Two-stage multi-task deep learning framework for simultaneous pelvic bone segmentation and landmark detection from CT images.
Haoyu ZhaiZhonghua ChenLei LiHairong TaoJinwu WangKang LiMoyu ShaoXiaomin ChengJing WangXiang WuChuan WuXiao ZhangLauri KettunenHongkai WangPublished in: International journal of computer assisted radiology and surgery (2023)
Using the multi-task networks and the coarse-to-fine strategy, this method achieved more accurate bone segmentation and landmark detection than the SOTA method, especially for diseased hip images. Our work contributes to accurate and rapid design of acetabular cup prostheses.
Keyphrases
- deep learning
- convolutional neural network
- loop mediated isothermal amplification
- artificial intelligence
- bone mineral density
- machine learning
- total hip arthroplasty
- high resolution
- real time pcr
- computed tomography
- label free
- soft tissue
- bone regeneration
- molecular dynamics
- air pollution
- dual energy
- magnetic resonance imaging
- rectal cancer
- molecular dynamics simulations
- sensitive detection
- image quality
- total knee arthroplasty
- quantum dots