Learning Domain-Agnostic Representation for Disease Diagnosis
Authors: Churan Wang, Jing Li, Xinwei Sun, Fandong Zhang, Yizhou Yu, Yizhou Wang
ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To verify the utility and effectiveness of our method, we perform our method on mammogram benign/malignant classification. Here the clinical attributes are those related to the masses, which are summarized in ACR (Sickles et al., 2013) and are easy to obtain. We consider four datasets (one public and three in-house) that are collected from different sources. The results on unseen domains show that our method can outperform others by 6.2%. Besides, our learned disease-related features can successfully encode the information on the lesion areas. |
| Researcher Affiliation | Collaboration | Churan Wang12, Jing Li1, Xinwei Sun7 , Fandong Zhang5, Yizhou Yu6, Yizhou Wang234 1 School of Computer Science, Peking University 2 CFCS, School of CS, Inst. for AI, Peking University 3 Nat l Key Lab. of GAI & Beijing Institute for GAI (BIGAI) 4 Nat l Eng. Research Center of Visual Technology 5 AI lab, Deepwise, Beijing, China 6 Department of Computer Science, The University of Hong Kong 7 School of data Science, Fudan University |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide a link to open-source code for the described methodology or explicitly state that the code is being released. |
| Open Datasets | Yes | Public dataset (DDSM (Bowyer et al., 1996)) and three in-house datasets (In H1, In H2, and In H3) what we use are from different centers (center4, 1, 2, 3 respectively). |
| Dataset Splits | Yes | For each dataset, we randomly split it into training, validation, and testing sets with an 8:1:1 patient-wise ratio. |
| Hardware Specification | No | The paper does not mention any specific hardware used for running the experiments (e.g., CPU, GPU models, or cloud computing instances with specifications). |
| Software Dependencies | No | The paper mentions software like "Py Torch" and "Adam" but does not provide specific version numbers for these dependencies. |
| Experiment Setup | Yes | The weight hyperparameter in variance regularizer β is 1 in our experiments. |