Unlocking Dark Vision Potential for Medical Image Segmentation
Authors: Hongpeng Yang, Xiangyu Hu, Yingxin Chen, Siyu Chen, Srihari Nelakuditi, Yan Tong, Shiqiang Ma, Fei Guo
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments demonstrate that the DVNet achieves SOTA performance in various medical image segmentation tasks. Extensive experiments have demonstrated that the DVNet can consistently enhance the performance of its baseline model by enhancing the model s visual activation of lesion areas through contrast enhancement. |
| Researcher Affiliation | Academia | 1Department of Computer Science and Engineering, University of South Carolina 2School of Computer Science and Engineering, Central South University 3Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences Corresponding author EMAIL, EMAIL, and EMAIL |
| Pseudocode | No | The paper describes the methodology in text and figures (Figure 1, Figure 2, Figure 3, Figure 4) but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing open-source code or a link to a code repository. |
| Open Datasets | Yes | The Bra TS 2020 and Bra TS 2023 datasets [Menze et al., 2014; Kazerooni et al., 2024; Henry et al., 2021] are widely recognized benchmarks for brain tumor segmentation tasks. The Abdomen MRI dataset was used for the MICCAI 2022 AMOS Challenge [Ji et al., 2022], designed for segmentation of abdominal organs. Microscopy Cell dataset was used for the Neur IPS 2022 Cell Segmentation Challenge [Ma et al., 2024b]. |
| Dataset Splits | Yes | The data is split randomly: 70% of the 3D volumes for training, 10% for validation, and the remaining 20% for testing. (for BraTS) we employed 60 labeled MRI scans for training and 50 scans for testing in the Abdomen MRI dataset and 1,000 images for training and 101 images for evaluation in the Microscopy Cell dataset. |
| Hardware Specification | No | The paper discusses implementation details such as input crop size, batch size, optimization process, learning rate scheduler, number of epochs, and data augmentations, but does not specify any particular hardware used for running the experiments. |
| Software Dependencies | No | DVNet is implemented based on Seg Mamba [Xing et al., 2024] for 3D tasks and x LSTM [Chen et al., 2024b] for 2D tasks. No specific version numbers provided for these or other software dependencies. |
| Experiment Setup | Yes | For 3D tasks, the input crop size is set to (64 64 64) with a batch size of 2. The optimization process uses cross-entropy loss with a stochastic gradient descent optimizer and a polynomial learning rate scheduler (initial learning rate of 0.01 and decay of 1 10 5). Training runs for 1,000 epochs with data augmentations including brightness, gamma, rotation, scaling, mirror, and elastic deformation. For 2D tasks, the loss function is the sum of Dice loss and cross-entropy loss, optimized with the Adam W optimizer (weight decay: 0.05). Learning rates are set empirically for each dataset: Abdomen MRI (0.005) and Microscopy (0.0015). Batch size is set to 30 for Abdomen MRI and 12 for Microscopy. All models are trained for 1,000 epochs for 2D tasks. |