Processing

Please wait...

PATENTSCOPE will be unavailable a few hours for maintenance reason on Tuesday 27.07.2021 at 12:00 PM CEST
Settings

Settings

Goto Application

1. WO2020135015 - METHOD, APPARATUS AND DEVICE FOR ESTABLISHING MEDICAL IMAGING MODEL, AND STORAGE MEDIUM

Publication Number WO/2020/135015
Publication Date 02.07.2020
International Application No. PCT/CN2019/124239
International Filing Date 10.12.2019
IPC
G16H 30/20 2018.1
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
30ICT specially adapted for the handling or processing of medical images
20for handling medical images, e.g. DICOM, HL7 or PACS
CPC
G06N 3/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
G06N 3/0454
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
0454using a combination of multiple neural nets
G16H 30/20
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
30ICT specially adapted for the handling or processing of medical images
20for handling medical images, e.g. DICOM, HL7 or PACS
Applicants
  • 深圳先进技术研究院 SHENZHEN INSTITUTES OF ADVANCED TECHNOLOGY [CN]/[CN]
Inventors
  • 梁栋 LIANG, Dong
  • 王珊珊 WANG, Shanshan
  • 柯子文 KE, Ziwen
  • 刘新 LIU, Xin
  • 郑海荣 ZHENG, Hairong
Agents
  • 北京品源专利代理有限公司 BEYOND ATTORNEYS AT LAW
Priority Data
201811654047.427.12.2018CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) METHOD, APPARATUS AND DEVICE FOR ESTABLISHING MEDICAL IMAGING MODEL, AND STORAGE MEDIUM
(FR) PROCÉDÉ, APPAREIL ET DISPOSITIF POUR ÉTABLIR UN MODÈLE D'IMAGERIE MÉDICALE, ET SUPPORT DE STOCKAGE
(ZH) 一种医学成像模型的建立方法、装置、设备及存储介质
Abstract
(EN)
A method, apparatus and device for establishing a medical imaging model, and a storage medium. The method comprises: obtaining K-space under-sampled data of a medical image as training samples (101); inputting the training samples into a pre-constructed original neural network for training, and optimizing the original neural network according to model loss functions (102); and when the model loss functions converge, stopping the training of a machine learning model and using the trained original neural network as a target medical imaging model (103), wherein the original neural network comprises a frequency domain network and an image domain network; the model loss functions comprise a main loss function and at least one additional loss function, and the additional loss function comprises a k-space loss function and/or an image domain loss function.
(FR)
L'invention concerne un procédé, un appareil et un dispositif permettant d'établir un modèle d'imagerie médicale, et un support de stockage. Le procédé consiste à : obtenir des données sous-échantillonnées d'espace k d'une image médicale en tant qu'échantillons d'apprentissage (101) ; entrer les échantillons d'apprentissage dans un réseau neuronal d'origine pré-construit pour l'apprentissage, et optimiser le réseau neuronal d'origine selon des fonctions de perte de modèle (102) ; et lorsque les fonctions de perte de modèle convergent, arrêter l'apprentissage d'un modèle d'apprentissage machine et utiliser le réseau neuronal d'origine entraîné en tant que modèle d'imagerie médicale cible (103), le réseau neuronal d'origine comprenant un réseau de domaine fréquentiel et un réseau de domaine d'image ; les fonctions de perte de modèle comprennent une fonction de perte principale et au moins une fonction de perte supplémentaire, et la fonction de perte supplémentaire comprend une fonction de perte d'espace k et/ou une fonction de perte de domaine d'image.
(ZH)
一种医学成像模型的建立方法、装置、设备及存储介质。该方法包括:获取医学图像的K空间欠采样数据作为训练样本(101);将训练样本输入至预先构建的原始神经网络进行训练,并根据模型损失函数优化原始神经网络(102);当模型损失函数收敛时,停止训练机器学习模型,将训练完成的原始神经网络作为目标医学成像模型(103);其中,原始神经网络包括频率域网络和图像域网络;模型损失函数包括主损失函数以及至少一个附加损失函数,附加损失函数包括k空间损失函数和/或图像域损失函数。
Also published as
Latest bibliographic data on file with the International Bureau