FSUMATH

This Week in Mathematics

TWIM RSS Feed
<< Back to Current Week [2019-01-13 - 2019-01-19] >> Beyond Next Week [2019-01-26+]
<< View Previously Scheduled Events
Next Week [Jan 20, 2019 - Jan 26, 2019]
January
S M T W R F S
  12345
6789101112
13141516171819
20212223242526
2728293031  
Entries for this week: 6
Monday January 21, 2019

Analysis & PDE
Time: 2:30-3:30 Room: 104 LOV
Wednesday January 23, 2019

Departmental Tea Time
C is for cookie, and shorthand for C[0,1] w/the sup norm
Time: 3: Room: 204 LOV
Biomathematics Seminar
    - Carolyn Eady, FSU
Time: 3:35 pm Room: LOV 200
Mathematics Colloquium [url]
Computational Neurology and Translational Modeling of Brain Disorders
    - Pedro Maia, UC San Francisco
Time: 3:35 pm Room: LOV 101
Abstract/Desc: There is an ample need for mathematical and computational models in neurology. Traumatic brain injuries and neurodegenerative diseases affect millions of people every year, and in all these disorders, pathologies occur across multiple spatial scales: at a micro scale, the accumulation of misfolded proteins lead to the neuron's axon to suffer traumatic injury or demyelination. As injured neurons fail to transmit electrical information, neuronal networks at a mesoscale lose their specialized functionality. At a whole-brain level, one may observe large-scale network disruption that will ultimately lead to measurable cognitive deficits. The emerging field of computational neurology provides an important window of opportunity for modeling of complex biophysical phenomena, for scientific computing, for understanding functionality disruption in neural networks, and for applying machine-learning methods for diagnosis and personalized medicine. In this talk, I will illustrate some of our latest results across different spatial scales spanning a broad array of mathematical techniques.
Friday January 25, 2019

Colloquium Tea
Time: 3:00 pm Room: 204 LOV
Mathematics Colloquium
Graph Regularizations for the EEG Inverse Problem
    - Jing Qin, Montana State
Time: 3:35 pm Room: LOV 101
Abstract/Desc: Regularization plays an important role in solving inverse problems in a wide spectrum of applications. In particular, regularization techniques for graph-structured datasets have a great potential to revolutionize imaging technologies such as Electroencephalogram (EEG). Estimation of the locations of brain sources from the EEG data, known as source localization, is a challenging ill-posed inverse problem. In this talk, we investigate several EEG source localization methods based on various spatial and temporal graph regularizations, including graph total generalized variation, graph fractional-order total variation, and temporal graph regularization. Numerical results have shown that the proposed methods localize source extents more effectively than the benchmark methods.

Problems? Email webmaster@math.fsu.edu.