The Artificial Intelligence in Biomedical Imaging Laboratory (AIBIL) focuses on development and application of machine learning and image analysis methods towards establishing imaging signatures of various diseases and disorders. These imaging signatures, based on both conventional machine learning and deep learning methods, aim to serve as individualized biomarkers for precision diagnosis and predictive modeling. The lab’s long standing record on use of machine learning in neuroimaging includes imaging signatures of brain aging, Alzheimer’s Disease, schizophrenia, and brain cancer, as well as of functional connectivity. Some of the current challenges and targets include dissecting disease heterogeneity using semi-supervised learning methods, establishing radiogenomic markers of genetic mutations, and relating imaging and pathology data.
The Computational Biomarker Imaging Group (CBIG) is a research group in the Radiology department at the University of Pennsylvania. Our goal is to act as a translational catalyst between computation imaging science and clinical cancer research by integrating image analysis, pattern recognition and artificial intelligence in clinically relevant cancer imaging applications. Our research program specifically focuses on investigating the role of imaging as a biomarker for improving personalized clinical decision-making for cancer screening, prognosis, and treatment. Towards this end, we are developing innovative methodologies to analyze multi-modality imaging data, as well as to integrate imaging with genomic, molecular, histopathologic, clinical, and epidemiologic data into high-dimensional integrated-diagnostic models to predict patient outcomes. CBIG collaborates with faculty within Radiology, the Abramson Cancer Center (ACC), the Institute for Biomedical Informatics (IBI), the Institute for the Translational Medicine and Therapeutics (ITMAT), and the Center for Clinical Epidemiology and Biostatistics (CCEB). Affiliated faculty includes experts in informatics, medical physics, genetics, pathology, oncology, biostatistics, epidemiology, and primary care. Our vision is to foster a vibrant, collaborative research environment in which basic scientists, graduate students, postdocs, medical trainees and clinical investigators will have the opportunity to work closely together to accelerate the translation biomedical imaging research towards precision-medicine approaches for improving the care of patients diagnosed with cancer.
The Center for Neuroimaging In Psychiatry (CNIP) at the University of Pennsylvania is at the forefront of brain imaging in psychiatric research. The CNIP team acquires, processes and analyzes data in numerous modalities including: high (3T) and ultra high (7T) field Blood Oxygen Level Dependent functional Magnetic Resonance Imaging (BOLD fMRI), Diffusion Tensor Imaging (DTI), perfusion, Magnetic Resonance Spectroscopy (MRS), and Positron Emission Tomography (PET). This neuroimaging data is complemented with additional behavioral, phsyiological, and genetic information. All neuroimaging data is validated and organized into a customized XNAT database that archives and facilitates data processing. The CNIP is led by expert investigators collaborating with leading psychiatric researchers across the world with the goal of developing and implementing multi-modal techniques to advance the field of neuroimaging.
Path BioResource was created by the Department of Pathology and Laboratory Medicine to provide administrative support to the departmentally-based shared resource laboratories. Through partnering with the School of Medicine administration, as well as Centers and Institutes within the University community, Path BioResource seeks to provide all investigators access to high quality, cost-effective advanced technology services as well as the scientific expertise to use these technologies effectively in their research efforts.
The Penn Statistical Imaging and Visualization Endeavor (PennSIVE) consists of a group of statisticians studying etiology and clinical practice through medical imaging. Based in the Center for Clinical Epidemiology and Biostatistics at the Perelman School of Medicine, we work closely with collaborators at the University of Pennsylvania and nationally in medical specialties including neurology, neurosurgery, and radiology. Our primary goals include 1)Developing robust and generalizable statistical methods for the analysis of multimodal biomedical imaging data. 2)Integrating complex medical imaging data and other biomarkers to study health. 3)Building clinical tools for the assessment of disease diagnosis, progression, and prognosis through cross-sectional and longitudinal imaging studies.
The Section for Biomedical Image Analysis (SBIA) is devoted to the development of computer-based image analysis methods, and their application to a wide variety of clinical research studies. Image analysis methodologies include image registration, segmentation, population-based statistical analysis, biophysical modeling of anatomical deformations, and high-dimensional pattern classification. Clinical research studies span a variety of clinical areas and organs, and are performed within a wide network of collaborations from within and outside Penn. They include brain diseases such as Alzheimer's and schizophrenia, evaluation of treatment effects in large clinical trials, diagnosis of cardiac diseases, and diagnosis prostate, breast and brain cancer. SBIA also performs small animal imaging research aiming to understand brain development in mouse models.
The Wang lab focuses on Alzheimer’s disease and other neurodegenerative disorders, aging, and psychiatric disorders including autism and bipolar disorder. Ongoing projects in the lab can be divided into the following three main directions: 1)Genetics and genomics of Alzheimer’s disease and other neurodegenerative disorders 2)Informatics and algorithm development for genome-scale experiments 3)Biomarker development for aging and neurodegenerative disorders. The Wang lab developed tools for the analysis of several large-scale genome-wide association (GWA) studies, which led to findings of new risk genes for frontotemporal dementia,progressive supranuclear palsy (PSP), and late-onset Alzheimer’s disease. The lab actively develops novel algorithms and computer programs that analyze GWA, DNA-seq and RNA-seq studies. The lab is also involved in biomarker development for aging and for Alzheimer's disease.