Open Access
13 December 2023 Broadband scattering properties of articular cartilage zones and their relationship with the heterogenous structure of articular cartilage extracellular matrix
Iman Kafian-Attari, Ervin Nippolainen, Florian Bergmann, Akuroma George, Petri Paakkari, Arash Mirhashemi, Florian Foschum, Alwin Kienle, Juha Töyräs, Isaac O. Afara
Author Affiliations +
Abstract

Significance

Articular cartilage exhibits a zonal architecture, comprising three distinct zones: superficial, middle, and deep. Collagen fibers, being the main solid constituent of articular cartilage, exhibit unique angular and size distribution in articular cartilage zones. There is a gap in knowledge on how the unique properties of collagen fibers across articular cartilage zones affect the scattering properties of the tissue.

Aim

This study hypothesizes that the structural properties of articular cartilage zones affect its scattering parameters. We provide scattering coefficient and scattering anisotropy factor of articular cartilage zones in the spectral band of 400 to 1400 nm. We enumerate the differences and similarities of the scattering properties of articular cartilage zones and provide reasoning for these observations.

Approach

We utilized collimated transmittance and integrating sphere measurements to estimate the scattering coefficients of bovine articular cartilage zones and bulk tissue. We used the relationship between the scattering coefficients to estimate the scattering anisotropy factor. Polarized light microscopy was applied to estimate the depth-wise angular distribution of collagen fibers in bovine articular cartilage.

Results

We report that the Rayleigh scatterers contribution to the scattering coefficients, the intensity of the light scattered by the Rayleigh and Mie scatterers, and the angular distribution of collagen fibers across tissue depth are the key parameters that affect the scattering properties of articular cartilage zones and bulk tissue. Our results indicate that in the short visible region, the superficial and middle zones of articular cartilage affect the scattering properties of the tissue, whereas in the far visible and near-infrared regions, the articular cartilage deep zone determines articular cartilage scattering properties.

Conclusion

This study provides scattering properties of articular cartilage zones. Such findings support future research to utilize optical simulation to estimate the penetration depth, depth-origin, and pathlength of light in articular cartilage for optical diagnosis of the tissue.

1.

Introduction

The behavior of light in biological tissues is governed by intrinsic fundamental and wavelength-dependent properties (optical properties), including absorption coefficient (μa), scattering coefficient (μs), extinction coefficient (μt=μs+μa), scattering anisotropy factor (g), reduced scattering coefficient (μs), and refractive index (n).13 The key factors determining the optical properties of biological tissues are chemical composition, concentration, size, distribution, and alignment of the tissue matrix’s biomolecules.3 Since these vary between biological tissues and within the same tissue due to biological or pathological processes, this variation often leads to differences in optical properties between healthy and pathological tissues. Thus, the optical properties of biological tissues have been adopted as biomarkers for the non-invasive screening of tissue pathologies in multiple organs, including muscles, brain, skin, heart, and breast.1

In musculoskeletal research, particularly relating to osteoarthritis, there is growing research interest in applying diffuse optical technologies for the diagnostic assessment of musculoskeletal tissues, particularly with a focus on early detection and characterization of degenerative diseases, such as osteoarthritis.49 One of the primary tissues of focus in musculoskeletal research is articular cartilage, the highly specialized connective tissue covering the ends of bones in articulating joints. Articular cartilage enables joint lubrication and transmission of load to the underlying bone, thus allowing smooth joint motion. The function of articular cartilage is inherently related to the structure and composition of its extracellular matrix. Articular cartilage comprises a biphasic matrix and chondrocytes—the sole cellular component of the tissue embedded within the matrix. Articular cartilage solid matrix mainly comprises of collagen fibers, proteoglycan macromolecules, and water.10 The size and orientation of collagen fibers in articular cartilage is a function of tissue depth, varying from the articular surface to the subchondral bone, resulting in three distinct zones, namely: superficial zone, middle zone, and deep zone.11 The collagen fibers are oriented parallel to the articular surface in superficial zone, exhibit a skewed orientation in middle zone, and have a perpendicular orientation in deep zone.10 Furthermore, there is a mineralized layer of calcified cartilage at the interface between cartilage matrix and subchondral bone.12 The unique alignment of collagen fibers and the high osmotic pressure created by proteoglycans equip cartilage matrix with unique biomechanical properties to withstand various mechanical stresses.

In the past two decades, various studies have reported the μa and μs of articular cartilage and assessed their potential for characterizing the tissue.1322 However, little attention is given to how articular cartilage zonal architecture affects light propagation through the tissue matrix. In particular, the scattering properties of articular cartilage zones and their impact on the overall optical properties and response of the bulk tissue are not well understood. Few studies in the literature have investigated μs and g values of articular cartilage bulk tissue and zones.13,23,24 Beek et al.13 reported g of bulk rabbit cartilage tissue at 632.8 nm using a double-integrating sphere and the inverse adding doubling method.25 Shyu et al.23 reported the effective g of bulk porcine articular cartilage estimated by optical coherence tomography. Jambor et al.24 utilized goniometric setup to estimate the μs and g of articular cartilage superficial, middle, and deep zones at 445 and 890 nm and showed how enzymatic degradation such as depletion of proteoglycan and disruption of the collagen mesh will affects these parameters.

The present study aims to provide the spectral values of μs and g of healthy articular cartilage zones from different anatomical origins within the bovine knee. We hypothesized the unique compositional and structural properties of articular cartilage zones affects the scattering properties of the tissue and this influence is represented in the scattering properties of articular cartilage zones. We carried out a series of optical spectroscopic and imaging measurements to validate this hypothesis.

2.

Materials and Methods

2.1.

Sample Preparation

In this study, intact bovine knee joints (n=15) collected from a local abattoir were used; thus, no ethical permission was required. In the collection process, the joints were selected from skeletally mature animals and no cofounding factors, such as age and sex, were controlled. The joints were obtained within 1 week of slaughter and kept at 4°C before sample extraction. Two sets of adjacent osteochondral samples were harvested from the lateral and medial sides of the femur, tibia, and patella of the joints. Samples in both sets (sets A and B) were extracted using a diameter punch (d=15  mm). After the osteochondral plug extraction, the bone end of each sample was filed until it was parallel to its surface. During the process, the cartilage surface of the plugs was continuously rinsed with phosphate-buffered saline solution (PBS, pH 7.4, containing inhibitors) to minimize deterioration, such as interstitial water evaporation.

Subsequently, the cartilage layer of samples in set A was extracted and subjected to integrating sphere measurement to determine the broadband μa and μs and polarized light microscopy imaging to estimate the collagen fiber orientation. For samples in set B, cartilage sections (thickness=100  μm) were extracted from the different depth-wise zones and subjected to collimated transmittance measurements to estimate their broadband μs. Table 1 describes the number of samples collected in each set and their anatomical origin.

Table 1

Details of the samples collected in sets A and B. FL, lateral femoral group; FM, medial femur group; PL, lateral patella group; PM, medial patella group; TL, lateral tibia plateau group; and TM, medial tibia plateau. SZ, superficial zone; MZ, medial zone; and DZ, deep zone of articular cartilage, respectively. IS, integrating sphere measurements; PLM, polarized light microscopy imaging; and CT, collimated transmittance measurements.

Measurement typeSample setAnatomical originTotal No.
FLFMPLPMTLTM
ISSet A141412119868
PLMSet A69234327
CTSet BSZ44524221
MZ56625428
DZ67726331

2.2.

microCT Imaging

To estimate the thickness and surface diameter of the articular cartilage samples, set A was subjected to microCT imaging immediately after extraction. The samples were imaged with a microCT scanner (XT H 225, Nikon Metrology, Leuven, Belgium). The images were acquired with 40×40×40  μm3 isotropic voxel size, but the voxel size was increased to 50×50×50  μm3 when reconstructions were calculated. The tube voltage was set to 80 kVp with a 1.0 mm aluminum filter.

After image acquisition and reconstruction, the thickness and surface diameter of the articular cartilage part of the osteochondral samples were estimated via processing the microCT images. In summary, the 3D osteochondral microCT images were segmented to obtain the volume of the articular cartilage segment. Considering that the microCT image histogram has three peaks, it can be readily segmented into three classes using histogram thresholding. The classes are bone, soft tissue (articular cartilage and moisture), and background (including air and the sample holder). Distinguishing moisture from articular cartilage is not possible using their intensity values alone. Therefore, the segmentation was done using morphological operators and surface normal vectors to geometrically isolate the articular cartilage tissue from the moisture.26 Figure 1 depicts the processing of the microCT images of the osteochondral samples. After microCT imaging, the articular cartilage portion of the osteochondral samples was mechanically detached from sunchondral bone using a scalpel, and then samples were stored in PBS at −20.

Fig. 1

(a) The schematics and image of the integrating sphere setup used for estimation of μa and μs of articular cartilage bulk tissue. (b) The schematic and image of the collimated transmittance setup used for estimation of μt of articular cartilage zones. (c) The distribution of the physical properties (thickness (mm), surface diameter (mm), and volume (mm3) of the articular cartilage bulk samples, estimated from the microCT images. (c) (right) The top surface of an articular cartilage sample after segmentation and the X-Z cross-section of the sample, which is segmented to articular cartilage (cyan), bone (yellow), and background medium (black). The error of the articular cartilage volume segmentation from the microCT images was <18%.

JBO_28_12_125003_f001.png

2.3.

Cartilage Cryosection Procedure

Samples in set B were further processed to extract tissue sections from the different cartilage zones. The bone end of each sample was mounted on a disc using a fixative solution (Optimal Cutting Temperature compound, Thermo Fisher Scientific Ltd., Runcorn, United Kingdom). The assembly was placed in a cryostat chamber (Leica CM3050 S, Leica Biosystems, Wetzlar, Germany) for 5 min at 18 to allow sample freezing. Afterward, the assembly was placed on the rotary microtome within the cryostat chamber. Before sectioning, the first 50  μm of the articular surface was cut and discarded to provide a uniformly smooth and flat surface for sectioning. A series of sections with a thickness of 100  μm was then consequently obtained. The sections were hydrated with PBS and sandwiched between a glass slide (thickness = 1 mm, Menzel-Gläser Frosted Microscope Slides, ThermoFisher Scientific, Oy, Finland) and a coverslip (thickness = 0.13 mm, Menzel Microscope Coverslips, ThermoFisher Scientific, Oy, Finland) and then stored in a humid box at −20 for collimated transmittance measurement. The selection of the sections to present the zonal structure of articular cartilage was carried out as follows: (I) the first section was always considered superficial zone, (II) the subsequent section was considered to represent middle zone, and (III) the deep zone layer was selected after the first 500  μm of the tissue.

2.4.

Integrating Sphere Measurements

The optical setup used in this study comprises an optical apparatus to measure the optical response of biological tissues (reflectance and transmittance) and an optical simulation software to estimate the optical properties (μa and μs) from the optical response of tissues. The optical apparatus is an optimized double-beam integrating sphere setup consisting of a halogen light source (Halostar Starlite, OSRAM, Germany), a visible-band spectrometer with 3-nm resolution (Maya2000Pro, Ocean Optics, USA), a near-infrared (NIR)-band spectrometer with 7-nm resolution (NIRQuest512-1.7, Ocean Optics, United States), an in-house 3D-printed integrating sphere, and a glass sample holder construct [Fig. 1(a)]. Technical details of the optical setup, including its calibration and validation of the measured optical properties, are reported elsewhere.22,26,27

To estimate the optical properties of articular cartilage, the spectral band of 400 to 1400 nm was used as μa of water is low and allows the detection of the spectral features of other chromophores in articular cartilage. In addition, given the higher values of μs in this band and high values of g, a higher penetration depth can be achieved that facilitates the extraction of biologically relevant features from deep within the tissue. Moreover, in the absence of spectral values of n and g for bulk articular cartilage, fixed values of 1.358 and 0.9 from the literature were used for n and g.28,29 The in-house optical simulation software was used to implement a 3D Monte Carlo Solver and Henyey–Greenstein scattering phase function26 to simulate the angular distribution of scattered photons. The optical simulation software considered the samples to be perfect cylinders. Hence, the lateral radius and axial thickness of articular cartilage samples from Set A, estimated using microCT imaging, were applied to build the cylindrical geometry of the sample in the software.

Before optical measurement, the samples from set A were thawed to room temperature for 30 min. Subsequently, they were placed within a cylindrical sample holder of glass cuvettes to reduce the refractive index mismatch between the tissue and the surrounding medium [Fig. 1(a)]. To estimate the values of μa and μs, we hypothesized that during the integrating sphere measurements, the surrounding medium, filling the gap between articular cartilage and the inner walls of the sample holder, is air to account for any potential morphological irregularities of the samples.

2.5.

Collimated Transmittance Measurements

The articular cartilage sections extracted from set B samples were subjected to collimated transmittance measurements. The optical setup consists of (I) a halogen lamp (HL-2000, Ocean Insight, United States) as the light source; (II) a spectrometer (Maya 2000 Pro, Ocean Insight, United States); (III) two lenses with regular achromats of f-60 mm and a diameter of 25.4 mm (LINOS Photonics Inc., United States); (IV) an illumination fiber with 100-μm core diameter; (V) detection fiber with a 200-μm core diameter; (VI) a diffusor disc; and (VII) an optical fiber with 1000-μm core diameter, connecting the diffusor disc to the spectrometer. The collimated transmittance measurements were carried out in the spectral band of 400 to 1400 nm. The optical resolution of the used spectrometer in the spectral, with an entrance slit of 100  μm, is around 3 nm. Figure 1(b) shows the schematics and images of the collimated transmittance setup.

The extinvtion coefficient, μt(mm1), of set B samples were estimated by applying Beer–Lambert’s Law3 on the collimated transmittance dataset. In particular, μt was obtained as

Eq. (1)

μt=log(TT0)d,
where T(=TsampleTDark) is the transmittance signal of the samples subtracted by the dark current noise of the system (TDark) and normalized by the detector dynamic range (T0=TReferenceTDark). TReference is the transmittance measurement when no sample is placed in the system. Finally, d(mm) is the thickness of the samples. Table 1 depicts the number of samples per anatomical location and zonal layer for collimated transmittance measurements.

2.6.

Polarized Light Microscopy

This study subjected the set A samples to histological processing to prepare thin axial tissue sections for acquiring information about the orientation of collagen fibers in articular cartilage using polarized light microscopy imaging.30 After the integrating sphere measurements, the specimens were placed into a fixative solution containing formaldehyde (4%, Merck, Darmstadt, Germany) and ethylenediaminetetraacetic acid (EDTA, 10%, Merck, Darmstadt, Germany), and were kept at room temperature for 21 days. Following fixation, the process continued by dehydrating the specimens with a series of graded ethanol-based solutions and embedding the dehydrated specimens in paraffin for sectioning. The embedded specimens were halved from sagittal plane, and one set of three unstained deparaffinized 5-μm axial sections was obtained from each half.

Fixed tissue sections obtained from set A samples were subjected to polarized light microscopic imaging,30,31 using an apparatus equipped with a microscope body (Leitz Ortholux II POL, Leitz Wetzlar, Germany), a monochromatic light source (λ=630±30  nm, Edmund Optics Inc., Barrington, New Jersey, United States), crossed polarizers (Techspec optics® XP42-200, Edmund Optics, Barrington, New Jersey, United States), and a monochrome camera (pixel size 3.5  μm, BFS-U3-88S6M-C FLIR Blackfly® S, FLIR Systems Inc., United States) with a 2.5× magnification objective. The setup consists of a sample placed between a polarizer and an analyzer at a 90 deg angle. Samples were measured at 21 different orientation angles in the band of (0 deg, 180 deg) with a 9 deg step size.

The collagen network of articular cartilage induces birefringence.32 Thus, by measuring the intensity of the polarized light at several angles, a parallelism index can be defined as a measure of collagen fiber orientation using Michelson contrast and the Stokes parameters (S0, S1, and S2). A detailed description of the algorithm for estimating the collagen fiber orientation is reported elsewhere.31

Table 1 depicts the number of samples per anatomical location used for polarized light microscopy imaging. Six measurements were carried out for each sample and the results were averaged to provide a depth-wise orientation of collagen fibers for the samples. On average, the superficial zone of the bovine cartilage sample used in this study was 8% of the total thickness, the middle zone was 21% of the total thickness, and the deep zone was 71% of the total thickness.

2.7.

Estimation of Scattering Parameters

When photons undergo a single scattering event in the tissue under investigation, the Mie–Rayleigh formula empirically describes μs of biological tissues as a mixture of photon scattering by the Rayleigh and Mie scatterers, weighted by their normalized contribution fraction. The intensity of Rayleigh-scattered photons varies proportionally to λ4, whereas that of Mie-scattered photons varies proportionally to λb, where b is a parameter related to the size of the Mie scatterers and λ is the wavelength of investigation.3 Hence, the Mie–Rayleigh formula describes μs as

Eq. (2)

μs=α[c×(λλ0)4+(1c)×(λλ0)b],
where α (= μs(λ0), mm1) is a parameter related to the density of the scatterer particles in the tissue, λ0 (= 500 nm) is a reference wavelength used for nondimensionalization, and c is the normalized contribution fraction of the Rayleigh scatterers to μs. Similarly, when photons undergo multiple scattering events in the tissue, μs can be described by the Mie–Rayleigh formula, hence

Eq. (3)

μs=α*[c*×(λλ0)4+(1c*)×(λλ0)b*].
α* (= μs(λ0), mm1), c*, and b* are the scatterer density parameter, normalized contribution fraction of Rayleigh scatterers to μs, and the size parameter of Mie scatterers, respectively.

For both cases of μs and μs, the Mie–Rayleigh parameters are estimated by fitting the associated Mie–Rayleigh equations to the experimental values of μt and μs, respectively. As articular cartilage has low μa, thus μt is assumed to be mainly determined by μs (μtμs). After removing outliers (n=2), the mean R2 and RMSE scores of Mie–Rayleigh fit for μs are 96.84% and 1.1576. The mean R2 and RMSE scores of Mie–Rayleigh fit for μs are 98.83% and 0.0909.

The scattering anisotropy factor (g) is defined as

Eq. (4)

g=1μsμs.

Note that g is dependent on the scattering phase function used for estimating μs. Given that the Henyey–Greenstein scattering phase function is used in this study to estimate μs, the estimated broadband values of g are representative of Mie type of scatterers that are best described by the Henyey–Greenstein phase function (g=gHG). Nevertheless, the estimated values of gHG can be transferred to another set of values (gmHG) that can account for the contribution of Rayleigh scatterers to the photon scattering in articular cartilage as follows:

Eq. (5)

gmHG=gHG(1c).
Where c is the normalized volume fraction of the Rayleigh scatterers. Equation (5), proposed by Graaff et al.,33 estimates the impact of Rayleigh scatterers on g of biological tissues.

Similar to the Mie–Rayleigh formula, the Mie-collagen formula describes the scattering properties of biological tissues as scattering due to Mie scatterers and small cylindrical particles with nm-scale diameter. The Mie-collagen power law for μs and μs can be described as

Eq. (6)

μs=α**[c**×(λλ0)3+(1c**)×(λλ0)b**],

Eq. (7)

μs=α***[c***×(λλ0)3+(1c***)×(λλ0)b***],
where α** (= μs(λ0), mm1) and α*** (= μs(λ0), mm1) are the scatterer density parameters; b** and b*** are the Mie scatterer size parameters; and c** and c*** are the contribution fraction of collagen scatterers to μs and μs, respectively.22 This empirical formula was suggested by Kienle et al.,22 as they observed through their research that the three-order power function fits the scattering coefficients better than the four-order power function (Rayleigh scattering) for biological tissues that are predominantly composed of collagen fibers. The results of data analysis on Mie-collagen scattering formula are presented in the Supplementary Material. After removing outliers (n=2), the mean R2 and RMSE scores of Mie-collagen fit for μs are 96.8% and 1.1598. The mean R2 and RMSE scores of Mie-collagen fit for μs are 98.84% and 0.0906.

2.8.

Statistical Analysis

Statistical tests were conducted to investigate if the difference in the Mie–Rayleigh parameters per articular cartilage zone and anatomical location are significant.

  • 1. Test 1: statistical difference of Mie–Rayleigh parameters between superficial, middle, and deep zones, and bulk articular cartilage across all anatomical locations.

  • 2. Test 2: statistical difference of Mie–Rayleigh parameters betweem superficial, middle, and deep zones and bulk tarticular cartilage within each anatomical location.

  • 3. Test 3: statistical difference of Mie–Rayleigh parameters between samples obtained from different anatomical locations across all articular cartilage zones.

  • 4. Test 4: statistical difference of Mie–Rayleigh parameters between samples obtained from different anatomical locations per articular cartilage zone.

The statistical test comprises the Kolmogorov–Smirnov normality test, Leven’s test for equality of variances, one-way ANOVA and Tukey’s tests to examine the difference in normally-distributed observations, and Kruskal-Wallis and Dunn’s tests for nonparameteric observations. All the computational and statistical analyses required for the present report were carried out in MATLAB (R2019b and R2020b) and Python v3.7 using standard libraries.

3.

Results

The values of μa of bulk articular cartilage suggest that the tissue possesses low light absorption strength with apparent water absorption peaks at 950 and 1150 nm [Fig. 2(a)]. On average, the decrease of values of bulk articular cartilage μs shows the presence of Rayleigh scatterers in the tissue and their contribution to μs [Fig. 2(b)]. The similarity of bulk articular cartilage μa and μs over different anatomical locations suggest anatomical variation has minimal impact on the bulk optical properties [Figs. 2(a) and 2(b)]. The values of μt of articular cartilage zones are predominantly influenced by the tissue’s μs, with minimal impact of μa, including the water absorption peaks at 950 and 1150 nm (Fig. 3). Mean values of μs and μt exhibited higher values in deep zone than middle zone and superficial zones over the anatomical sites of medial femur and patella, and lateral and medial tibia. In addition, similar to μs, the decrease of μs of articular cartilage zones suggests the presence of Rayleigh scattering and its contribution to μs [Fig. 3(b)].

Fig. 2

Absorption coefficient (a) (μa, mm1) and reduced scattering coefficient (b) (μs, mm1) of bulk articular cartilage tissue. FL and FM are the lateral and medial femur sites, PL and PM are the lateral and medial patella sites, and TL and TM are the lateral and medial tibia sites. The values represent mean ± standard deviation.

JBO_28_12_125003_f002.png

Fig. 3

The extinction coefficient (a) (μt, mm1) and scattering coefficient (b) (μs, mm1) of articular cartilage zones of different anatomical locations. μt was experimentally obtained from collimated transmittance measurements while μs was obtained by fitting the Mie–Rayleigh formula to the experimental μt. SZ, MZ, and DZ are the superficial, middle, and deep zones of articular cartilage, respectively. FL and FM are the lateral and medial femur sites, PL and PM are the lateral and medial patella sites, and TL and TM are the lateral and medial tibia sites. The values represent mean ± standard deviation.

JBO_28_12_125003_f003.png

When the Mie–Rayleigh parameters of μs of cartilage zones are considered [Fig. 4(a)], on average and across all anatomical locations, the scatterer density parameter (α) increases consistently with tissue depth. That is, α is highest in deep zone and also higher in middle zone than in superficial zone (Table 2). The scatterer size parameter (b) shows a similar trend to α when averaged across all anatomical locations. However, the contribution of Rayleigh scatterer to μs (c) is highest in middle zone with a mean value of 10.65% across all anatomical locations (Table 2). When the Mie–Rayleigh parameters of bulk articular cartilage μs are considered [Fig. 4(b)], on average, the scatterer density parameter (α*) is higher in the lateral groups (lateral femur, tibia, and patella) than in the medial groups (medial femur, tibia, and patella). The scatterer size parameter (b*) shows a similar trend in the lateral and medial groups. However, when the contribution of Rayleigh scatterer to μs (c*) is considered, the medial groups possess larger values of c* than the lateral groups (Table 2).

Fig. 4

(a) The Mie–Rayleigh parameters obtained from fitting the μs of articular cartilage zones to Mie–Rayleigh formula over different anatomical location. (b) The Mie–Rayleigh parameters obtained from fitting the μs of bulk articular cartilage tissue to Mie–Rayleigh formula over different anatomical location. SZ, MZ, and DZ are the superficial, middle, and deep zones of articular cartilage, respectively. FL and FM are the lateral and medial femur sites, PL and PM are the lateral and medial patella sites, and TL and TM are the lateral and medial tibia sites. α and α* are the scatterer-density parameter of μs and μs; b and b* are the scatterer-size parameter of μs and μs; and c and c* are the normalized contribution of Rayleigh scatterers to μs and μs, respectively.

JBO_28_12_125003_f004.png

Table 2

The mean values of the scatterer-density parameter (α and α*), the scatterer-diameter parameter (b and b*) and normalized contribution of Rayleigh scatterers (c and c*) to the scattering coefficients of articular cartilage zones and bulk tissue across different anatomical location. μs and μs′ are the single and reduced scattering coefficients. SZ, MZ, and DZ are the superficial, middle, and deep zones of articular cartilage, respectively. FL and FM are the lateral and medial femur sites; PL and PM are the lateral and medial patella sites; and TL and TM are the lateral and medial tibia sites. The Mie–Rayleigh parameters were obtained by fitting Eqs. (2) and (3) to μs and μs′ values, respectively. The mean R2 and RMSE scores of Mie–Rayleigh fit for μs are 96.84% and 1.1576. The mean R2 and RMSE scores of Mie–Rayleigh fit for μs′ are 98.83% and 0.0909.

Mie–Rayleigh parameterZonal structureAnatomical location
FLFMPLPMTLTM
α (μs)SZ22.5 ± 7.9918.23 ± 4.7017.84 ± 0.012.75 ± 1.6321.7 ± 3.868.08 ± 0.0
MZ33.25 ± 30.6422.60 ± 9.3625.02 ± 0.011.95 ± 2.2930.23 ± 9.4114.08 ± 7.38
DZ34.83 ± 8.6634.94 ± 11.0220.05 ± 0.019.64 ± 3.5645.44 ± 9.8229.86 ± 8.03
α* (μs)Bulk2.53 ± 0.162.06 ± 0.43.0 ± 0.02.34 ± 0.33.45 ± 0.622.23 ± 0.13
b (μs)SZ1.27 ± 0.311.68 ± 0.22.14 ± 0.01.22 ± 0.271.58 ± 0.140.55 ± 0.0
MZ0.85 ± 0.631.30 ± 0.471.75 ± 0.01.69 ± 0.051.49 ± 0.311.02 ± 0.11
DZ1.07 ± 0.291.22 ± 0.451.61 ± 0.01.74 ± 0.301.12 ± 0.170.98 ± 0.03
b* (μs)Bulk2.56 ± 0.182.53 ± 0.182.68 ± 0.02.56 ± 0.372.52 ± 0.032.41 ± 0.27
c (μs)SZ0.05 ± 0.070.10 ± 0.030.06 ± 0.00.27 ± 0.190.0 ± 0.00.12 ± 0.0
MZ0.07 ± 0.10.16 ± 0.10.05 ± 0.00.14 ± 0.140.0 ± 0.00.22 ± 0.06
DZ0.02 ± 0.040.04 ± 0.080.08 ± 0.00.0 ± 0.00.02 ± 0.030.07 ± 0.03
c* (μs)Bulk0.05 ± 0.060.08 ± 0.130.0 ± 0.00.17 ± 0.170.0 ± 0.00.0 ± 0.0

The results of the statistical analysis indicate a nonuniform statistically significant difference between the Mie–Rayleigh parameters of μs of the articular cartilage zones and anatomical sites (Table 3). In particular, α is significantly different between superficial and deep zones (p-value=0.001) and between middle and deep zones (p-value=0.015). However, b and c are not statistically significant across the different zones (Table 3).

Table 3

The p-value of posthoc method of group statistical test for assessing the statistically significant difference of Mie–Rayleigh parameters of μs over different articular cartilage zones and anatomical locations. μs is the single scattering coefficient of articular cartilage tissue. α is the scatterer-density parameter, b is the scatterer-size parameter, and c is the contribution of the Rayleigh scatterers to μs of articular cartilage zones. SZ, MZ, and DZ are the superficial, middle, and deep zones of articular cartilage, respectively. FL and FM are the lateral and medial femur sites, PL and PM are the lateral and medial patella sites, and TL and TM are the lateral and medial tibia sites. The italicized values highlight the statistical significance difference.

Optical propertyMie–Rayleigh parameterZoneZone
SZMZDZ
μsαSZ1.00.96940.0010
MZ0.96941.00.015
DZ0.0010.0151.0
bSZ1.01.00.3655
MZ1.01.01.0
DZ0.36551.01.0
cSZ1.01.00.1082
MZ1.01.00.1328
DZ0.10820.13281.0
Mie–Rayleigh parameterAnatomical locationAnatomical location
FLFMPLPMTLTM
αFL1.01.01.00.00030.00090.6379
FM1.01.01.00.000030.00010.4552
PL1.01.01.00.00.00.0687
PM0.00030.000030.01.01.01.0
TL0.00090.00010.01.01.01.0
TM0.63790.45520.06871.01.01.0
bFL1.01.01.00.00030.00090.6379
FM1.01.01.00.000030.00010.4552
PL1.01.01.00.00.00.0687
PM0.00030.000030.01.01.01.0
TL0.00090.00010.01.01.01.0
TM0.63790.45520.06871.01.01.0
cFL1.01.01.00.00030.00090.6379
FM1.01.01.00.000030.00010.4552
PL1.01.01.00.00.00.0687
PM0.00030.000030.01.01.01.0
TL0.00090.00010.01.01.01.0
TM0.63790.45520.06871.01.01.0

In addition, the statistical analysis suggests significant difference of Mie–Rayleigh parameters of articular cartilage μs (from all zones) when considered over anatomical locations. α, b, and c are significantly different between lateral femur and medial patella, between lateral femur and tibia, between medial femur and patella, between medial femur and lateral tibia, between lateral and medial patella, and between lateral patella and tibia, all with p-values <0.001 (Table 3). No statistically significant difference were observed for the Mie–Rayleigh scattering parameters of μs over different anatomical locations. Furthermore, similar statistical analysis was carried out for the Mie-collagen parameters of μs and μs. The results are presented in the Supplementary Material.

As observed in Fig. 5(c), deep zone forms a significant part of articular cartilage (80%) with collagen angular alignment of 60  deg. Whereas superficial (collagen angular alignment of 30  deg) and middle zone (30  deg collagen angular alignment of 60  deg) contribute roughly to 20% of articular cartilage tissue depth across different anatomical locations. The values of g (gHG and gmHG) for the different zones [Figs. 5(a) and 5(b)] suggest that light in the visible and NIR bands follows a forward-direction propagation in articular cartilage zones with higher gHG and gmHG in the spectral band of 800 to 1400 nm than the short visible band (400 to 800 nm). In addition, across all anatomical locations, gHG shows a monotonically increasing trend as function of tissue depth. That is gHG is highest in deep zone; middle zone exhibits the lowest gHG values across the anatomical sites lateral femur, and medial femur, patella, and tibia, whereas superficial zone possess the lowest gHG values across the lateral patella and tibia. When the contribution of Rayleigh scatterers are considered, superficial zone has the lowest gmHG values over the anatomical locations of lateral patella, medial patella, and lateral tibia. Whereas middle zone exhibits lowest gmHG values over the anatomical locations lateral femur, medial femur, and medial tibia.

Fig. 5

The Henyey–Greenstein scattering anisotropy factor (gHG) of articular cartilage zones (SZ, MZ, and DZ) for different anatomical origins (a). The modified Henyey-Greenstein scattering anisotropy factor (gmHG) of articular cartilage zones (SZ, MZ, and DZ), accounted for the impact of the Rayleigh scatterers on gHG (b). The depth-wise alignment of collagen fiber of articular cartilage of different anatomical origins (c); the orange shaded areas depicts the percentage of the articular cartilage zones of the total thickness according to the alignment of collagen fibers. SZ, MZ, and DZ are the superficial, middle, and deep zones of articular cartilage defined as the regions where the collagen fibers have alignment of 0 deg to 30 deg, 30 deg to 60 deg, and 60 deg to 90 deg, respectively. FL and FM are the lateral and medial femur sites, PL and PM are the lateral and medial patella sites, and TL and TM are the lateral and medial tibia sites. gHG and gmHG are unitless parameters. The values were plotted as mean ± standard deviation.

JBO_28_12_125003_f005.png

4.

Discussion

This study addresses the gap in knowledge on the scattering properties of articular cartilage depth-wise zones by determining its values of μs and g in the different zones, in addition to the depth-wise Mie–Rayleigh parameters. The present findings enhance understanding on the scattering properties of healthy bulk articular cartilage and its depthwise zones. In addition, the depth-wise angular distribution of articular cartilage collagen fibers were provided to investigate the impact of collagen orientation on articular cartilage g. In this study, we hypothesized that normal variation in articular cartilage zones will influence its scattering properties. Our results indicate that α, the Mie-Rayleigh parameter representing scatterer density in articular cartilage, exhibits statistically significant variations throughout the zones of articular cartilage. Furthermore, gHG and gmHG of articular cartilage zones show that they vary across the tissue depth.

In our view, various factors affect the values of articular cartilage g, including the depth-wise normalized fraction of the Rayleigh scatterers contribution to the scattering coefficients, the intensity of light scattered by the Rayleigh and Mie scatterers as a function of wavelength, and the heterogeneity of the tissue’s structure. When the contribution of Rayleigh scatterers to μs of bulk articular cartilage tissue is considered, on average, 5.14% of μs can be attributed to these scatterers. Hence, the impact of Rayleigh scatterers on bulk articular cartilage g is expected to be weak (gHGgmHG). When the contribution of Rayleigh scatterers to μs of the different cartilage zones is considered, their impact on g becomes more apparent. On average, these scatterers contribute to 10.19%, 10.65%, and 3.87% of μs of articular cartilage superficial, middle, and deep zones, respectively. Therefore, as observed in Fig. 5, the broadband values of articular cartilage g at superficial and middle zones are lower than those at deep zone. When g is considered in the different zones and anatomical locations (Fig. 5), our findings suggest that c (the contribution of the Rayleigh scatterers) acts as a scaling factor on the magnitude of g (gmHGcgHG), which is supported by Eq. (5).

To understand how articular cartilage g changes as a function of wavelength in the spectral band of 400 to 1400 nm, we investigated this behavior in the different tissue zones, separately from the bulk tissue. This is because collagen orientation within each zone is relatively homogeneous, except in middle zone, but the angular distribution of collagen fibers in the bulk tissue exhibits a wide range of variation, as high as 70 deg [Fig. 5(c)]. Therefore, it is plausible that the spectral behavior of g within articular cartilage zones stems from the ratio between the intensity of light scattered by the Rayleigh and Mie scatterers, whereas in the bulk tissue, the heterogeneity of angular distribution of collagen fibers acts as a confounding factor. As stated earlier, the behavior of g is primarily dependent upon the scattering phase function. The scattering phase function describes the intensity of the scattered light, which is affected by the Rayleigh and Mie scatterers. The intensity of light scattered by the Rayleigh and Mie scatterers varies in proportion to λ4 and λb, respectively. On average, the parameter b was estimated as 1.4068, 1.3508, and 1.2925 for articular cartilage superficial, middle, and deep zones, respectively (Table 2). The 4% to 8% difference in b suggests similarities across the zones, which in turn leads to similar trends of μs and g (both gHG and gmHG) in superficial, middle, and deep zones (Figs. 3 and 5).

When the impact of the zonal structure of articular cartilage on g is considered, the importance of the angular distribution of collagen fibers across the tissue depth becomes apparent. 80% of bovine articular cartilage matrix consists of deep zone with collagen fibers oriented between 60 deg and 80 deg [Fig. 5(c)]. Whereas superficial zone (fiber orientation of 0°-30°) and middle zone (fiber orientation of 30 deg to 60 deg) together comprise roughly the first 20% of bovine cartilage matrix. Hence, we speculate that in the short visible spectral band (400 to 700 nm), the structural and scattering features of superficial and middle zones, including collagen orientation and contribution of Rayleigh scatterers to the scattering coefficients, will affect the g values of articular cartilage bulk tissue. This is because of the higher intensity of Rayleigh scattering in this spectral band and the higher contribution of Rayleigh scatterers to μs of cartilage superficial and middle zones. In addition, the change of collagen fiber orientation from parallel to the articular surface in superficial zone to a skewed orientation in middle zone is likely to alter the light propagation behavior and cause the values of g to be dependent on the alignment of collagen fibers. This speculation can also be supported by the scattering mean free path (=1/μs). Due to large values of μs in the short visible band, the scattering mean free path decreases while the number of scattering events in superficial and middle zones increases. Hence, the impact of structural and scattering properties of superficial and middle zones on light scattering is higher in this spectral band.

In the far visible and NIR bands (700 to 1400 nm), the values of g for bulk cartilage is affected by the structural and scattering properties of deep zone predominantly. As the photon wavelength shifts from the short visible to the NIR region, the magnitude of μs declines and the size of the scattering mean free path increases. Given that articular cartilage is mostly comprised of deep zone with collagen fibers oriented perpendicular to the surface, most of the scattering events occur in a forward direction in deep zone. Hence, bulk articular cartilage is expected to have g values like those of superficial and middle zones in the short visible region and g values close to those of deep zone in the far visible and NIR regions.

Detailed knowledge of the light propagation characteristics in articular cartilage is critical for a better understanding of the origin of the optical response of the tissue. Light penetration depth, depth-origin, and path length of photons propagating the articular cartilage matrix are the key parameters that can shed light on the contribution of the different zones of the articular cartilage on its optical response. This study provides critical input for developing Monte Carlo models of light propagation in articular cartilage to estimate photon penetration depth, depth origin, and path length and subsequently the wavelength-dependent optical response of the tissue.

In this study, certain reasonable assumptions were made. Light propagation in biological tissues are complex. That is, due to complex architecture of their matrix, light propagation becomes anisotropic and tissue scattering coefficients and phase function become dependent on the light’s incident direction. Skin, muscle, and dentin are example of such tissues that exhibit anisotropic light propagation.33,34 We speculate that articular cartilage, due to its zonal architecture, is an example of such biological tissues too. However, we assumed an isotropic light propagation model to investigate the light interaction in articular cartilage and estimate its optical properties. The underlying reason is due to numerous unknowns associated with the anisotropic model of light propagation in articular cartilage that makes its usage not feasible.

In general, the reported μs values are from different layers of articular cartilage, whereas the μs values are obtained from the bulk of articular cartilage tissue, hence μs is a lump-sum representation of the μs of articular cartilage layers. However, the low thickness of the tissue’s superficial and middle zones (less than a few hundred micrometers) makes it impossible to extract sufficiently thick sections from superficial and middle zones to estimate their μs. Therefore, our approach was to estimate the μs of bulk articular cartilage and μs of articular cartilage zones and investigate the relationship between the scattering properties of the bulk tissue and its zones through [Eq. (4)]. In this context, we assumed μs,bulk tissue/μs,cartilage zone represents a normalized contribution of individual cartilage zones to the scattering properties of bulk tissue.

As the broadband values of n and scattering phase function of bulk articular cartilage tissue are not known and the estimation of μa and μs depends on these parameters, we used fixed values of n and g (applying to the Henyey–Greenstein scattering phase function used in this study) to estimate its broadband μa and μs values. Bergmann et al.22 reported that the variation of n in the range of 1.3 to 1.5 could result in a relative change of less than 5% for μs and <13% for μa of biological tissues. More so, in our previous study,29 we showed that using fixed value of g (0.8g0.99) will not result in statistically significant change in the estimation of articular cartilage μa and μs. Thus, we expect that using fixed value of g for estimation of bulk articular cartilage μs has minimal impact on the reported values of g for different zones of articular cartilage.

Furthermore, the choice of scattering phase function used in the estimation of μs plays a crucial role in determining the spectral values of g, which is inherently dependent upon the scattering phase function, and defined as the first Legendre moment of the scattering phase function.1 In this study, the Henyey–Greenstein phase function was used to estimate the μs of bulk articular cartilage. As we utilized the relationship between μs and μs [Eq. (4)] to estimate articular cartilage g, we expect that the estimated values of g will also be dependent upon the Henyey-Greenstein scattering phase function. The Henyey–Greenstein phase function is biased toward the forward scattering of light by Mie scatterers.1,3 Therefore, the impact of the Rayleigh scatterers on the isotropic scattering of light and subsequently g might not be properly captured by the Henyey–Greenstein scattering phase function. To minimize the impact of this bias, Eqs. (2) and (5) were utilized to estimate the volume fraction of the Rayleigh scatterers and to account for their impact on articular cartilage g. We estimated the volume fraction of the Rayleigh scatterers in each zone of articular cartilage from different anatomical origins and showed how the depth-wise normalized contribution of the Rayleigh scatterers to μs of articular cartilage will impact the corresponding g values [Table 2 and Fig. 5(b)]. We also provided the spectral values of g based on the Henyey–Greenstein phase function (gHG) and when the contribution of Rayleigh scatterers is accounted for (gmHG). We then enumerated the similarities and differences between gHG and gmHG across cartilage zones and various anatomical locations.

Although it is often theoretically expected to observe decreasing broadband values of g with increasing wavelength, our results show an increasing trend in the broadband values of g for articular cartilage. We performed Monte Carlo simulations, for which we considered the actual experimental setup, showing that for our apparatus and μt50  mm1 and a g0.99, there was no significant contribution from multi-scattered light using the Henyey–Greenstein phase function. Hence, the measurement geometry and the resulting values for μt should be precise. Therefore, the increasing trend of g observed in this study could be potentially due to the alignment of collagen fibers, a special diameter distributions of the involved collagen fibers, and further scattering structures, such as the chondrocytes. Formulating the impact of these structures in articular cartilage μs and its broadband g requires an independent investigation, which is out of the scope of the present study.

5.

Conclusion

This study reports the broadband μs and g of articular cartilage zonal structure in the spectral band of 400 to 1400 nm. To achieve this, we carried out integrating sphere measurements on bulk tissue of articular cartilage and collimated transmittance measurements on thin layers of articular cartilage that represent its zonal structure. Our findings suggest that the depth-wise distribution of the volume fraction of Rayleigh scatterers, the intensity of the light scattered by the Rayleigh and Mie scatterers, and the orientation of collagen fibers across the articular cartilage matrix are the key parameters that affect the broadband values of g of articular cartilage zones and bulk tissue.

Disclosures

The authors declare no conflict of interest.

Code and Data Availability

The data reported in this article is available upon reasonable request.

Acknowledgments

Dr. Afara acknowledges funding from the Academy of Finland (Grant Nos. 315820, 320135, and 345670) and the Jane and Aatos Erkko Foundation (Grant No. 190001). Iman Kafian-Attari acknowledges funding from the Kuopio University Hospital (VTR grant) and the Finnish Academy of Science and Letters. The authors also acknowledge facility support from Advanced X-ray Tomography Laboratory for supporting the reform of local R&D&I operations project, Regional Council of Pohjois-Savo (Grant No. A74798), and ERDF Sustainable Growth and Jobs - Structural Fund Programme of Finland 2014–2020.

References

1. 

S. L. Jacques and B. W. Pogue, “Tutorial on diffuse light transport,” J. Biomed. Opt., 13 (4), 041302 https://doi.org/10.1117/1.2967535 JBOPFO 1083-3668 (2008). Google Scholar

2. 

L. Rieppo, J. Töyräs and S. Saarakkala, “Vibrational spectroscopy of articular cartilage,” Appl. Spectrosc. Rev., 52 (3), 249 –266 https://doi.org/10.1080/05704928.2016.1226182 APSRBB 0570-4928 (2017). Google Scholar

3. 

S. L. Jacques, “Optical properties of biological tissues: a review,” Phys. Med. Biol., 58 (11), R37 –R61 https://doi.org/10.1088/0031-9155/58/11/R37 PHMBA7 0031-9155 (2013). Google Scholar

4. 

I. O. Afara et al., “Characterization of connective tissues using near-infrared spectroscopy and imaging,” Nat. Protoc., 16 (2), 1297 –1329 https://doi.org/10.1038/s41596-020-00468-z 1754-2189 (2021). Google Scholar

5. 

H. T. Kokkonen et al., “Computed tomography detects changes in contrast agent diffusion after collagen cross-linking typical to natural aging of articular cartilage,” Osteoarthr. Cartil., 19 (10), 1190 –1198 https://doi.org/10.1016/j.joca.2011.07.008 (2011). Google Scholar

6. 

S. B. Mahbub et al., “Non-invasive monitoring of functional state of articular cartilage tissue with label-free unsupervised hyperspectral imaging,” Sci. Rep., 9 (1), 4398 https://doi.org/10.1038/s41598-019-40942-7 SRCEC3 2045-2322 (2019). Google Scholar

7. 

S. Nebelung et al., “Morphometric grading of osteoarthritis by optical coherence tomography - an ex vivo study: morphometric grading of osteoarthritis,” J. Orthop. Res., 32 (10), 1381 –1388 https://doi.org/10.1002/jor.22673 JOREDR 0736-0266 (2014). Google Scholar

8. 

R. Shaikh et al., “Raman spectroscopy is sensitive to biochemical changes related to various cartilage injuries,” J. Raman Spectrosc., 52 (4), 796 –804 https://doi.org/10.1002/jrs.6062 JRSPAF 0377-0486 (2021). Google Scholar

9. 

University of California Irvine, 3418 Engineering Hall, Irvine, CA 92697, USAet al., “Nondestructive fluorescence lifetime imaging and time-resolved fluorescence spectroscopy detect cartilage matrix depletion and correlate with mechanical properties,” Eur. Cell. Mater., 36 30 –43 https://doi.org/10.22203/eCM.v036a03 (2018). Google Scholar

10. 

A. J. Sophia Fox, A. Bedi and S. A. Rodeo, “The basic science of articular cartilage: structure, composition, and function,” Sports Health Multidiscip. Approach, 1 (6), 461 –468 https://doi.org/10.1177/1941738109350438 (2009). Google Scholar

11. 

F. H. Epstein and D. Hamerman, “The biology of osteoarthritis,” N. Engl. J. Med., 320 (20), 1322 –1330 https://doi.org/10.1056/NEJM198905183202006 NEJMAG 0028-4793 (1989). Google Scholar

12. 

N. Arden and M. Nevitt, “Osteoarthritis: epidemiology,” Best Pract. Res. Clin. Rheumatol., 20 (1), 3 –25 https://doi.org/10.1016/j.berh.2005.09.007 (2006). Google Scholar

13. 

J. F. Beek et al., “In vitro double-integrating-sphere optical properties of tissues between 630 and 1064 nm,” Phys. Med. Biol., 42 (11), 2255 –2261 https://doi.org/10.1088/0031-9155/42/11/017 PHMBA7 0031-9155 (1997). Google Scholar

14. 

M.-A. Descalle et al., “Measurements of ligament and cartilage optical properties at 351 nm, 365 nm, and in the visible range (440 to 800 nm),” 280 –286 (1998). Google Scholar

15. 

S. J. Madsen, E. A. Chu and B. J. F. Wong, “The optical properties of porcine nasal cartilage,” IEEE J. Sel. Top. Quantum Electron., 5 (4), 1127 –1133 https://doi.org/10.1109/2944.796339 IJSQEN 1077-260X (1999). Google Scholar

16. 

J.-I. Youn et al., “Optical and thermal properties of nasal septal cartilage,” Lasers Surg. Med., 27 (2), 119 –128 https://doi.org/10.1002/1096-9101(2000)27:2<119::AID-LSM3>3.0.CO;2-V LSMEDI 0196-8092 (2000). Google Scholar

17. 

V. N. Bagratashvili et al., “Change in the optical properties of hyaline cartilage heated by the near-IR laser radiation,” Quantum Electron., 31 (6), 534 –538 https://doi.org/10.1070/QE2001v031n06ABEH001996 QUELEZ 1063-7818 (2001). Google Scholar

18. 

D. W. Ebert, “Articular cartilage optical properties in the spectral range 300–850 nm,” J. Biomed. Opt., 3 (3), 326 https://doi.org/10.1117/1.429893 JBOPFO 1083-3668 (1998). Google Scholar

19. 

T. Reuter et al., “Determination of the optical properties of native joint cartilage with NIR – spectroscopy,” Biomed. Eng. Biomed. Tech., 58 743 –744 https://doi.org/10.1515/bmt-2013-4357 (2013). Google Scholar

20. 

A. V. Yuzhakov et al., “Optical properties of costal cartilage and their variation in the process of non-destructive action of laser radiation with the wavelength 1.56 μm,” Quantum Electron., 44 (1), 65 –68 https://doi.org/10.1070/QE2014v044n01ABEH015343 QUELEZ 1063-7818 (2014). Google Scholar

21. 

I. Kafian-Attari et al., “Tissue optical properties combined with machine learning enables estimation of articular cartilage composition and functional integrity,” Biomed. Opt. Express, 11 (11), 6480 https://doi.org/10.1364/BOE.402929 BOEICL 2156-7085 (2020). Google Scholar

22. 

F. Bergmann et al., “Ex vivo determination of broadband absorption and effective scattering coefficients of porcine tissue,” Photonics, 8 (9), 365 https://doi.org/10.3390/photonics8090365 (2021). Google Scholar

23. 

J.-J. Shyu et al., “Diagnosis of articular cartilage damage by polarization sensitive optical coherence tomography and the extracted optical properties,” Prog. Electromagn. Res., 91 365 –376 https://doi.org/10.2528/PIER09022602 PELREX 1043-626X (2009). Google Scholar

24. 

A. N. Jambor et al., “Assessing collagen alterations in enzymatic degradation models of osteoarthritis via second harmonic generation microscopy,” Osteoarthr. Cartil., 29 (11), 1590 –1599 https://doi.org/10.1016/j.joca.2021.08.004 (2021). Google Scholar

25. 

S. A. Prahl et al., “Determining the optical properties of turbid media by using the adding–doubling method,” Appl. Opt., 32 (4), 559 –568 https://doi.org/10.1364/AO.32.000559 (1993). Google Scholar

26. 

I. Kafian-Attari et al., “Impact of experimental setup parameters on measurement of articular cartilage optical properties in the visible and short near-infrared spectral bands,” Biomed. Opt. Express, 14 (7), 3397 –3412 https://doi.org/10.1364/BOE.488801 BOEICL 2156-7085 (2023). Google Scholar

27. 

F. Foschum, F. Bergmann and A. Kienle, “Precise determination of the optical properties of turbid media using an optimized integrating sphere and advanced Monte Carlo simulations. Part 1: theory,” Appl. Opt., 59 (10), 3203 https://doi.org/10.1364/AO.386011 APOPAI 0003-6935 (2020). Google Scholar

28. 

F. Bergmann et al., “Precise determination of the optical properties of turbid media using an optimized integrating sphere and advanced Monte Carlo simulations. Part 2: experiments,” Appl. Opt., 59 (10), 3216 https://doi.org/10.1364/AO.385939 APOPAI 0003-6935 (2020). Google Scholar

29. 

S.-Z. Wang et al., “Assessment of depth and degeneration dependences of articular cartilage refractive index using optical coherence tomography in vitro,” Connect. Tissue Res., 51 (1), 36 –47 https://doi.org/10.3109/03008200902890161 CVTRBC 0300-8207 (2010). Google Scholar

30. 

S. B. Mehta, M. Shribak and R. Oldenbourg, “Polarized light imaging of birefringence and diattenuation at high resolution and high sensitivity,” J. Opt., 15 (9), 094007 https://doi.org/10.1088/2040-8978/15/9/094007 (2013). Google Scholar

31. 

J. Rieppo et al., “Practical considerations in the use of polarized light microscopy in the analysis of the collagen network in articular cartilage,” Microsc. Res. Tech., 71 (4), 279 –287 https://doi.org/10.1002/jemt.20551 MRTEEO 1059-910X (2008). Google Scholar

32. 

V. V. Tuchin, “Polarized light interaction with tissues,” J. Biomed. Opt., 21 (7), 071114 https://doi.org/10.1117/1.JBO.21.7.071114 JBOPFO 1083-3668 (2016). Google Scholar

33. 

R. Graaff et al., “Optical properties of human dermis in vitro and in vivo,” Appl. Opt., 32 (4), 435 https://doi.org/10.1364/AO.32.000435 APOPAI 0003-6935 (1993). Google Scholar

34. 

F. Martelli et al., Light Propagation Through Biological Tissue and Other Diffusive Media: Theory, Solutions, and Validations, 2nd ed.SPIE Press, Bellingham, Washington (2022). Google Scholar

Biographies of the authors are not available.

CC BY: © The Authors. Published by SPIE under a Creative Commons Attribution 4.0 International License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.
Iman Kafian-Attari, Ervin Nippolainen, Florian Bergmann, Akuroma George, Petri Paakkari, Arash Mirhashemi, Florian Foschum, Alwin Kienle, Juha Töyräs, and Isaac O. Afara "Broadband scattering properties of articular cartilage zones and their relationship with the heterogenous structure of articular cartilage extracellular matrix," Journal of Biomedical Optics 28(12), 125003 (13 December 2023). https://doi.org/10.1117/1.JBO.28.12.125003
Received: 22 June 2023; Accepted: 3 November 2023; Published: 13 December 2023
Advertisement
Advertisement
KEYWORDS
Cartilage

Tissues

Collagen

Light scattering

Rayleigh scattering

Scattering

Anatomy

Back to Top