Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Learning to Zoom with Anatomical Relations for Medical Structure Detection
Authors: Bin Pu, Liwen Wang, Xingbo Dong, Xingguo Lv, ZHE JIN
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Empirical validation across three diverse medical imaging benchmarks demonstrates that ZR-DETR consistently outperforms strong baselines in both single-domain and unsupervised domain adaptation scenarios. Extensive experiments across diverse medical imaging benchmarks demonstrate that our approach consistently surpasses robust baseline models in terms of detection accuracy and uncertainty calibration, especially within the context of unsupervised domain adaptation (UDA) scenarios. 4 Experiments |
| Researcher Affiliation | Academia | Bin Pu1, Liwen Wang2, Xingbo Dong2, Xingguo Lv1, Zhe Jin2 1Hunan University 2Anhui University |
| Pseudocode | No | The paper describes the methodology using mathematical equations and textual explanations (Sections 3.1-3.5) but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain an explicit statement about the release of source code for the described methodology, nor does it provide a link to a code repository within the main content. |
| Open Datasets | Yes | Fetal Cardiac Structure (FCS) [9] is a diversified ultrasound dataset... Early Pregnancy View (EPV) [33] is a challenging early pregnancy ultrasound dataset... MM-WHS [34] consists of 20 unpaired MRI and 20 CT volumes... |
| Dataset Splits | Yes | The FCS, EPV and MM-WHS datasets were divided into a training set, a validation set, and a test set in the ratio of 7:1:2, and all the settings remained the same. |
| Hardware Specification | Yes | trained for 20 epochs and 2 batch size with one RTX3090 GPU. |
| Software Dependencies | No | For a fair comparison, we use Res Net-50 [36] as the backbone for all experiments, which is implemented in Py Torch and trained for 20 epochs and 2 batch size with one RTX3090 GPU. ... we trained the model using the Adam W optimizer... The paper mentions PyTorch and Adam W optimizer but does not specify their version numbers or versions for other libraries used. |
| Experiment Setup | Yes | For a fair comparison, we use Res Net-50 [36] as the backbone for all experiments, which is implemented in Py Torch and trained for 20 epochs and 2 batch size with one RTX3090 GPU. For data augmentation, we use random horizontal flipping, random color jittering, grayscale, gaussian blurring, and cutout patches for image augmentation. We uniformly resized medical images to 800 1333 for all stages, and we trained the model using the Adam W optimizer with an initial learning rate of 0.01 with the weight decay of 1 10 4. The FCS, EPV and MM-WHS datasets were divided into a training set, a validation set, and a test set in the ratio of 7:1:2, and all the settings remained the same. |