Related Experiment Video
Updated: Jan 25, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.3K
St-Swin TransNet: a spatiotemporal swin transformer-based network for self-supervised depth estimation in
Derong Yu1, Wenyuan Sun1, Junchen Wang2
1Institute of Medical Robotics, Shanghai Jiao Tong University, Dongchuan Road, Shanghai, 200240, China.
International Journal of Computer Assisted Radiology and Surgery
|January 24, 2026
Summary
This study introduces a novel spatiotemporal Swin transformer network (ST-Swin TransNet) for accurate depth estimation in stereoscopic laparoscopic videos, improving surgical navigation. The method significantly outperforms existing approaches in depth estimation and augmented reality navigation tasks.
Area of Science:
- Computer-assisted intervention
- Medical imaging analysis
- Surgical robotics
Background:
- Depth estimation from stereoscopic laparoscopic videos is crucial for computer-assisted interventions and surgical navigation.
- Existing methods often overlook temporal information present in stereoscopic laparoscopic videos, limiting accuracy.
- Accurate depth perception is vital for enhancing surgical precision and safety.
Purpose of the Study:
- To develop a novel deep learning network for accurate depth estimation from stereoscopic laparoscopic videos.
- To leverage spatiotemporal features within video sequences for improved depth prediction.
- To evaluate the network's performance in depth estimation and its application in augmented reality surgical navigation.
Main Methods:
- A spatiotemporal Swin (ST-Swin) transformer-based network (ST-Swin TransNet) utilizing a symmetric encoder-decoder architecture was proposed.
- The network employs 12 ST-Swin blocks designed to capture spatiotemporal features via self-attention mechanisms.
- Hierarchical spatiotemporal features from binocular laparoscopic videos are exploited to predict disparity maps for depth estimation.
Main Results:
- The ST-Swin TransNet achieved a mean absolute depth error (mADE) of 3.33 mm in depth estimation tasks.
- The method demonstrated a mean absolute distance (mAD) of 1.07 mm in video see-through augmented reality (VST-AR) navigation.
- Experiments on two public datasets confirmed the superior performance of the proposed method over state-of-the-art techniques.
Conclusions:
- A novel spatiotemporal Swin transformer network (ST-Swin TransNet) was successfully developed for self-supervised depth estimation in binocular laparoscopic videos.
- The proposed method significantly outperforms existing state-of-the-art approaches in depth estimation accuracy.
- ST-Swin TransNet shows promise for enhancing computer-assisted interventions and surgical navigation systems.
Related Concept Videos
01:33Bacterial Transformation
59.5K
In 1928, bacteriologist Frederick Griffith worked on a vaccine for pneumonia, which is caused by Streptococcus pneumoniae bacteria. Griffith studied two pneumonia strains in mice: one pathogenic and one non-pathogenic. Only the pathogenic strain killed host mice.
Griffith made an unexpected discovery when he killed the pathogenic strain and mixed its remains with the live, non-pathogenic strain. Not only did the mixture kill host mice, but it also contained living pathogenic bacteria that...
Griffith made an unexpected discovery when he killed the pathogenic strain and mixed its remains with the live, non-pathogenic strain. Not only did the mixture kill host mice, but it also contained living pathogenic bacteria that...
59.5K
02:26Protein Networks
4.5K
An organism can have thousands of different proteins, and these proteins must cooperate to ensure the health of an organism. Proteins bind to other proteins and form complexes to carry out their functions. Many proteins interact with multiple other proteins creating a complex network of protein interactions.
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
4.5K
01:06What are Estimates?
8.2K
It isn't easy to measure a parameter such as the mean height or the mean weight of a population. So, we draw samples from the population and calculate the mean height or mean weight of the individuals in the sample. This sample data acts as a representative measure of the population parameter. These sample statistics are known as estimates.
The estimate for the mean of a sample is denoted by ͞x, whereas the mean of the population is designated as μ. Further, parameters such...
The estimate for the mean of a sample is denoted by ͞x, whereas the mean of the population is designated as μ. Further, parameters such...
8.2K
02:18Network Covalent Solids
16.1K
Network covalent solids contain a three-dimensional network of covalently bonded atoms as found in the crystal structures of nonmetals like diamond, graphite, silicon, and some covalent compounds, such as silicon dioxide (sand) and silicon carbide (carborundum, the abrasive on sandpaper). Many minerals have networks of covalent bonds.
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
16.1K
01:27Uniform Depth Channel Flow
540
Uniform depth channel flow keeps fluid depth consistent along channels such as irrigation canals. In natural channels, such as rivers, approximate uniform flow is often assumed. This condition occurs when the channel’s bottom slope matches the energy slope, balancing potential energy lost from gravity with head loss due to shear stress. This balance prevents depth changes along the channel length, resulting in a steady, uniform flow.Uniform flow in open channels with a constant cross-section...
540
01:15Depth Perception and Spatial Vision
1.9K
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
1.9K