larger matrix. The correlation value for older skin is higher than that for younger skin. Moreover, the entropy (a measure of disorder) of collagen fibrils rises with distance in both younger and older skin (see Figure 5). By the same token, the entropy value of younger skin is higher than that of older skin. Photoaged skin has the lowest entropy.

Figure 4. Normalized correlation values with pixel distance for variably aged skin in vivo.

Figure 5. Normalized entropy values with pixel distance for variably aged skin in vivo.

In summary, results of texture analysis suggest that in younger skin, collagen texture is distinct and the pixel matrix well differentiated. The collagen bundle has a characteristically fine texture. In contrast, in older skin, and particularly in photoaged skin, collagen texture is rougher, obscure, and uniform in the pixel matrix. These findings indicate that texture analysis by GLCM can be used as an aid in characterizing the structure of different types of aging skin, and consequently to further understand mechanisms of skin aging. We intend to apply the results of our findings to monitoring and improving treatment for aging skin in the near future.

Shulian Wu, Hui Li, Xiaoman Zhang, Zhifang Li, Shufei Xu
School of Physics and OptoElectronics Technology, Fujian Normal University
Fuzhou, China

Shulian Wu is a PhD candidate in optical engineering.

Hui Li is the corresponding author and a professor of optical engineering. His research interests include tissue optics, laser-tissue interactions, and photoacoustic tomography.

Xiaoman Zhang is a research assistant in optics.

Zhifang Li is a PhD candidate in optical engineering.

Shufei Xu is an MS candidate in optics.

4. S. Neerken, G. W. Lucassen, M. A. Bisschop, E. Lenderink, T. Nuijs, Characterization of age-related effects in human skin: a comparative study that applies confocal laser scanning microscopy and optical coherence tomography, J. Biomed. Opt. 9, no. 2pp. 274-281, 2004. doi:10.1117/1.1645795
11. http://rsbweb.nih.gov/ij/  Image Processing and Analysis in Java (ImageJ) download site. Accessed 12 December 2010.