BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//GREYC UMR CNRS 6072 - Groupe de Recherche en Informatique, Image, et Instrumentation de Caen - ECPv5.7.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:GREYC UMR CNRS 6072 - Groupe de Recherche en Informatique, Image, et Instrumentation de Caen
X-ORIGINAL-URL:https://www.greyc.fr
X-WR-CALDESC:évènements pour GREYC UMR CNRS 6072 - Groupe de Recherche en Informatique, Image, et Instrumentation de Caen
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20260329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20261025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20260521T140000
DTEND;TZID=Europe/Paris:20260521T153000
DTSTAMP:20260507T224233
CREATED:20260429T133641Z
LAST-MODIFIED:20260505T085221Z
UID:12089-1779372000-1779377400@www.greyc.fr
SUMMARY:Séminaire Image: Shouhei Hanaoka et Rie Tanaka
DESCRIPTION:Nous aurons le plaisir d’écouter Shouhei Hanaoka\, Associate Professor (Graduate School of Medicine\, The University of Tokyo) et Rie Tanaka (Pharmaceutical and Health Sciences\, Kanazawa University).\nIl donneront un séminaire IMAGE le jeudi 21 mai 2026 à 14h en salle de séminaire F-200. \nTitre (Shouhei Hanaoka) : « Clinical importance of trees and graphs: An example of HoTPiG for cerebral aneurysm detection » \nRésumé :  \nThe human body has many tree-like structures such as blood vessels\, bronchi\, bile duct and nervous system.  In this talk\, I am going to present my old work for detecting local abnormality in such tree-like structures.\nThe proposed method was named as HoTPiG (Histogram of Triangular Paths in Graph).\nGiven a graph structure extracted from a binarized volume\, the proposed feature extraction algorithm can effectively encode both the morphological characteristics and the local branching pattern of the structure around each graph node (e.g.\, each voxel in the vessel).\nThe features are derived from a 3-D histogram whose bins represent a triplet of shortest path distances between the target node and all possible node pairs near the target node.\nThe extracted feature set is a vector with a fixed length and is readily applicable to state-of-the-art machine learning methods.\nI will show some examples of lesion detections in the lung and the cerebral vessel.\n\nTitre (Rie Tanaka) : « From Static X-ray to Functional Imaging: AI-driven Advances in Thoracic Radiography » \nRésumé :  \nChest radiography is the most widely used imaging modality for screening and follow-up. Dynamic chest radiography (DCR) extends conventional radiography by enabling functional assessment using flat-panel detector (FPD)-based systems. Quantitative and time-series analysis of DCR allows evaluation of respiratory and circulatory dynamics through lung density changes\, diaphragm motion\, and tracheal diameter. However\, these approaches remain limited by the two-dimensional projection nature of X-ray imaging\, motivating the integration of artificial intelligence and in silico–based strategies to enable higher-dimensional understanding of spatiotemporal patterns.\nWe present a unified framework integrating DCR\, artificial intelligence\, and in silico–based training strategies. We first enhance two-dimensional projections via image decomposition\, then estimate lung volume and respiratory function\, and finally reconstruct four-dimensional (4D) representations from dynamic X-ray time-series data using deep learning. Developed using large-scale virtual datasets and validated on clinical cases\, these approaches may enable functional and quantitative chest radiography\, bridging the gap between low-cost X-ray imaging and advanced modalities such as CT.
URL:https://www.greyc.fr/event/seminaire-image-shouhei-hanaoka-et-rie-tanaka/
LOCATION:ENSICAEN – Batiment F – Salle F-200\, 6 Bd Maréchal Juin\, Caen\, 14050\, France
CATEGORIES:General,Image,Seminaire Image
END:VEVENT
END:VCALENDAR