BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20241027T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RDATE:20251026T030000
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.29340.field_data.0@www.diag.uniroma1.it
DTSTAMP:20260411T223826Z
CREATED:20250417T095711Z
DESCRIPTION:Short Course part of Seminars in AI and RoboticsNon-rigid image
  registration is one of the fundamental pillars of medical image analysis\
 , enabling the alignment of anatomical structures across subjects\, time p
 oints\, and imaging modalities. Among its various formulations\, diffeomor
 phic registration stands out for ensuring smooth and invertible deformatio
 ns that preserve anatomical topology. These properties are essential for a
 ccurately capturing complex morphological variations while maintaining the
  structural integrity of the anatomies being analyzed.Computational Anatom
 y is a recently emerging discipline that aims to model and analyze anatomi
 cal structures through mathematical and computational frameworks\, enablin
 g quantitative studies of shape variability\, disease progression\, and po
 pulation trends. A fundamental tool in this field is diffeomorphic image r
 egistration. Traditional methods\, such as Large Deformation Diffeomorphic
  Metric Mapping (LDDMM) and its Euler-Poincaré formulation (EPDiff-LDDMM)\
 , provide a solid theoretical foundation but often suffer from high comput
 ational costs. To mitigate these limitations\, faster approximations like 
 Stationary LDDMM\, diffeomorphic Demons\, and FLASH introduce notable effi
 ciency gains while maintaining registration fidelity. With the advent of d
 eep learning\, supervised and non-supervised data-driven methods such as Q
 uickSilver and VoxelMorph have reshaped the field\, offering rapid inferen
 ce while maintaining diffeomorphic constraints. More recently\, Implicit N
 eural Representations (INRs) and Neural Ordinary Differential Equations (N
 ODEs) have emerged as a promising paradigms\, bridging traditional physics
 -based models with modern deep-learning frameworks. This short course prov
 ides a structured journey through the evolution of diffeomorphic registrat
 ion and its application into Computational Anatomy\, covering key methodol
 ogies from classical optimization-based techniques to state-of-the-art dee
 p-learning approaches.Through theoretical insights and hands-on demonstrat
 ions\, participants will gain a comprehensive understanding of the LDDMM-v
 erse\, its challenges\, and its past and future directions in medical imag
 e analysis.
DTSTART;TZID=Europe/Paris:20250508T100000
DTEND;TZID=Europe/Paris:20250508T140000
LAST-MODIFIED:20250417T105603Z
LOCATION:Aula Magna\, DIAG
SUMMARY:Insights into traditional and deep-learning diffeomorphic registrat
 ion methods: to infinity and beyond in the LDDMM-verse - Monica Hernandez 
 Gimenez\, Dep. Computer Science\, Univ. Zaragoza
URL;TYPE=URI:https://www.diag.uniroma1.it/node/29340
END:VEVENT
END:VCALENDAR