Curriculum Hierarchical Knowledge Distillation for Bias-Free Survival Prediction
Authors: Chaozhuo Li, Zhihao Tang, Mingji Zhang, Zhiquan Liu, Litian Zhang, Xi Zhang
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our proposal is extensively evaluated over popular datasets and experimental results demonstrate its superiority. Extensive experiments across multiple datasets validate its superiority. Table 2 presents survival prediction results using the C-index and STAGE-5 metrics. We conduct ablation studies on the NLST and LUAD datasets to evaluate the contribution of each module in Patho KD, with results presented in Table 4. |
| Researcher Affiliation | Academia | 1Key Laboratory of Trustworthy Distributed Computing and Service (Mo E), Beijing University of Posts and Telecommunications, Beijing 100876, China 2Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital, Fuzhou 350014, China 3Jinan University, Guangzhou 510632, China EMAIL, EMAIL, EMAIL, EMAIL |
| Pseudocode | No | The paper describes the methodology using textual explanations and mathematical equations (e.g., Eq. 1 through Eq. 18), and provides a framework overview in Figure 2. However, it does not include any explicitly labeled 'Pseudocode', 'Algorithm', or structured code-like blocks. |
| Open Source Code | No | The paper states: 'All baseline survival models are re-implemented faithfully from their original publications and available open-source repositories.' This refers to the code for baseline models used for comparison, not the authors' own implementation of the proposed Patho KD method. There is no explicit statement or link indicating that the source code for Patho KD is publicly available. |
| Open Datasets | Yes | We evaluate our proposal on five real-world cancer datasets. The first is the National Lung Screening Trial (NLST) dataset [Team, 2011], which includes cases of adenocarcinoma (ADC) and squamous cell carcinoma (SCC). The remaining four datasets are from The Cancer Genome Atlas (TCGA) [Tomczak et al., 2015], encompassing lung cancer (LUSC and LUAD), breast cancer (BRCA), and bladder cancer (BLCA). |
| Dataset Splits | Yes | For a fair comparison, we adopt a 5-fold cross-validation scheme: in each fold, 10% of the training split is held out as a validation set for early stopping [Bai et al., 2021]. |
| Hardware Specification | Yes | All experiments are carried out in Py Torch 2.2.0 on NVIDIA V100 GPUs with 32 GB of memory. All models were run on two Xeon E5-2690 v4 processors (2.60 GHz) and four NVIDIA V100 GPUs. |
| Software Dependencies | Yes | All experiments are carried out in Py Torch 2.2.0 on NVIDIA V100 GPUs with 32 GB of memory. |
| Experiment Setup | Yes | Models are trained for up to 200 epochs (with patience of 10 epochs on validation loss) using the Adam W optimizer (weight decay 1e 4) and a fixed batch size of 64. Patho KD performs best with a mask WSI number of 3 and a mask patch ratio of 50%. Patho KD performs best with an expanding WSI number of 3 and a mask patch ratio of 30%. |