量子核在医学基础模型嵌入中的相对经典坍缩优势

摘要

我们在MIMIC-CXR胸部X光片的二分类保险风险评估中，通过使用基于三种医学基础模型（MedSigLIP-448、RAD-DINO、ViT-patch32）冻结嵌入的量子支持向量机（QSVM），提供了无噪声模拟下的量子核优势证据。我们提出双层公平比较框架，两类分类器均采用相同的PCA-q特征。在第一层级（未调参QSVM vs 未调参线性SVM，双方C=1），QSVM在所有18组测试配置中均获得少数类F1优势（17组p<0.001，1组p<0.01）。经典线性核在所有量子比特数下均出现90-100%种子数的多数类预测崩溃，而QSVM始终保持有效召回率。在q=11（MedSigLIP-448平台中心）时，未经超参数调优的QSVM实现平均F1=0.343，显著优于经典方法的F1=0.050（F1增益+0.293，p<0.001）。在第二层级（未调参QSVM vs C值调优的RBF SVM）中，QSVM在全部七组测试配置中获胜（平均增益+0.068，最大增益+0.112）。特征谱分析表明量子核在q=11时有效秩达到69.80，远超线性核秩，且经典方法崩溃现象具有C不变性。全量子比特扫描揭示了模型间架构依赖的浓度起始现象。代码地址：https://github.com/sebasmos/qml-medimage

English

We provide evidence of quantum kernel advantage under noiseless simulation in binary insurance classification on MIMIC-CXR chest radiographs using quantum support vector machines (QSVM) with frozen embeddings from three medical foundation models (MedSigLIP-448, RAD-DINO, ViT-patch32). We propose a two-tier fair comparison framework in which both classifiers receive identical PCA-q features. At Tier 1 (untuned QSVM vs. untuned linear SVM, C = 1 both sides), QSVM wins minority-class F1 in all 18 tested configurations (17 at p < 0.001, 1 at p < 0.01). The classical linear kernel collapses to majority-class prediction on 90-100% of seeds at every qubit count, while QSVM maintains non-trivial recall. At q = 11 (MedSigLIP-448 plateau center), QSVM achieves mean F1 = 0.343 vs. classical F1 = 0.050 (F1 gain = +0.293, p < 0.001) without hyperparameter tuning. Under Tier 2 (untuned QSVM vs. C-tuned RBF SVM), QSVM wins all seven tested configurations (mean gain +0.068, max +0.112). Eigenspectrum analysis reveals quantum kernel effective rank reaches 69.80 at q = 11, far exceeding linear kernel rank, while classical collapse remains C-invariant. A full qubit sweep reveals architecture-dependent concentration onset across models. Code: https://github.com/sebasmos/qml-medimage