Multi-view Evidential Learning-based Medical Image Segmentation
Authors: Chao Huang, Yushu Shi, Waikeung Wong, Chengliang Liu, Wei Wang, Zhihua Wang, Jie Wen
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on three datasets show that our method obtains superior segmentation performance. ... The proposed method is evaluated on four medical image segmentation datasets: Mo Nu Seg (Kumar et al. 2017), Gla S (Sirinukunwattana et al. 2017), TNBC (Naylor et al. 2018), and DRIVE (Staal et al. 2004). ... Dice score and Io U (Wang et al. 2022) are used to evaluate the performance. ... Quantitative comparison of our method and other SOTA methods ... Ablation study on the effectiveness of the proposed components |
| Researcher Affiliation | Academia | 1School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University 2School of Fashion and Textiles, Hong Kong Polytechnic University 3Laboratory for Artificial Intelligence in Design, Hong Kong 4Department of Computer Science and Engineering, Hong Kong University of Science and Technology 5Department of Computer Science, City University of Hong Kong 6School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen |
| Pseudocode | No | The paper describes the methodology in detail, including mathematical formulations for the State Space Model, but it does not present any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any specific links to a code repository or an explicit statement about releasing the source code for the described methodology. |
| Open Datasets | Yes | The proposed method is evaluated on four medical image segmentation datasets: Mo Nu Seg (Kumar et al. 2017), Gla S (Sirinukunwattana et al. 2017), TNBC (Naylor et al. 2018), and DRIVE (Staal et al. 2004). |
| Dataset Splits | Yes | Mo Nu Seg contains 30 digital microscopic tissue images of several patients. The training, validation, and test sets are organized as (Wang et al. 2022). Gla S has 85 images for training and 80 images for testing. ... DRIVE dataset is divided into 20 images for training and 20 images for testing. |
| Hardware Specification | Yes | All experiments are conducted on a single NVIDIA A100 GPU with 80GB memory. |
| Software Dependencies | No | Our model is optimized by Adam W (Loshchilov and Hutter 2017) and the cosine annealing learning rate scheduler is adopted. The paper does not specify version numbers for any software libraries or frameworks. |
| Experiment Setup | Yes | Our model is optimized by Adam W (Loshchilov and Hutter 2017) and the cosine annealing learning rate scheduler is adopted. The learning rate is set to 1e-3. The size of input image and batch size are 224 224 and 2, respectively. |