BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//BioMedIA - ECPv6.15.11//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:BioMedIA
X-ORIGINAL-URL:https://biomedia.doc.ic.ac.uk
X-WR-CALDESC:Events for BioMedIA
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20150101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20160518T120000
DTEND;TZID=UTC:20160518T130000
DTSTAMP:20260417T012014
CREATED:20160427T184017Z
LAST-MODIFIED:20160512T210704Z
UID:2743-1463572800-1463576400@biomedia.doc.ic.ac.uk
SUMMARY:BioMedIA Talk: Enzo Ferrante
DESCRIPTION:Title: Graph-based deformable registration: slice-to-volume mapping and context specific methods \nAbstract: Image registration methods\, which aim at aligning two or more images into one coordinate system\, are among the oldest and most widely used algorithms in computer vision. A particular type of registration algorithm\, known as graph-based deformable registration\, has become popular during the last decade given its robustness\, scalability\, efficiency and theoretical simplicity. In this thesis\, we extend this flexible framework to new scenarios\, and propose novel methodological contributions. \nWe start by formulating\, within the graph-based deformable registration framework\, the challenging slice-to-volume registration problem. We introduce a scalable\, modular and flexible formulation accommodating low-rank and high order terms\, which simultaneously selects the plane and estimates the in-plane deformation through a single shot optimization approach. The other two contributions included in this thesis are related to how semantic information can be encompassed within the registration process. Currently\, most of the methods rely on a single metric function explaining the similarity between the source and target images. We argue that incorporating semantic information to guide the registration process will further improve the accuracy of the results\, particularly in the presence of semantic labels making the registration a domain specific problem.
URL:https://biomedia.doc.ic.ac.uk/event/biomedia-talk-enzo-ferrante/
LOCATION:Huxley 144
CATEGORIES:Talks & Seminars
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20160519T160000
DTEND;TZID=UTC:20160519T170000
DTSTAMP:20260417T012014
CREATED:20160427T184136Z
LAST-MODIFIED:20160512T210610Z
UID:2744-1463673600-1463677200@biomedia.doc.ic.ac.uk
SUMMARY:BioMedIA Talk: Ioannis Katramados (COSMONiO)
DESCRIPTION:Title: The challenges of developing an active deep-learning platform \nAbstract:\nDeep learning is usually linked to Big Data. However\, there are several scientific\, engineering and medical imaging problems with limited data available (e.g. rare medical conditions). Can Deep Learning prove a useful tool in such cases? COSMONiO is designing NOUS\, an active deep learning platform that aims to make high-accuracy predictions using significantly smaller training datasets. NOUS aims to allow experts from any field to train neural networks without any prior experience. We will discuss the main challenges and how we address them.
URL:https://biomedia.doc.ic.ac.uk/event/biomedia-talk-ioannis-katramados-cosmonio/
LOCATION:Huxley 144
CATEGORIES:Talks & Seminars
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20160524T120000
DTEND;TZID=UTC:20160524T130000
DTSTAMP:20260417T012014
CREATED:20160520T171016Z
LAST-MODIFIED:20160520T171212Z
UID:2807-1464091200-1464094800@biomedia.doc.ic.ac.uk
SUMMARY:BioMedIA Talk: Leo Grady (HeartFlow)
DESCRIPTION:Personalized Blood Flow Simulation from an Image-Derived Model: Changing the Paradigm for Cardiovascular Diagnostics \nAbstract: Coronary heart disease is the leading cause of mortality worldwide\, accounting for 1/3 of all global deaths.  Treatment of stable coronary heart disease is typically performed by medication/lifestyle for a lower disease burden or PCI (stenting) for a greater disease  burden.  The choice between these treatments is best determined by an invasive diagnostic test that measures blood flow through a diseased area.  Unfortunately\, this invasive diagnostic test is expensive\, dangerous and usually finds a lower disease burden.  We are working to change the diagnostics paradigm with blood flow simulation in a personalized heart model that is derived from cardiac CT angiography images.  This simulation-based diagnostic is the first clinically available diagnostic that utilizes personalized simulation and is much safer and more comfortable for the patient as well as less expensive.  Our diagnostic depends on a hyperaccurate image segmentation of the coronary arteries\, physiological modeling and accurate computational fluid dynamics.  In this talk I will discuss the algorithms that drive this technology\, the machine learning that we’re doing with our database of segmented images and personalized hemodynamics\, and the successful clinical trials that have proven the diagnostic accuracy and benefit to patients. \n  \nBio: Leo Grady is the Vice President of Research and Development at HeartFlow since 2012.  Prior to joining HeartFlow\, he worked at Siemens Corporate Research for nine years as a Principal Research Scientist following his PhD at Boston University.  His work has focused on a range of computer vision and medical imaging applications in image segmentation and machine learning.  He has written two books on computer vision and data analysis using graph theory\, is an editor of several journals in computer vision and was recently inducted as a Fellow of the American Institute of Medical and Biomedical Engineers.
URL:https://biomedia.doc.ic.ac.uk/event/biomedia-talk-leo-grady-heartflow/
LOCATION:Huxley 144
CATEGORIES:Talks & Seminars
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20160525T120000
DTEND;TZID=UTC:20160525T130000
DTSTAMP:20260417T012014
CREATED:20160520T171139Z
LAST-MODIFIED:20160520T171139Z
UID:2809-1464177600-1464181200@biomedia.doc.ic.ac.uk
SUMMARY:BioMedIA Talk: Archontis Giannakidis (Royal Brompton)
DESCRIPTION:Improving clinical application of cardiac diffusion tensor MRI\nAbstract  Magnetic resonance diffusion tensor imaging (MRDTI)\, also known as DTI\, has emerged as a powerful non-invasive tool for mapping the orientation-dependent microanatomical organization of fibrous organs such as brain white matter and myocardium. To do so\, it elegantly relates the self-diffusion of water molecules that undergo Brownian motion to proton spin relaxation MR signals. In this talk it will be shown how DTI can shed some light on the left-ventricular micro-structural remodeling following hypertensive disease. A framework for the voxelwise registration-based characterization of cardiac disease will be considered. Population and longitudinal studies may benefit from such a scheme. Results from a comparison study will be presented juxtaposing the performance of three tensor distance functions. The selection of a tensor distance function resides in the foundation of the tensor-variate framework\, critically affecting many operations in the six-dimensional space of diffusion tensors. This seminar will end by discussing how sparsifying transforms can be used in conjunction with compressive sensing reconstruction to shorten acquisition times.
URL:https://biomedia.doc.ic.ac.uk/event/biomedia-talk-archontis-giannakidis-royal-brompton/
LOCATION:Huxley 144
CATEGORIES:Talks & Seminars
END:VEVENT
END:VCALENDAR