Three-dimensional computerized reconstruction. Illustration of incremental articular cartilage thinning.

RATIONALE AND OBJECTIVES: The authors have addressed the ability of magnetic resonance (MR) imaging to resolve incremental thinning of articular cartilage by assessment of three-dimensional (3-D) and two-dimensional (2-D) representations. METHODS: Using a porcine knee model, sequential cartilage shavings were characterized using a 3-D fat suppressed spoiled gradient-echo (SPGR) MR imaging protocol that provided good contrast between high-signal articular cartilage and lower signal surrounding tissues. Lesion dimensional measurements were made on both MR images and 3-D computerized reconstructions. Volumes of cartilage removed were approximately 0.06 mL. RESULTS: Incremental articular cartilage thinning typically was apparent on 3-D reconstructed images. Three-dimensional articular cartilage reconstructions were effective in depicting location and orientation of shaved cartilage regions. Average percent error associated with length and with measurements based on 2-D MR images was approximately 19% for observer 1 and 33% for observer 2 when compared with direct measurements of the shaved cartilage. Average percent error of thickness measurements based on 2-D MR was approximately 21% for observer 1 and 37% for observer 2. Overall average errors associated with length, width, and thickness measurements were approximately 25%. CONCLUSIONS: Incremental thinning of articular cartilage can be tracked qualitatively and quantitatively using 3-D computerized reconstructions and 2-D MR images. Errors associated with the quantitative measurements can be attributed to limitations of measurement methods and intrinsic limitation of MR resolution.

Analysis of three-dimensional computerized representations of articular cartilage lesions.

RATIONALE AND OBJECTIVES: The authors assess the accuracy of three-dimensional (3D) computer representations based on magnetic resonance images of articular cartilage lesions, using actual cartilage lesions as reference standards. METHODS: Grade 2 and grade 3 articular lesions were created on articular surfaces of five porcine knee joints. The knees were then imaged using 3D fat-suppressed SPGR acquisition at four different slice thicknesses. Magnetic resonance imaging data sets were transferred to a computer workstation for image processing and 3D reconstruction. Lesion dimensions (length, width, and depth) based on the 3D reconstructed image were compared with the dimensions measured using actual lesions. RESULTS: The average percent error of lesion length, width, and depth based on the 3D images ranged from approximately 8% to 12% when using the thinnest magnetic resonance slice thickness (0.7 mm). CONCLUSIONS: Three-dimensional reconstructed images derived from thin-slice magnetic resonance imaging can provide reasonable representations of true articular cartilage lesion dimensions.

Texture analysis of cerebral white matter in SIV-infected macaque monkeys.

Image texture analysis is used in a wide variety of applications in medical research. Neurovirulent simian immunodeficiency virus (SIV) infection in monkeys is considered a good model for HIV-1 infection in humans and causes neuropathological changes in white matter which can include diffuse myelin pallor, subtle white matter astrocytosis, perivascular macrophage infiltrates, and microglial nodules with multinucleated giant cells. The ability of image texture analysis to quantify these changes was evaluated. Sections of thionin-stained brain tissue from eight male rhesus macaques ranging in age from 42-59 months were used. Four animals served as controls and four animals were infected with neurovirulent SIVmac239/17E-R71 by bone marrow inoculation. Images of cerebral white matter were captured and analyzed by calculating 13 textural features based on statistical analysis of spatial co-occurrence matrices. Statistical analysis of the results included multiple comparisons using the Newman-Keuls multiple range test. The effect of variation in background illumination used at image acquisition was also evaluated. Ten of the 13 textural features used in this study successfully discriminated between tissue from control and SIV-infected animals and were consistent with independent neuropathological assessment. Three textural features were highly sensitive to variation in background illumination and found not useful in this application.

Treating hop plants with (Z)-jasmone increases colonization by Phorodon humuli (Hemiptera: Aphididae) spring migrants.

Hop plants were sprayed with (Z)-jasmone, at a rate of 50 g ha(-1), during the spring migration of the damson-hop aphid Phorodon humuli (Schrank) in 2002 and 2003. Numbers of P. humuli spring migrants colonizing hop plants, Humulus lupulus L., 2-6 and 7-11 days after applying this treatment were assessed in both years. During the first five-day period, significantly more spring migrants were found on hop plants treated with (Z)-jasmone, in comparison with control plants, in 2002. By contrast, no significant difference was evident in the second five-day period. Although the migration in 2003 was much lighter than in 2002, greater numbers of migrants were again removed from treated plants. Indeed, more spring migrants were removed from plants sprayed with (Z)-jasmone in this year during both five-day periods (11 and 44%, respectively) compared with the 23% greater numbers removed in the first five-day period in 2002. Therefore, unlike some other species of aphid, where numbers were consistently lower on plots sprayed with (Z)-jasmone, spraying the secondary host of P. humuli with this compound appears to increase colonization by spring migrants.

Textural parameters based on the Spatial Gray Level Dependence Method applied to melanocytic nevi.

The purpose of this study is to evaluate the Spatial Gray Level Dependence Method (SGLDM) of texture analysis with respect to its ability to discriminate between melanocytic nevi and normal skin. Thirteen textural features based on the SGLDM were evaluated with respect to their relative sensitivities to both texture and tone. Ten features were found to be more sensitive to texture than tone and were selected for further study. Twenty-four digitized images of benign melanocytic nevi were obtained from six volunteers. Ten textural features were analyzed for each nevus and for surrounding sections of normal skin. Of these 10 features, 8 features can distinguish between textural properties of melanocytic nevi and surrounding skin.