MEDICAL IMAGE PROCESSING DEVICE AND MEDICAL IMAGE PROCESSING METHOD

To collect two-dimensional medical image data of a second modality for learning on the basis of three-dimensional medical image data of a first modality easily and in large quantities.SOLUTION: A medical image processing device includes an allocation unit, a generation unit, and a learning unit. The...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: OCHIAI RIE, YAMAHANA AKIKO, NAGAI SEIICHIRO, SHIMA KAZUNARI, MAEHAMA TOMIO, KOBAYASHI HIROMASA, OHASHI RITA
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To collect two-dimensional medical image data of a second modality for learning on the basis of three-dimensional medical image data of a first modality easily and in large quantities.SOLUTION: A medical image processing device includes an allocation unit, a generation unit, and a learning unit. The allocation unit allocates substance information including information on density and composition for each part of a subject included in three-dimensional data on a medical image of the subject generated by a first modality. The generation unit generates a plurality of pieces of simulated two-dimensional image data by simulating imaging of the three-dimensional data in which the substance information is set by a second modality using a plurality of simulation imaging parameter sets. The learning unit constructs a learned model by using the plurality of pieces of simulated two-dimensional image data and definite diagnosis information associated with the three-dimensional data as a training dataset.SELECTED DRAWING: Figure 2 【課題】第1モダリティの3次元医用画像データにもとづいて学習用に第2モダリティの2次元医用画像データを容易に数多く収集すること。【解決手段】実施形態に係る医用画像処理装置は、割当部と、生成部と、学習部とを備える。割当部は、第1モダリティで生成された被検体の医用画像の3次元データに含まれる被検体の各部について、密度と組成の情報を含む物質情報を割り当てる。生成部は、物質情報が設定された3次元データを複数の模擬撮影パラメータセットを用いて第2モダリティで撮影したと模すことにより、複数の模擬2次元画像データを生成する。学習部は、複数の模擬2次元画像データと、3次元データに関連付けられた確定診断情報と、をトレーニングデータセットとして用いて学習済みモデルを構築する。【選択図】 図2