BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260610T091810EDT-5735ng0uiV@132.216.98.100
DTSTAMP:20260610T131810Z
DESCRIPTION:Speaker: Archana Venkataraman\, PhD\n\nJohn C. Malone Assistant
  Professor\, John Hopkins University\, USA\n\nTalk abstract\n\nDeep learni
 ng has disrupted nearly every major field of study from computer vision to
  genomics. The unrivaled power of these models offers a new and exciting w
 ay to study the intricacies of the brain via functional neuroimaging. On t
 he flipside\, deep networks lack interpretability and are notorious for ov
 erfitting on smaller datasets. My lab tackles these challenges through car
 eful problem formulation\, architectural design\, and model training. This
  talk will highlight two ongoing projects that demonstrate both the scient
 ific and translational potential of deep learning for fMRI. In the first p
 roject\, we develop a novel deep network architecture to integrate imaging
  and genetics data\, as guided by diagnosis. Our model consists of a coupl
 ed autoencoder and classifier. The encoder learns a non-linear subspace sh
 ared between the input data modalities. The classifier and the decoder act
  as regularizers to ensure that the low-dimensional space captures predict
 ive differences between patients and controls. We use a learnable dropout 
 layer to extract interpretable biomarkers from the data\, and our unique t
 raining strategy can easily accommodate missing data modalities across sub
 jects. In the second project\, we develop an end-to-end deep learning fram
 ework that uses dynamic functional connectivity to simultaneously localize
  the language and motor areas of the eloquent cortex in brain tumor patien
 ts. Our model leverages specialized convolutional filters that extract gra
 ph-based features from the dynamic connectivity matrices\, an LSTM attenti
 on network to weigh the relevant time points and multitask classification 
 to simultaneously localize different eloquent subsystems.\n\nThe Feindel V
 irtual Brain and Mind (VBM) Seminar Series\n\nAdvancing the vision of Dr. 
 William Feindel (1918–2014)\, former Director of the Neuro (1972–1984) and
  Founding Director of the McConnell Brain Imaging Centre (1984)\, to const
 antly bridge the clinical and research realms. The talks will highlight th
 e latest advances and discoveries in neuropsychology\, cognitive neuroscie
 nce\, and neuroimaging.\n\nSpeakers will include scientists from across Th
 e Neuro\, as well as colleagues and collaborators locally and from around 
 the world. The series is intended to provide a virtual forum for scientist
 s and trainees to continue to foster interdisciplinary exchanges on the me
 chanisms\, diagnosis and treatment of brain and cognitive disorders. If yo
 u\, or someone you know\, would be interested in presenting please contact
  us to secure a spot on the schedule.\n\nTo register: Eventbrite\n
DTSTART:20201125T210000Z
DTEND:20201125T220000Z
LOCATION:CA\, ZOOM
SUMMARY:Deep Learning for Multimodal and Dynamic Functional Neuroimaging
URL:https://www.mcgill.ca/cim/channels/event/deep-learning-multimodal-and-d
 ynamic-functional-neuroimaging-326399
END:VEVENT
END:VCALENDAR