公开(公告)号：US12062168B2

公开(公告)日：2024-08-13

申请号：US17305631

申请日：2021-07-12

Applicant: SIEMENS HEALTHINEERS AG

Inventor： Felix Meister ,  Tiziano Passerini ,  Tommaso Mansi ,  Eric Lluch Alvarez ,  Chloé Audigier ,  Viorel Mihalef

IPC: G06T7/00 ,  A61B6/00 ,  G06F18/22 ,  G06F18/231 ,  G06F18/243 ,  G06T7/11 ,  A61B6/50

CPC classification number: G06T7/0012 ,  A61B6/5217 ,  G06F18/22 ,  G06F18/231 ,  G06F18/24323 ,  G06T7/11 ,  A61B6/50 ,  G06T2207/20076 ,  G06T2207/30061 ,  G06V2201/03

Abstract: Systems and methods for estimating local conductivities from anatomical information derived from MR images, ECG, and sparse contact maps are provided. ECG features and sparse measurements are mapped to an anatomical model represented as a graph. Graph convolutional layers and a multilayer perceptron are applied to extract local and global features respectively. The local and global features are combined and further processed by a series of fully connected layers to regress a set of vertex conductivities.