Weakly Supervised Segmentation of Vertebral Bodies with Iterative Slice-propagation
Vertebral body (VB) segmentation is an important preliminary step towards medical visual diagnosis for spinal diseases. However, most previous works require pixel/voxel-wise strong supervisions, which is expensive, tedious and time-consuming for experts to annotate. In this paper, we propose a Weakly supervised Iterative Spinal Segmentation (WISS) method leveraging only four corner landmark weak labels on a single sagittal slice to achieve automatic volumetric segmentation from CT images for VBs. WISS first segments VBs on an annotated sagittal slice in an iterative self-training manner. This self-training method alternates between training and refining labels in the training set. Then WISS proceeds to segment the whole VBs slice by slice with a slice-propagation method to obtain volumetric segmentations. We evaluate the performance of WISS on a private spinal metastases CT dataset and the public lumbar CT dataset. On the first dataset, WISS achieves distinct improvements with regard to two different backbones. For the second dataset, WISS achieves dice coefficients of $91.7\%$ and $83.7\%$ for mid-sagittal slices and 3D CT volumes, respectively, saving a lot of labeling costs and only sacrificing a little segmentation performance..
| Medienart: |
|
|---|
Preprint| Erscheinungsjahr: |
2024 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
arXiv.org - (2024) vom: 13. Feb. Zur Gesamtaufnahme - year:2024 |
|---|
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Peng, Shiqi [VerfasserIn] |
|---|
| Links: |
Volltext [kostenfrei] |
|---|
| Themen: |
000 |
|---|
| Förderinstitution / Projekttitel: |
|
|---|
| PPN (Katalog-ID): |
XAR042504988 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | XAR042504988 | ||
| 003 | DE-627 | ||
| 005 | 20240215080332.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 240215s2024 xx |||||o 00| ||eng c | ||
| 035 | |a (DE-627)XAR042504988 | ||
| 035 | |a (arXiv)2402.08892 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Peng, Shiqi |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Weakly Supervised Segmentation of Vertebral Bodies with Iterative Slice-propagation |
| 264 | 1 | |c 2024 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Vertebral body (VB) segmentation is an important preliminary step towards medical visual diagnosis for spinal diseases. However, most previous works require pixel/voxel-wise strong supervisions, which is expensive, tedious and time-consuming for experts to annotate. In this paper, we propose a Weakly supervised Iterative Spinal Segmentation (WISS) method leveraging only four corner landmark weak labels on a single sagittal slice to achieve automatic volumetric segmentation from CT images for VBs. WISS first segments VBs on an annotated sagittal slice in an iterative self-training manner. This self-training method alternates between training and refining labels in the training set. Then WISS proceeds to segment the whole VBs slice by slice with a slice-propagation method to obtain volumetric segmentations. We evaluate the performance of WISS on a private spinal metastases CT dataset and the public lumbar CT dataset. On the first dataset, WISS achieves distinct improvements with regard to two different backbones. For the second dataset, WISS achieves dice coefficients of $91.7\%$ and $83.7\%$ for mid-sagittal slices and 3D CT volumes, respectively, saving a lot of labeling costs and only sacrificing a little segmentation performance. | ||
| 650 | 4 | |a Computer Science - Computer Vision and Pattern Recognition |7 (dpeaa)DE-84 | |
| 650 | 4 | |a Computer Science - Machine Learning |7 (dpeaa)DE-84 | |
| 650 | 4 | |a 000 |7 (dpeaa)DE-84 | |
| 700 | 1 | |a Lai, Bolin |4 aut | |
| 700 | 1 | |a Yao, Guangyu |4 aut | |
| 700 | 1 | |a Zhang, Xiaoyun |4 aut | |
| 700 | 1 | |a Zhang, Ya |4 aut | |
| 700 | 1 | |a Wang, Yan-Feng |4 aut | |
| 700 | 1 | |a Zhao, Hui |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t arXiv.org |g (2024) vom: 13. Feb. |
| 773 | 1 | 8 | |g year:2024 |g day:13 |g month:02 |
| 856 | 4 | 0 | |u https://arxiv.org/abs/2402.08892 |z kostenfrei |3 Volltext |
| 912 | |a GBV_XAR | ||
| 951 | |a AR | ||
| 952 | |j 2024 |b 13 |c 02 | ||