Segmentation of the distal femur in ultrasound images
Publication Info
- Author
B. Hohlmann
J. Glanz
K. Radermacher - Conference
CURAC
- Publisher
De Gruyter
- Date
2020 - Link
https://www.degruyter.com/document/doi/10.1515/cdbme-2020-0034/html - PDF „CURAC 2020“
Objectives
Ultrasound is a widely used imaging technology that allows for fast diagnosis of a broad range of illnesses and injuries of the musculoskeletal system. However, interpreting ultrasound images remains a challenging task that requires expert knowledge and years of training for each exam. One crucial step for the long-term goal of automatic diagnosis is pixel wise semantic segmentation.
Methods
In this work, several state-of-the-art semantic segmentation networks were trained on a new dataset of manually annotated ultrasound images depicting the distal femur.
Results
PSP-Net achieved the best overall performance with an average surface distance error (SDE) of 0.64 mm.
Conclusions
We recommend the PSP-Net architecture for semantic segmentation of bone surfaces.
The average surface distance error (SDE) over epochs on the validation set. U-Net in magenta, HR-Net in gray, HR + OCR in red, MobileNetv2 in teal and PSP-Net in petrol. Non-smoothed values in transparent.