Automated annotation and quantitative description of ultrasound videos of the fetal heart

详细信息    查看全文

• 作者：Christopher P. Bridge ; ^a ; ^{christopher.bridge@eng.ox.ac.uk} ; Christos Ioannou^b ; ^{christos.ioannou@ouh.nhs.uk} ; J. Alison Noble^a ; ^{alison.noble@eng.ox.ac.uk}
• 关键词：Ultrasound ; Fetal ; Heart ; Cardiac ; View detection ; Rotation-invariant ; Random forests ; Particle filter
• 刊名：Medical Image Analysis
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：36
• 期：Complete
• 页码：147-161
• 全文大小：2701 K
• 卷排序：36

文摘

A model for automating interpretation of fetal heart ultrasound videos is proposed. Predictions are made in each frame using random forests. These are then linked through time using a dynamics model and a particle filter. Heart position, orientation, viewing plane and cardiac phase are estimated. High accuracy is achieved on a challenging dataset.