BEGIN:VCALENDAR
VERSION:2.0
PRODID:Icfo
X-PUBLISHED-TTL:P1W
BEGIN:VEVENT
UID:69b7eadeb516f
DTSTART:20250305T110000Z
SEQUENCE:0
TRANSP:OPAQUE
DTEND:20250305T120000Z
LOCATION:Seminar Room
SUMMARY:ICFO | JESÚS PINEDA
CLASS:PUBLIC
DESCRIPTION:Geometric deep learning has revolutionized fields like social n
 etwork analysis\, molecular chemistry\, and neuroscience\, but its applica
 tion to microscopy data analysis remains a significant challenge. The hurd
 les stem not only from the scarcity of high-quality data but also from the
  intrinsic complexity and variability of microscopy datasets. This present
 ation introduces two groundbreaking geometric deep-learning frameworks des
 igned to overcome these barriers\, advancing the integration of graph neur
 al networks (GNNs) into microscopy and unlocking their full potential. Fir
 st\, we present MAGIK\, a cutting-edge framework for analyzing biological 
 system dynamics through time-lapse microscopy. Leveraging a graph neural n
 etwork augmented with attention-based mechanisms\, MAGIK processes object 
 features using geometric priors. This enables it to perform a range of tas
 ks\, from linking coordinates into trajectories to uncovering local and gl
 obal dynamic properties with unprecedented precision. Remarkably\, MAGIK e
 xcels under minimal data conditions\, maintaining exceptional performance 
 and robust generalization across diverse scenarios. Next\, we introduce MI
 RO\, a novel algorithm powered by recurrent graph neural networks. MIRO pr
 e-processes Single Molecule Localization (SML) datasets to enhance the eff
 iciency of conventional clustering methods. Its ability to handle clusters
  of varying shapes and scales enables more accurate and consistent analyse
 s across complex datasets. Furthermore\, MIRO&rsquo\;s single- and few-sho
 t learning capabilities allow it to generalize effortlessly across scenari
 os\, making it an efficient\, scalable\, and versatile tool for microscopy
  data analysis. Together\, MAGIK and MIRO address critical limitations in 
 microscopy data analysis\, offering innovative solutions for multi-scale d
 ata analysis and advancing the boundaries of what is currently achievable 
 with geometric deep learning in the field.
DTSTAMP:20260316T113454Z
END:VEVENT
END:VCALENDAR