Faculty Bibliography

Currently, there is a lack of computational methods for the evaluation of mild traumatic brain injury (mTBI) from magnetic resonance imaging (MRI). Further, the development of automated analyses has been hindered by the subtle nature of mTBI abnormalities, which appear as low contrast MR regions. This paper proposes an approach that is able to detect mTBI lesions by combining both the high-level context and low-level visual information. The contextual model estimates the progression of the disease using subject information, such as the time since injury and the knowledge about the location of mTBI. The visual model utilizes texture features in MRI along with a probabilistic support vector machine to maximize the discrimination in unimodal MR images. These two models are fused to obtain a final estimate of the locations of the mTBI lesion. The models are tested using a novel rodent model of repeated mTBI dataset. The experimental results demonstrate that the fusion of both contextual and visual textural features outperforms other state-of-the-art approaches. Clinically, our approach has the potential to benefit both clinicians by speeding diagnosis and patients by improving clinical care.

Repetitive mild traumatic brain injury (rmTBI) is an important medical concern for active sports and military personnel. Multiple mild injuries may exacerbate tissue damage resulting in cumulative brain injury and poor functional recovery. In the present study, we investigated the time course of brain vulnerability to rmTBI in a rat model of mild cortical controlled impact. An initial mild injury was followed by a second injury unilaterally at an interval of 1, 3, or 7 days. RmTBI animals were compared to single mTBI and sham treated animals. Neuropathology was assessed using multi-modal magnetic resonance imaging (MRI), followed by ex vivo tissue immunohistochemistry. Neurological and behavioral outcomes were evaluated in a subset of animals receiving rmTBI 3 days apart and shams. RmTBI 1 or 3 days apart but not 7 days apart revealed significantly exacerbated MRI-definable lesion volumes compared to single mTBI and shams. Increases in cortical tissue damage, extravascular iron and glial activation assessed by histology/immunohistochemistry correlated with in vivo MRI findings where shorter intervals (1 or 3 days apart) resulted in greater tissue pathology. There were no neurological deficits associated with rmTBI 3 day animals. At 1 mo post-injury, animals with rmTBI 3 days apart showed reduced exploratory behaviors and subtle spatial learning memory impairments were observed. Collectively, our findings suggest that the mildly-impacted brain is more vulnerable to repetitive injury when delivered within 3 days following initial mTBI.

Implantation of an endometriotic lesion within a pelvic or abdominal wall scar is an uncommon but well-described condition that may be the underlying cause of acute or chronic recurrent abdominal or pelvic pain, especially after cesarean section. Radiologists may not consider scar endometriosis when it is encountered at cross-sectional imaging. Cesarean section scars are the most common site of extraovarian or extrauterine endometriosis. The condition also has been identified in other uterine surgery-related scars and in the skin, subcutaneous tissues, and abdominal and pelvic wall musculature adjacent to these scars. The most plausible cause of scar endometriosis is implantation of endometrial stem cells at the surgical site at the time of uterine surgery. Patients with scar endometriosis may be asymptomatic or present with cyclical pain corresponding to the menstrual cycle. Cross-sectional imaging findings vary from the nonspecific to those suggestive of the diagnosis when combined with clinical history. In particular, the presence of blood products in an anterior abdominal wall mass at magnetic resonance (MR) imaging with no other explanation is strongly suggestive of scar endometriosis. Ultrasonography, computed tomography, and MR imaging may be used to depict an endometriotic lesion, exclude endometriosis, or provide evidence for an alternative diagnosis.

Mild traumatic brain injury (mTBI) has become an increasing public health concern as subsequent injuries can exacerbate existing neuropathology and result in neurological deficits. This study investigated the temporal development of cortical lesions using magnetic resonance imaging (MRI) to assess two mTBIs delivered to opposite cortical hemispheres. The controlled cortical impact model was used to produce an initial mTBI on the right cortex followed by a second injury induced on the left cortex at 3 (rmTBI 3d) or 7 (rmTBI 7d) days later. Histogram analysis was combined with a novel semi-automated computational approach to perform a voxel-wise examination of extravascular blood and edema volumes within the lesion. Examination of lesion volume 1d post last injury revealed increased tissue abnormalities within rmTBI 7d animals compared to other groups, particularly at the site of the second impact. Histogram analysis of lesion T2 values suggested increased edematous tissue within the rmTBI 3d group and elevated blood deposition in the rm TBI 7d animals. Further quantification of lesion composition for blood and edema containing voxels supported our histogram findings, with increased edema at the site of second impact in rmTBI 3d animals and elevated blood deposition in the rmTBI 7d group at the site of the first injury. Histological measurements revealed spatial overlap of regions containing blood deposition and microglial activation within the cortices of all animals. In conclusion, our findings suggest that there is a window of tissue vulnerability where a second distant mTBI, induced 7d after an initial injury, exacerbates tissue abnormalities consistent with hemorrhagic progression.

Pediatric ALK-positive anaplastic large cell lymphoma (ALK+ ALCL) is usually associated with a favorable prognosis. ALK+ ALCL associated with a leukemic phase is uncommon, but has been associated with an aggressive clinical course and unfavorable prognosis. Overexpression of c-myc has been shown to be a consistent finding in ALK+, but not ALK-negative ALCL (ALK- ALCL), and the c-myc gene is considered a downstream target of deregulated ALK signaling. We describe a pediatric ALK+ ALCL with a leukemic phase at relapse. Similar to other rare cases described in the literature, it followed an aggressive clinical course despite multiple regimens of chemotherapy and bone marrow transplantation. Lymphoma cells showed aberrant ALK expression and c-myc overexpression. In addition to the characteristic t(2;5)(p23;q35) translocation, a t(3;8)(q26.2;q24) translocation was also present, and c-myc gene rearrangement was confirmed by FISH analysis. The findings in this case demonstrate the association of peripheral blood leukemic involvement and aggressive clinical course, and suggest that other factors, such as c-myc rearrangement, may be responsible for the aggressive clinical behavior in ALK+ ALCL.

Carcinosarcomas are very uncommon tumors, which are comprised of both malignant epithelial and mesenchymal elements. They occur most commonly in the head and neck, respiratory tract, and female reproductive organs. In the gastrointestinal tract, they are most often found in the oropharynx, esophagus, and, to a lesser extent, in the stomach. Carcinosarcomas rarely originate from the colon, but when they do, they are extremely aggressive malignancies. We report the radiologic and pathologic findings of a patient with a carcinosarcoma believed to have arisen from the colon and which involved the adjacent mesentery and omentum.

Juvenile granulosa cell tumor (GCT) of the ovary is a rare neoplasm occurring in premenarchal girls and young women. Juvenile GCT that occurs in premenarchal girls usually produces sexual precocity as a consequence of estrogen secretion. Juvenile GCTs are more likely to grow to a relatively large size with a much smaller likelihood of peritoneal spread, unlike their counterpart, epithelial ovarian neoplasms. We report the radiology and pathology of a patient with juvenile GCT and review the literature of this rare tumor.

RATIONALE AND OBJECTIVES: The purpose of this study was to determine whether a simple rapid blood test can obviate computed tomography (CT) in a sizable percentage of patients suspected of having pulmonary embolism, based on the hypothesis that negative D-dimer results could eliminate any further search for pulmonary embolism. MATERIALS AND METHODS: At the authors' institution, 2,121 sequential patients underwent a whole-blood antibody agglutination test for cross-linked fibrin degradation products (D-dimer). Of these patients, 844 had positive test results and were not further considered. A retrospective review included reports of all multisection combined CT venographic and pulmonary angiographic studies obtained within 48 hours of the D-dimer assay for the 1,277 patients with negative D-dimer results; 229 (18%) of these 1,277 patients underwent combined CT venography and pulmonary angiography, usually within 24 hours. RESULTS: Retrospective review of the imaging examinations that were discrepant with the D-dimer results revealed only three false-negative D-dimer results. Of the 229 patients in whom combined CT venography and pulmonary angiography was performed for suspected pulmonary embolism, 226 (98.7%) had no evidence of acute pulmonary embolism or deep venous thrombosis. The negative predictive value of a negative D-dimer result was therefore 98.7% (confidence interval, 96.2%-99.7%). CONCLUSION: The D-dimer assay is a simple rapid blood test that is sensitive to the presence of acute thrombosis. Very few patients with negative results have acute deep venous thrombosis or pulmonary embolism, with combined CT venography and pulmonary angiography used as the reference standard.