[ABE-L] Seminário DEST/UFMG em 18/08/2023

[Marcos Prates]

Caros,

Na próxima sexta (18 de Agosto, às 13:30h) retomaremos o ciclo de
Seminários do Departamento de Estatística da UFMG, com a apresentação do
prof. Shariq Mohammed do Departamento de Bioestatística da
Boston University.

Shariq received his bachelor’s degree in mathematics with honors from the
Indian Statistical Institute (ISI), Bangalore, India, followed by an M.Sc.
in applications of mathematics from Chennai Mathematical Institute (CMI),
Chennai, India. He obtained his PhD in Statistics from the University of
Connecticut (UConn), Storrs in August 2018 before joining the Department of
Biostatistics and the Department of Computational Medicine & Bioinformatics
at the University of Michigan, Ann Arbor, as a post-doctoral fellow.
Currently, he is an Assistant Professor in the Biostatistics department
from Boston University. His research interests reside in Bayesian modeling,
variable selection, and spatial statistics with applications to medical
imaging data. Shariq’s work has focused on the development, implementation,
and use of sound statistical methodology coupled with domain knowledge to
extract information from medical images that can be used for disease
classification and disease prediction.

Título: Layered Variable Selection for Multivariate Bayesian Regression: A
Case Study in Imaging-Genomics
Abstract: We propose a statistical framework to integrate radiological
magnetic resonance imaging (MRI) and genomic data to identify the
underlying radiogenomic associations in lower grade gliomas (LGG). We
devise a novel imaging phenotype by dividing the tumor region into
concentric spherical layers that mimics the tumor evolution process. MRI
data within each layer is represented by voxel-intensity-based probability
density functions which capture the complete information about tumor
heterogeneity. Under a Riemannian-geometric framework these densities are
mapped to a vector of principal component scores which act as imaging
phenotypes. Subsequently, we build Bayesian variable selection models for
each layer with the imaging phenotypes as the response and the genomic
markers as predictors. Our novel hierarchical prior formulation
incorporates the interior-to-exterior structure of the layers, and the
correlation between the genomic markers. We employ a computationally
efficient Expectation-Maximization-based strategy for estimation. With a
focus on the cancer driver genes in LGG, we discuss some biologically
relevant findings.

O seminário será presencial na sala 2076 do ICEx/UFMG.

Att,

Marcos Prates
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://lists.ime.usp.br/pipermail/abe/attachments/20230813/df34e5a0/attachment.htm>