Penn Image Computing & Science Lab

Biomechanics

Utilizing various image registration techniques in conjunction with traditional and novel medical imagery, we derive biomechanical properties of various anatomies. Such kinematic quantities allow for quantitative assessment of tissue viability.

Diffusion Imaging

The diffusion imaging theme exists to develop novel methodologies for the analysis of tissue microstructure with diffusion MR, to explore clinical hypotheses using the diffusion modality, and to provide expertise and high-quality software for the analysis of diffusion images in conjunction with other imaging modalities.

Shape Analysis

The Shape Analysis Working Group is searching for best ways to leverage geometrical information in statistical morphometry and analysis of multimodality imaging data, including fMRI and diffusion imaging.

Segmentation

Identification of anatomical structures within medical images is an essential precursor to subsequent analysis. The segmentation techniques developed at PICSL include those that have been applied to a wide range of segmentation problems, including three-tissue brain segmentation, atlas-based segmentation, shape-based segmentation of subcortical structures such as the hippocampus and others.

Registration

Open Source Software

The OSIRiS group embodies PICSL's philosophy that playing a lead role within the academic image research community entails encouragement of the translational aspect of our work in making software available to collaborators and the research community at large.

Hippocampus SPM

We are developing a new class of techniques that focus statistical analysis of functional neuroimaging data on specific structures such as the hippocampus, using the shape of the structures as the guide by which to combine information across subjects in ways that is more meaningful, sensitive and specific than in whole-brain normalization.

Cardiac Medial Modeling

The aims of this collaborative project with the Computational Imaging Lab at the Pompeu Fabra University are to use a medial model of the myocardium to generate stronger shape priors for segmentation, richer features for shape analysis and a shape-based coordinate parameterization for the analysis of heart wall dynamics.

Continuous Medial Representation

This project involves further theoretical development of the continuous medial representation, which provides a rich geometric shape description for anatomical structures and forms the underlying framework of some of PICSL's morphometry and functional imaging analysis projects.

Traumatic Brain Injury

This collaboration with the Moss Rehabilitation Research Institute at the Albert Einstein Healthcare Network involves the development of methologies for examining both structural and connective properties in the brains of TBI survivors.

Autopsy Brain Imaging

This collaborative project involves collecting a database of postmortem brain scans and histological sections that will serve as an invaluable source of anatomical information for building atlases of brain anatomy, learning the relationships between molecular and morphological phenotypes of neurodegenerative diseases, as well as validating image analysis methodology.

NNC/CfN Morphology Core

This core provides technical and scientific support to the Penn Center for Functional Neuroimaging (CfN) in the areas of morphometric analysis, statistics, advanced normalization and template generation for multi-subject studies, and visualization of results.

Modeling Complex Axon Fiber Architecture

The multi-diffusion model reveals anatomically plausible axon fiber orientations for both single and multiple fiber voxels. The fiber crossings and bifurcations can be solved by processing high angular DW-MRI data.