用于CT肝脏肿瘤分割的基础模型参数高效微调
Parameter-Efficient Fine-Tuning of Foundation Models for Liver Tumor Segmentation in CT
浏览论文内容
中文总结 AI 辅助
本研究评估了SAM在肝脏肿瘤CT分割中的参数高效微调，比较多种适配器，发现LoRA和Conv-Adapter准确率最高，DiSCo效率最优，支持在资源受限时采用PEFT。
中文摘要 AI 辅助
我们评估了Segment Anything Model（SAM）在结直肠癌肝转移腹部CT中肝脏肿瘤分割的参数高效微调（PEFT）。我们比较了低秩适应（LoRA）、4位量化LoRA（QLoRA）、卷积适配器（Conv-Adapter）以及我们提出的方向性谱Top-K适配器（DiSCo），在冻结SAM主干的同时仅训练适配器。DiSCo从行归一化权重的奇异值分解中推导谱基，并学习秩门控谱系数、每输出幅度偏移和谱增益，在推理时可选Top-K秩选择，可训练参数为0.14M。我们基准测试了五种提示模式：无提示、单点、多点以及交并比为0.50和0.75的边界框。Conv-Adapter和LoRA达到了最高准确率（总体Dice为0.793和0.792；单点Dice为0.795和0.792；95百分位豪斯多夫距离（HD95）为32mm）。QLoRA接近（总体Dice为0.766；单点Dice为0.768；HD95为36.41mm），可训练参数为0.91M，延迟为120ms，峰值内存为4.9GB。DiSCo实现了每百万可训练参数的最高Dice（4.66），总体Dice为0.653，单点Dice为0.698，HD95为49.53mm。这些结果表明了准确性与效率之间的权衡，并支持在计算和标注数据有限的情况下，通过降低适应成本进行肝脏肿瘤分割的PEFT。代码：此https URL
英文摘要
We evaluated parameter-efficient fine-tuning (PEFT) of the Segment Anything Model (SAM) for liver tumor segmentation in abdominal CT of colorectal liver metastases. We compared Low-Rank Adaptation (LoRA), 4-bit Quantized LoRA (QLoRA), a convolutional adapter (Conv-Adapter), and our Directional Spectral Top-K adapter (DiSCo), training only adapters while freezing the SAM backbone. DiSCo derives spectral bases from singular value decomposition of row-normalized weights and learns rank-gated spectral coefficients, per-output magnitude offsets, and a spectral gain, with optional Top-K rank selection at inference and 0.14 M trainable parameters. We benchmarked five prompting regimes: no prompt, single-point, multi-point, and bounding boxes at intersection over union 0.50 and 0.75. Conv-Adapter and LoRA achieved the highest accuracy (overall Dice 0.793 and 0.792; single-point Dice 0.795 and 0.792; 95th-percentile Hausdorff distance (HD95) 32 mm). QLoRA was close (overall Dice 0.766; single-point Dice 0.768; HD95 36.41 mm), with 0.91 M trainable parameters, 120 ms latency, and 4.9 GB peak memory. DiSCo achieved the highest Dice per million trainable parameters (4.66), with overall Dice 0.653, single-point Dice 0.698, and HD95 49.53 mm. These results show an accuracy-efficiency trade-off and support PEFT for liver tumor segmentation with reduced adaptation costs when compute and labeled data are limited. Code: https://github.com/Ramtin-Mojtahedi/PEFT-SAM-Liver-CT
发表机构
- University Health Network（大学健康网络）
- Memorial Sloan Kettering Cancer Center（纪念斯隆凯特琳癌症中心）
- University of Alberta（阿尔伯塔大学）
- Alberta Machine Intelligence Institute（阿尔伯塔机器智能研究所）
- Islamic Azad University（伊斯兰阿扎德大学）
机构由 AI 辅助整理，请以论文原文为准。
