Research map / Architectures and efficiency / Quantization
Privacy-Preserving SAM Quantization for Efficient Edge Intelligence in Healthcare
Zhikai Li, Jing Zhang, Qingyi Gu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add pseudo-positive label evolution (ablation) | data synthesis from Gaussian noise · AbdomenCT1k | IoU (%): +1.42 | Improved |
| add patch similarity to synthesis (ablation) | data synthesis · AbdomenCT1k | IoU (%): +2.46 | Improved |
| add scale reparameterization to calibration (ablation) | 4-bit SAM quantization · AbdomenCT1k | IoU (%): +1.71 | Improved |
| calibrate with DFQ-SAM synthesized data instead of Gaussian noise | 4-bit SAM · AbdomenCT1k | IoU (%): 62.03 -> 74.32 | Improved |
| calibrate with synthesized data instead of Gaussian noise | 4-bit SAM · MRI average | IoU (%): 65.89 -> 69.20 | Improved |
| calibrate with synthesized data instead of real data | 4-bit SAM · AbdomenCT1k | IoU (%): 69.83 -> 74.32 | Improved |
| calibrate with synthesized data instead of real data | 4-bit SAM · SA-Xray | IoU (%): 47.70 -> 45.47 | Hurt |
| calibrate with synthesized data instead of real data | 4-bit SAM · EndoVis | IoU (%): 66.31 -> 66.23 | No clear effect |
| calibrate with synthesized data instead of real data | 4-bit SAM · ISIC | IoU (%): 82.71 -> 83.23 | Improved |
| quantize to 4-bit instead of full precision | SAM with DFQ-SAM · CT average | IoU (%): 76.28 -> 74.45 | Mixed |
| vary balance hyperparameters alpha and beta in grid search (ablation) | DFQ-SAM · grid search | accuracy: 0.732 to 0.744 | No clear effect |
Closest papers by meaning
- Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decoding (CVPR 2024 · similarity 0.87)
- Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.86)
- BLO-SAM: Bi-level Optimization Based Finetuning of the Segment Anything Model for Overfitting-Preventing Semantic Segmentation (ICML 2024 · similarity 0.85)
- FairSeg: A Large-Scale Medical Image Segmentation Dataset for Fairness Learning Using Segment Anything Model with Fair Error-Bound Scaling (ICLR 2024 · similarity 0.84)
- Towards a Comprehensive, Efficient and Promptable Anatomic Structure Segmentation Model Using 3D Whole-Body CT Scans (AAAI 2025 · similarity 0.83)
- VQ-Seg: Vector-Quantized Token Perturbation for Semi-Supervised Medical Image Segmentation (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
