Journal: Journal of Medical Imaging

Article Title: GIFTed Demons: deformable image registration with local structure-preserving regularization using supervoxels for liver applications

doi: 10.1117/1.JMI.5.2.024001

Figure Lengend Snippet: Example of sparse image representation using SLIC: (a) coronal view of 3-D CT lung and liver volume, (b) projection through 3-D supervoxel representation with supervoxel boundaries and (c) with assignment of mean intensity. The SLIC algorithm with different values of the parameter K = 11,000 (top) and K = 5500 (bottom) shows that clustering is consistent in image regions with sufficient structural information (close to edges, e.g., the sliding surfaces of lungs), while different clusters are generated in homogeneous image regions.

Article Snippet: The computation time per registration using the presented framework is ≈ 3 min per 3-D pair (on a standard CPU, running nonoptimized C++ code, MATLAB™ mex compiler) and is several times faster than our previous bilateral filtering procedure ( ≈ 60 min ) or the locally adaptive anisotropic regularization (several hours).

Techniques: Generated

Journal: Journal of Medical Imaging

Article Title: GIFTed Demons: deformable image registration with local structure-preserving regularization using supervoxels for liver applications

doi: 10.1117/1.JMI.5.2.024001

Figure Lengend Snippet: Main anatomical views of 3-D CT registration results for case #P0 of the liver dataset: (a) coronal, (b) axial, and (c) sagittal views for the color-coded (red-cyan) intensity differences between volume pair before registration (left), after registration using Demons with isotropic Gaussian kernel, iso-dem, (middle), and guided image filtering with random SLIC clustering, rdn-gif, (right). Registration using our method (right) improves registration accuracy especially close to the lung and liver surfaces (depicted by corresponding red dotted and green solid arrows, respectively).

Article Snippet: The computation time per registration using the presented framework is ≈ 3 min per 3-D pair (on a standard CPU, running nonoptimized C++ code, MATLAB™ mex compiler) and is several times faster than our previous bilateral filtering procedure ( ≈ 60 min ) or the locally adaptive anisotropic regularization (several hours).

Techniques:

Journal: Journal of Medical Imaging

Article Title: GIFTed Demons: deformable image registration with local structure-preserving regularization using supervoxels for liver applications

doi: 10.1117/1.JMI.5.2.024001

Figure Lengend Snippet: Main anatomical views of resulting 3-D displacement fields for case #P0 of the liver CT dataset: (a) coronal, (b) axial, and (c) sagittal views for the color-coded magnitude of the displacement field estimated using Demons with isotropic Gaussian kernel, iso-dem, (middle) and guided image filtering with random SLIC clustering, rdn-gif, (right). (left) The reference image with the corresponding blue contour is shown for a guidance to the displacement field. Registration using our method (right) produces a visually smooth displacement field inside the lungs and liver, and at the same, estimates sliding motion at the lung and liver interface [depicted by corresponding red dotted (for lungs) and green solid (for liver) arrows].

Article Snippet: The computation time per registration using the presented framework is ≈ 3 min per 3-D pair (on a standard CPU, running nonoptimized C++ code, MATLAB™ mex compiler) and is several times faster than our previous bilateral filtering procedure ( ≈ 60 min ) or the locally adaptive anisotropic regularization (several hours).

Techniques:

Journal: Oncotarget

Article Title: Inhibition of HAS2 and hyaluronic acid production by 1,25-Dihydroxyvitamin D 3 in breast cancer

doi: 10.18632/oncotarget.27587

Figure Lengend Snippet: ( A ) Images of VDR, HAS2, and HA localization in Hs578T cells. ( B ) Lysates from Hs578T cells treated with vehicle or 100 nM 1,25D3 for 48 hours were blotted with antibodies against VDR, HAS2, or loading control GAPDH. ( C ) RNA was isolated from Hs578T cells treated with 100 nM 1,25D3 for 24 hours. CYP24A1 , HAS2 and CD44 were quantitated by RT-qPCR, normalized to 18S and expressed relative to vehicle treated cells. Secreted HA was evaluated by ELISA of conditioned media removed 48 hours after 1,25D3 or vehicle treatment. Bars represent mean ± standard error, with quadruplicates for RT-qPCR and triplicates for ELISA. * p < 0.05, control vs 1,25D3 treated in each cell line as evaluated by Student’s t test. ( D ) Phase contrast images of Hs578T cells treated with 100 nM 1,25D3 or vehicle for 96 hours. ( E ) Crystal violet assay for culture density was conducted in HS578T cells treated with 100 nM 1,25D3 for 144 hours. ( F ) Crystal violet assay for culture density was conducted in HS578T cells treated with HA synthesis inhibitors 4MU or DON for 96 hours. Data from growth assays are mean ± standard error of quadruplicates. * p < 0.05, control vs treated as evaluated by Student’s t test. ( G ) Percentages of cells in G1, S and G2 phases of the cycle after 96 hours treatment with 100 nM 1,25D ± 100 μM 4MU. Data are mean of duplicates with 5000 cells analyzed per run. ( H ) Viable cell count (mean ± standard deviation, n = 4) after 96 hours treatment with 100 nM 1,25D ± 100 μM 4MU. Bars annotated with different letters are significant at p < 0.05 by one-way ANOVA and Bonferroni post -test. ( I ) Percentage of Annexin positive cells indicative of apoptosis (mean ± standard deviation, n = 4) after 96 hours treatment with 100 nM 1,25D ± 100 μM 4MU. Bars annotated with different letters are significant at p < 0.05 by one-way ANOVA and Bonferroni post -test.

Article Snippet: RT-qPCR was performed for HAS1 , HAS2 , HAS3 , HYAL1 , HYAL2 , HYAL3 , VDR , CYP24A1 , and CD44 gene expression using primers from Origene and SYBR green master mix (Life Technologies Applied Biosystems).

Techniques: Isolation, Quantitative RT-PCR, Enzyme-linked Immunosorbent Assay, Crystal Violet Assay, Cell Counting, Standard Deviation