BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.5//EN TZID:America/Los_Angeles X-WR-TIMEZONE:America/Los_Angeles BEGIN:VEVENT UID:57@escience.washington.edu DTSTART;TZID=America/Los_Angeles:20231128T163000 DTEND;TZID=America/Los_Angeles:20231128T172000 DTSTAMP:20231106T174104Z URL:https://escience.washington.edu/events/uwdss-garyfallidis/ SUMMARY:UW Data Science Seminar: Eleftherios Garyfallidis DESCRIPTION:\n\nPlease join us for a UW Data Science Seminar on Tuesday\, November 28th from 4:30 to 5:20 p.m. PST. The seminar will feature Dr. Ele ftherios Garyfallidis\, an Associate Professor of Intelligent Systems Engi neering (ISE) at Indiana University. \n\n\nThis event will take place in the Physics/Astronomy Auditorium 102 (PAA A102) on the University of Washi ngton campus.\n\n \;\n"AI breakthroughs for pre-processing structural and diffusion MRI data"\nAbstract: During my talk I will focus on two nove l approaches for segmentation and denoising.\nSegmentation: Brain extracti on is a computational necessity for researchers using brain imaging data. However\, the complex structure of the interfaces between the brain\, meni nges and human skull have not allowed a highly robust solution to emerge. While previous methods have used machine learning with structural and geom etric priors in mind\, with the development of Deep Learning (DL)\, there has been an increase in Neural Network based methods. Most proposed DL mod els focus on improving the training data despite the clear gap between gro ups in the amount and quality of accessible training data between. We prop ose an architecture we call Efficient V-net with Additional Conditional Ra ndom Field Layers (EVAC+). EVAC+ has 3 major characteristics: (1) a smart augmentation strategy that improves training efficiency\, (2) a unique way of using a Conditional Random Fields Recurrent Layer that improves accura cy and (3) an additional loss function that fine-tunes the segmentation ou tput. We compare our model to state-of-the-art non-DL and DL methods.\nDen oising: Patch2Self (P2S)\, which performs self-supervised denoising of dMR I data using the statistical independence of noise\, has previously shown state-of-the-art results by performing a series of regression analyses on the so-called Casorati matrix. P2S however is resource intensive\, both in terms of running time and memory usage. This work exploits the redundancy imposed by P2S to alleviate its performance issues and inspect regions th at influence the noise disproportionately. Specifically\, this study makes a two-fold contribution: (1) We present Patch2Self2 (P2S2)\, a method tha t uses matrix sketching to perform self-supervised denoising. By solving a sub-problem on a smaller sub-space\, so called\, coreset\, we show how P2 S2 can yield a significant speedup in training time while using less memor y. (2) We show how the so-called statistical leverage scores can be used t o interpret the denoising of dMRI data\, a process that was traditionally treated as a black-box. Our experiments are conducted on simulated and rea l data and clearly demonstrate that P2S2 does not lead to any loss in deno ising quality\, while providing significant speedup and improved memory us age by training on only a small fraction of the data.\nBiography: Dr. Gary fallidis holds the position of Associate Professor of Intelligent Systems Engineering (ISE) at Indiana University (IU). Prof. Garyfallidis works on the interface between machine learning\, medical imaging and engineering v isualization. He is the inventor of multiple ground breaking algorithms in cluding QuickBundles. QuickBundles was the first fast and unsupervised alg orithm in neuroimaging for grouping tractographies using streamlines. Prof . Garyfallidis is the inventor of SLR. SLR is the most accurate method for affinely registering bundles or tractograms. His research has been fundam ental in understanding the challenges of brain tractography. Due to a meth od called RecoBundles\, in 2015 Garyfallidis enabled the evaluation of tra ctographies in data with distortions. Prof. Garyfallidis is the founder an d lead engineer of DIPY. The pioneering work that Dr. Garyfallidis started and is today championed by his graduate students. See for example his lab s work on Patch2Self denoising\, EVAC+ and Bundle Analytics (BUAN). Prof. Garyfallidis is organizing yearly workshops (see DIPY workshops) to train faculty and students to use the latest methods in neuroimaging.\n\n \; \nThe UW Data Science Seminar is an annual lecture series at the Univers ity of Washington that hosts scholars working across applied areas of data science\, such as the sciences\, engineering\, humanities and arts along with methodological areas in data science\, such as computer science\, app lied math and statistics. Our presenters come from all domain fields and i nclude occasional external speakers from regional partners\, governmental agencies and industry.\nThe 2023-2024 seminars will be held in person\, an d are free and open to the public. LOCATION:Physics/Astronomy Auditorium A102\, 3910 15th Ave NE\, Seattle\, W A\, United States X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=3910 15th Ave NE\, Seattle\ , WA\, United States;X-APPLE-RADIUS=100;X-TITLE=Physics/Astronomy Auditori um A102:geo:0,0 END:VEVENT BEGIN:VTIMEZONE TZID:America/Los_Angeles X-LIC-LOCATION:America/Los_Angeles BEGIN:STANDARD DTSTART:20231105T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 TZNAME:PST END:STANDARD END:VTIMEZONE END:VCALENDAR