Shape-Constrained Multi-Atlas Segmentation of Spleen in CT.

Spleen segmentation on clinically acquired CT data is a challenging problem given the complicity and variability of abdominal anatomy. Multi-atlas segmentation is a potential method for robust estimation of spleen segmentations, but can be negatively impacted by registration errors. Although labeled atlases explicitly capture information related to feasible organ shapes, multi-atlas methods have largely used this information implicitly through registration. We propose to integrate a level set shape model into the traditional label fusion framework to create a shape-constrained multi-atlas segmentation framework. Briefly, we (1) adapt two alternative atlas-to-target registrations to obtain the loose bounds on the inner and outer boundaries of the spleen shape, (2) project the fusion estimate to registered shape models, and (3) convert the projected shape into shape priors. With the constraint of the shape prior, our proposed method offers a statistically significant improvement in spleen labeling accuracy with an increase in DSC by 0.06, a decrease in symmetric mean surface distance by 4.01 mm, and a decrease in symmetric Hausdorff surface distance by 23.21 mm when compared to a locally weighted vote (LWV) method.

Variables influencing the neural correlates of perceived risk of physical harm.

Many human activities involve a risk of physical harm. However, not much is known about the specific brain regions involved in decision making regarding these risks. To explore the neural correlates of risk perception for physical harms, 19 participants took part in an event-related fMRI study while rating risky activities. The scenarios varied in level of potential harm (e.g., paralysis vs. stubbed toe), likelihood of injury (e.g., 1 chance in 100 vs. 1 chance in 1,000), and format (frequency vs. probability). Networks of brain regions were responsive to different aspects of risk information. Cortical language- processing areas, the middle temporal gyrus, and a region around the bed nucleus of stria terminalis responded more strongly to high- harm conditions. Prefrontal areas, along with subcortical ventral striatum, responded preferentially to high- likelihood conditions. Participants rated identical risks to be greater when information was presented in frequency format rather than probability format. These findings indicate that risk assessments for physical harm engage a broad network of brain regions that are sensitive to the severity of harm, the likelihood of risk, and the framing of risk information.