Poster 116 - Morphometry Analysis Workflow for Micro-Computed Tomography Images of Osteochondral Defect Repair

Saturday, October 26, 2024

10:00 AM - 12:00 PM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Emmy Shulman (she/her/hers)

High School Student
The Emery / Weiner School
Houston, Texas, United States

Last Author(s)

RS

Robert Sah

Professor
University of California San Diego, United States

Co-Author(s)

AC

Albert Chen, PhD

Project Scientist
University of California San Diego, United States

Introduction: A variety of methods to repair focal articular cartilage defects are being studied [2]. Characterization of repair has traditionally been with histology, histochemical stains, and immunohistochemistry; with semi-quantitative grading systems, the presence and extent of tissues including bone, articular-like cartilage, and undesirable fibrous tissue are scored [3]. Histomorphometry of tissue sections provides quantification of the relative amounts of such tissues, overall at the defect site, and in the surface region that normally is articular cartilage [6].

Micro-computed tomography (microCT) provides 3-D information on tissue structure. Standard microCT, without contrast, sensitively localizes mineralized tissue [1]. With Hexabrix contrast, microCT can additionally distinguish fibrous tissue from cartilaginous tissue, based on the latter repelling the contrast agent [5]. Hexabrix-enhanced microCT has been useful to characterize the cartilaginous tissue in osteochondral grafts, where the osteochondral unit is typically intact [4]. However, in more complex repairs, the cartilage and bone regions need to regenerate the osteochondral unit. Alone, contrast-enhanced microCT images of such repair may be difficult to interpret, as regions of high opacity may either indicate mineralized or fibrous that has taken up contrast. Thus, combined analysis of registered non-contrast and contrast microCT images can, in principle, help distinguish the three types of tissue, mineralized, cartilaginous, and fibrous.

The aim of this study was to develop an image processing workflow for pairs of microCT scans, one without contrast and one with Hexabrix stain, that is suitable for the normal osteochondral unit as well as for focal defect repair.

Materials and

Methods: 2D image slices from 3D microCT scans of normal distal femurs and repaired articular cartilage defects were obtained. There was one pair of aligned images for each sample, one image without contrast and one image with Hexabrix contrast. Image processing was performed with imageJ. An area of interest (AOI) was defined to be 3mm wide x 3.5mm tall encompassing air, bone, and the repair region.

First, masks were created from non-contrast images and used to separate soft tissue and air from mineralized tissues. To create the mask, images were smoothed, a threshold between histogram peaks, representing air and bone, was applied, and the resultant image was dilated to reduce partial volume effects. Next, masks were created from Hexabrix images and used to separate air from soft tissue. To create the mask, a threshold between histogram peaks, representing air and soft tissue, was applied, and the resultant image also dilated to reduce partial volume effects. Then, in the soft tissue region, cartilage and fibrous tissues were distinguished by thresholding at a level in between the Hexabrix stain intensity and fluid, similar to unstained soft tissue. Finally, to assess the region normally occupied by articular cartilage, a mask of that region was created and applied. Here, the air mask was advanced by the cartilage thickness, in both normal and repair samples.

From the masked regions, the absolute and relative amounts of mineralized tissue, fibrous tissue, and cartilaginous tissue, were obtained.

Results, Conclusions, and Discussions: The AOIs, cropped from the original 2D image slices showed distinctive patterns for Normal and Repair samples (Fig.1). Both unstained (Fig.1AB) and Hexabrix-stained (Fig.1CD) images contained bright interconnected struts (top) consistent with normal or repairing trabecular bone. The Normal (Fig.1AC), but not Repair (Fig.1BD), sample contained a continuous subchondral plate region, capped below by dim signal consistent with articular cartilage. In contrast, the Repair sample had a larger region of dim signal (Fig 1B), that variably brightened with Hexabrix (Fig 1D)

Masks effectively isolated regions of soft tissue. Application of masks of mineralized tissue (Fig 2AB) to Hexabrix-stained images (Fig 1C,D) removed the bright regions of mineralization (Fig.2C,D). To those images, additional application of masks of the air region (Fig.3AB) resulted in a clear border at the soft tissue surface and no signal below (Fig.3CD). Finally, these Hexabrix-stained regions of soft tissue in the AOI were trimmed by application of an articular cartilage mask (Fig.4AB) to obtain Hexabrix-stained soft tissue in the normal location of articular cartilage.

I am currently quantifying the absolute and relative area and intensities of mineralized tissue (from Fig 2AB) and soft tissues (from Fig.3CD and 4CD), and distinguishing between cartilaginous vs fibrous tissues. In addition, I am evaluating how mask thresholds affect the images and metrics. Overall, the current results establish the foundations of an automated workflow for analysis of microCT scans of normal osteochondral tissue and repairing cartilage defects.

Acknowledgements (Optional):

References (Optional): 1. Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R. Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res. 25:1468-86, 2010. doi: 10.1002/jbmr.141

2. Hinckel BB, Thomas D, Vellios EE, Hancock KJ, Calcei JG, Sherman SL, Eliasberg CD, Fernandes TL, Farr J, Lattermann C, Gomoll AH. Algorithm for treatment of focal cartilage defects of the knee: Classic and new procedures. Cartilage. 13S1:473-95, 2021. doi: 10.1177/1947603521993219.

3. Hoemann CD, Kandel R, Roberts S, Saris DB, Creemers L, Mainil-Varlet P, Methot S, Hollander AP, Buschmann MD. International Cartilage Repair Society (ICRS) recommended guidelines for histological endpoints for cartilage repair studies in animal models and clinical trials. Cartilage. 2:153-72, 2011. doi: 10.1177/1947603510397535

4. Pallante AL, Gortz S, Chen AC, Healey RM, Chase DC, Ball ST, Amiel D, Sah RL, Bugbee WD. Treatment of articular cartilage defects in the goat with frozen versus fresh osteochondral allografts: effects on cartilage stiffness, zonal composition, and structure at six months. J Bone Joint Surg Am. 94:1984-95, 2012. doi: 10.2106/JBJS.K.00439

5. Palmer AW, Guldberg RE, Levenston ME. Analysis of cartilage matrix fixed charge density and three-dimensional morphology via contrast-enhanced microcomputed tomography. Proc Natl Acad Sci U S A. 103:19255-60, 2006. doi: 10.1073/pnas.0606406103

6. Yoshioka M, Coutts RD, Amiel D, Hacker SA. Characterization of a model of osteoarthritis in the rabbit knee. Osteoarthritis Cartilage. 4:87-98, 1996. doi: 10.1016/s1063-4584(05)80318-8