DenseNet-based transfer learning for LV shape Classification: Introducing a novel information fusion and data augmentation using statistical Shape/Color modeling

Farshid Babapour Mofrad,Gelareh Valizadeh

Expert Systems with Applications（2023）

引用 1|浏览13

暂无评分

摘要

•To suggest extracting PPVs from 3D surface points suitable for feeding into CNNs.•To fuse information by assigning additional 3D/3D + t information in color to PPVs.•To introduce a novel data augmentation method via statistical shape/color modeling.•To adopt DenseNet model as a backbone structure for MI shape screening through TL.•To study various fine-tuning strategies to define an optimal one for MI detection.

查看译文

关键词

Transfer learning,Deep convolutional neural network,Feature fusion,Data augmentation,Statistical shape modeling,Myocardial infarction,Fine-tuning

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要