MEDICAL IMAGE PROCESS APPARATUS AND MEDICAL IMAGE LEARNING METHOD AND MEDICAL IMAGE PROCESS METHOD

A medical image learning method performed by a medical image processing apparatus according to an embodiment includes the steps of: obtaining a learning image dataset including a first synthetic medical image in which a metal artifact area image is synthesized within a medical image of a body, and a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: KIM NAMKUG, JEONG JIHEON
Format: Patent
Sprache:eng ; kor
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A medical image learning method performed by a medical image processing apparatus according to an embodiment includes the steps of: obtaining a learning image dataset including a first synthetic medical image in which a metal artifact area image is synthesized within a medical image of a body, and a second synthetic medical image obtained by adding a metal artifact to the first synthetic medical image; and a training an artificial neural network model with the learning image dataset to obtain a processed medical image with reduced metal artifacts from a target medical image including an actual metal artifact area. 일 실시예에 따른 의료영상 처리 장치가 수행하는 의료영상 학습 방법은, 신체에 대한 의료영상 내에 금속인공물 영역 영상이 합성된 제 1 합성 의료영상과 제 1 합성 의료영상에 금속 아티팩트(metal artifact)가 부가된 제 2 합성 의료영상을 포함하는 학습용 영상 데이터세트를 획득하는 단계와, 실제 금속인공물 영역이 포함된 대상 의료영상으로부터 금속 아티팩트를 감소시킨 처리 의료영상을 획득할 수 있도록, 인공신경망 모델이 상기 학습용 영상 데이터세트를 학습하는 단계를 포함한다.