利用DICOM头信息自监督学习，提升医学影像模型跨设备泛化能力。
The Protocol Genome A Self Supervised Learning Framework from DICOM Headers
- 从DICOM头字段中学习协议关联，结合图像特征进行对比学习与掩码预测。
- 外部验证下AUC达0.901，校准误差降至0.036，优于基线0.847和0.058。
- 适用于多模态、多厂商场景，降低协议边界误报，适合临床部署。
本文提出Protocol Genome，一种基于DICOM头信息的自监督学习框架，通过分析扫描参数（如设备型号、序列、千伏、回波时间等）与图像质量之间的隐含关联，构建协议感知且临床鲁棒的图像表征。该方法在126万例研究（7个医疗机构、31台扫描仪、3个厂商；包含CT、MRI、X光）上验证，涵盖胸部CT肺栓塞分诊、脑部MRI胶质瘤分级及胸片心脏扩大检测任务。相比强基线模型（SimCLR、MAE）及ImageNet迁移，其外部AUC提升0.046~0.058，校准误差下降25%-37%（p<0.01，DeLong检验）。即使仅使用10%-20%标注数据，性能仍稳定。该技术可嵌入PACS系统，支持DICOM C-FIND/C-MOVE与DICOMweb接口，提供去标识化与偏见审计，发布完整模型卡与部署指南。
原文摘要 · Abstract (English)
In this paper, we introduce the Protocol Genome, a self-supervised learning system that learns correlations from DICOM headers and achieves AUROC 0.901 (vs 0.847 baseline) and ECE 0.036 (vs 0.058) on fully held-out external validation. Our method also improves calibration and robustness across modalities (CT, MRI, CXR) and vendors. Clinical imaging is funneled through PACS/DICOM, where procedure choices (scanner make/model, sequence, kernel, kVp, TR/TE, and slice thickness) have consequences for contrast, noise, and artifact. These latent confounders impede the generalization of image-only networks across sites. We consider structured DICOM headers as a label and learn protocol-aware but clinically robust image representations. Protocol Genome obtains tokenized embeddings of de-identified header fields and models them along with image features using: (1) protocol-image contrastive learning, (2) masked protocol prediction, and (3) protocol-protocol translation. With 1.26M studies (7 health systems, 31 scanners, 3 vendors; CT, MR, CR/DR), we experiment on: (A) chest CT triage for PE, (B) brain MRI glioma grading, and (C) chest radiograph cardiomegaly detection. Relative to strong SSL baselines (SimCLR, MAE) as well as ImageNet transfer, Protocol Genome (+0.046: PE, +0.058: glioma, +0.041: cardiomegaly) is associated with higher external AUROC; 25-37% calibration improvements are obtained (p < 0.01, DeLong tests). While the gains may be task-dependent, they are preserved with 10-20% of labeled data. From a clinical point of view, the technique reduces false positives at protocol borders and is applicable in a PACS (DICOM C-FIND/C-MOVE, DICOMweb QIDO/WADO). We publish a model card and deployment guide, complete with both de-identification and bias audits.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。