Atherosclerotic plaque (AP) composition is an important factor influencing plaque rupture. Intraplaque hemorrhage (IPH) is a marker of complicated-plaque formation, responsible for many of the clinical manifestations of atherosclerotic disease. Magnetic resonance imaging (MRI) has emerged as a modality to image carotid AP. The in-vivo high-resolution MR imaging of carotid complicated-plaque forms the basis of this thesis. In the first part, Magnetic Resonance Imaging of Intraplaque Hemorrhage (MRIPH), an in-vivo high-spatial-resolution 3-dimensional MRI sequence specifically designed to depict complicated-plaque in the carotid arteries is described. MRIPH was applied, as compared with histologic analysis (gold standard), to demonstrate that T1-hyperintense intraplaque signal represented blood products. Strong agreement was seen between T1-hyperintensity and histologically-identified hemorrhage, with high sensitivity/specificity/positive- and negative-predictive values for T1-hyperintense detection of hemorrhage. While IPH increases plaque rupture risk, high degrees of calcification promote stability. Calcium can generate T1-hyperintensity in some gradient-echo (GRE) sequences. Therefore, distinction between these two components is crucial. In the second part, T1-hyperintensity in MRIPH was shown to be almost exclusively due to hemorrhage and not calcification by directly comparing in-vivo T1-hyperintensity with calcification in ex-vivo specimens imaged with microCT. T1- hyperintesity showed very good albeit inverse agreement with calcification and excellent agreement with lack of calcification as seen on microCT. IPH is thought to be the result of rupture/leakage of the vasa vasora. In the third part, we tested the hypothesis that if IPH were due to vasa vasorum rupture/leakage, the majority of the IPH would be associated with the adventitial rather than the luminal surface of the plaque. Deep (closer to vessel wall) and superficial (closer to vessel lumen) regions of complicated plaques were identified. Very good inter-rater agreement was seen for the location of IPH using MRIPH, with IPH being more frequently present in the deeper compared to superficial segments of the plaque. In summary, an in-vivo MR technique to detect IPH at high spatial-resolution in carotid complicated-plaque was developed; demonstrating T1-hyperintensity in MRIPH is the result of IPH and not calcification. The predilection of IPH for the deeper segments of the plaques suggests that IPH is due to vasa vasorum rupture/leakage.

Atherosclerotic plaque (AP) composition is an important factor influencing plaque rupture. Intraplaque hemorrhage (IPH) is a marker of complicated-plaque formation, responsible for many of the clinical manifestations of atherosclerotic disease. Magnetic resonance imaging (MRI) has emerged as a modality to image carotid AP. The in-vivo high-resolution MR imaging of carotid complicated-plaque forms the basis of this thesis. In the first part, Magnetic Resonance Imaging of Intraplaque Hemorrhage (MRIPH), an in-vivo high-spatial-resolution 3-dimensional MRI sequence specifically designed to depict complicated-plaque in the carotid arteries is described. MRIPH was applied, as compared with histologic analysis (gold standard), to demonstrate that T1-hyperintense intraplaque signal represented blood products. Strong agreement was seen between T1-hyperintensity and histologically-identified hemorrhage, with high sensitivity/specificity/positive- and negative-predictive values for T1-hyperintense detection of hemorrhage. While IPH increases plaque rupture risk, high degrees of calcification promote stability. Calcium can generate T1-hyperintensity in some gradient-echo (GRE) sequences. Therefore, distinction between these two components is crucial. In the second part, T1-hyperintensity in MRIPH was shown to be almost exclusively due to hemorrhage and not calcification by directly comparing in-vivo T1-hyperintensity with calcification in ex-vivo specimens imaged with microCT. T1- hyperintesity showed very good albeit inverse agreement with calcification and excellent agreement with lack of calcification as seen on microCT. IPH is thought to be the result of rupture/leakage of the vasa vasora. In the third part, we tested the hypothesis that if IPH were due to vasa vasorum rupture/leakage, the majority of the IPH would be associated with the adventitial rather than the luminal surface of the plaque. Deep (closer to vessel wall) and superficial (closer to vessel lumen) regions of complicated plaques were identified. Very good inter-rater agreement was seen for the location of IPH using MRIPH, with IPH being more frequently present in the deeper compared to superficial segments of the plaque. In summary, an in-vivo MR technique to detect IPH at high spatial-resolution in carotid complicated-plaque was developed; demonstrating T1-hyperintensity in MRIPH is the result of IPH and not calcification. The predilection of IPH for the deeper segments of the plaques suggests that IPH is due to vasa vasorum rupture/leakage.

Objective: The purpose of this study was to compare the accuracy of cone beam CT (CBCT) imaging with digital orthopantomograms for determining bone quantity in the mandibular ramus. Methods: Twenty-nine cadaveric mandibles marked bilaterally with three fiducial markers were imaged using both CBCT and digital orthopantomography. After sectioning, four cross sectional measurements were made on the specimens and on the CBCT images. Two corresponding linear measurements were made on the orthopantomograms. Statistical analysis was used to compare the CBCT and orthopantomogram measurements with measurements from the anatomic specimens. Results: CBCT measurements were found to significantly differ from those made on the anatomic specimens (P<0.05). Linear measurements from the orthopantomograms varied by 15.9 percent compared to the anatomic specimens. Conclusion: CBCT and orthopantomogram measurements were significantly different from those of the anatomic specimens suggesting inaccuracies in the radiographic technology or a lack of precision in landmark identification.

Objective: The purpose of this study was to compare the accuracy of cone beam CT (CBCT) imaging with digital orthopantomograms for determining bone quantity in the mandibular ramus. Methods: Twenty-nine cadaveric mandibles marked bilaterally with three fiducial markers were imaged using both CBCT and digital orthopantomography. After sectioning, four cross sectional measurements were made on the specimens and on the CBCT images. Two corresponding linear measurements were made on the orthopantomograms. Statistical analysis was used to compare the CBCT and orthopantomogram measurements with measurements from the anatomic specimens. Results: CBCT measurements were found to significantly differ from those made on the anatomic specimens (P<0.05). Linear measurements from the orthopantomograms varied by 15.9 percent compared to the anatomic specimens. Conclusion: CBCT and orthopantomogram measurements were significantly different from those of the anatomic specimens suggesting inaccuracies in the radiographic technology or a lack of precision in landmark identification.

APLF (Aprataxin and Polynucleotide kinase/phosphatase-Like Factor) is an FHA (forkhead-associated)-domain-containing nuclear protein that facilitates the repair of single-strand and double-strand breaks (SSBs and DSBs). Specifically, the APLF-FHA domain mediates interactions with XRCC1 and XRCC4, factors involved in SSB and DSB repair, respectively. A novel substitution was identified in pancreatic cancer patients, where a conserved histidine was substituted to leucine (H42L) in the APLF-FHA domain. The functional and biological characterization of this “variant of unknown significance” was investigated. This thesis shows that the H42L substitution affects APLF phosphorylation, impairs APLF retention at laser-induced DNA breaks, disrupts protein-binding to XRCC1 but not to XRCC4, and reduces cell survival following irradiation and etoposide exposure. Collectively, these data suggest that the H42L substitution impacts APLF participation in global DNA damage repair. The findings from this work could provide insight into the role of APLF in genomic integrity and, moreover, in cancer predisposition.

APLF (Aprataxin and Polynucleotide kinase/phosphatase-Like Factor) is an FHA (forkhead-associated)-domain-containing nuclear protein that facilitates the repair of single-strand and double-strand breaks (SSBs and DSBs). Specifically, the APLF-FHA domain mediates interactions with XRCC1 and XRCC4, factors involved in SSB and DSB repair, respectively. A novel substitution was identified in pancreatic cancer patients, where a conserved histidine was substituted to leucine (H42L) in the APLF-FHA domain. The functional and biological characterization of this “variant of unknown significance” was investigated. This thesis shows that the H42L substitution affects APLF phosphorylation, impairs APLF retention at laser-induced DNA breaks, disrupts protein-binding to XRCC1 but not to XRCC4, and reduces cell survival following irradiation and etoposide exposure. Collectively, these data suggest that the H42L substitution impacts APLF participation in global DNA damage repair. The findings from this work could provide insight into the role of APLF in genomic integrity and, moreover, in cancer predisposition.

This project examined the effect of splinting adjacent dental implants of different designs [Sintered porous surface (SPS), Threaded dual acid etched (THR-DAE), Threaded sand-blasted large grit acid etched (THR-SLA)] placed in posterior sites and compared two intraoral radiographic techniques to assess peri-implant crestal bone levels. 799 implants located in 345 patients qualified and were examined retrospectively using an inclusion criteria of a minimum one year in function and having a recent periapical (PA) and vertical bitewing radiograph (vBW). SPS had less bone loss than THR-DAA and THR-SLA. PA views showed approximately 0.10 mm (Range 0.07-0.40 mm) less crestal bone loss than vBW (P<0.01). Splinted implants had more crestal bone loss (0.30 mm; range 0.16-0.38 mm) than non-splinted implants. 49 implants failed with an overall failure rate of 4.9% with a significant difference in the failure rate of SPS and THR-DAA (P<0.0005) and in the time to fail (P<0.0036).

Autism and associated gene mutations can be studied with genetic mouse models. Magnetic Resonance Imaging (MRI) of these animal models quantifies the impact of genetics on brain morphology. Using MRI, three genetic mouse models of autism were imaged: Neuroligin 3 R451C knock-in, Methyl-CpG binding protein-2 308-truncation and Integrin β-3 knock-out. Morphological differences were identified using a newly developed MRI mouse cerebellum atlas. The results show all three genes to alter cerebellar anatomy. Each studied gene affected a unique set of cerebellar structures. I hypothesize that the results and known behavioural phenotypes of the models are linked, with anatomy contributing to specific behaviours. In the future work section, a surface-based analysis method is presented to investigate the variance in cerebellum foliation across disease models and inbred strains. This work shows that autism risk-genes alter distinct regions of the cerebellum.

This project examined the effect of splinting adjacent dental implants of different designs [Sintered porous surface (SPS), Threaded dual acid etched (THR-DAE), Threaded sand-blasted large grit acid etched (THR-SLA)] placed in posterior sites and compared two intraoral radiographic techniques to assess peri-implant crestal bone levels. 799 implants located in 345 patients qualified and were examined retrospectively using an inclusion criteria of a minimum one year in function and having a recent periapical (PA) and vertical bitewing radiograph (vBW). SPS had less bone loss than THR-DAA and THR-SLA. PA views showed approximately 0.10 mm (Range 0.07-0.40 mm) less crestal bone loss than vBW (P<0.01). Splinted implants had more crestal bone loss (0.30 mm; range 0.16-0.38 mm) than non-splinted implants. 49 implants failed with an overall failure rate of 4.9% with a significant difference in the failure rate of SPS and THR-DAA (P<0.0005) and in the time to fail (P<0.0036).

Autism and associated gene mutations can be studied with genetic mouse models. Magnetic Resonance Imaging (MRI) of these animal models quantifies the impact of genetics on brain morphology. Using MRI, three genetic mouse models of autism were imaged: Neuroligin 3 R451C knock-in, Methyl-CpG binding protein-2 308-truncation and Integrin β-3 knock-out. Morphological differences were identified using a newly developed MRI mouse cerebellum atlas. The results show all three genes to alter cerebellar anatomy. Each studied gene affected a unique set of cerebellar structures. I hypothesize that the results and known behavioural phenotypes of the models are linked, with anatomy contributing to specific behaviours. In the future work section, a surface-based analysis method is presented to investigate the variance in cerebellum foliation across disease models and inbred strains. This work shows that autism risk-genes alter distinct regions of the cerebellum.