Rigorous Benchmarking of Variational Quantum Classifiers for Cancer Classification: Encoding Bandwidth and a Matched, Tuned Repeated Cross-Validation Study
Variational quantum classifiers (VQCs) are often compared with classical models under unequal data partitions, tuning budgets, or statistical treatment. We benchmark a 4-qubit VQC against logistic regression, an RBF-SVM, random forest, a quantum-kernel SVM (QSVM), and a majority-class control on Breast Cancer Wisconsin (n=569), Coimbra (n=116), and Golub leukemia (n=72), using five stratified folds across five seeds (25 paired blocks per dataset), identical training-fold-fitted preprocessing, and Nadeau–Bengio corrected paired tests with Holm correction. A first arm with a frozen configuration and the common [0, π] angle encoding left the VQC at 0.563, 0.515, and 0.576 accuracy (majority 0.617, 0.552, 0.652). In a second arm, in which every model was tuned by the same inner cross-validation and the encoding scale was a tuned hyperparameter, VQC accuracy rose to 0.864, 0.616, and 0.747. On Wisconsin the VQC stayed significantly below every matched classical model (differences 0.079–0.097, Holm p≤0.006) but above the majority class; on Golub it was significantly below only the SVM (Holm p=0.016); on Coimbra no comparison was significant. A QSVM with the same feature map exceeded the VQC numerically on every dataset, and three optimizers reached the same training loss in pilot diagnostics, pointing to the encoding and the variational head rather than the optimizer. The Wisconsin arms also differ in sample size (120 versus 569). Results apply to this low-qubit, PCA-compressed, exactly simulated setting and do not indicate a quantum advantage.
Variational quantum classifiers (VQCs) are often compared with classical models under unequal data partitions, tuning budgets, or statistical treatment. We benchmark a 4-qubit VQC against logistic regression, an RBF-SVM, random forest, a quantum-kernel SVM (QSVM), and a majority-class control on Breast Cancer Wisconsin (n=569), Coimbra (n=116), and Golub leukemia (n=72), using five stratified folds across five seeds (25 paired blocks per dataset), identical training-fold-fitted preprocessing, and Nadeau–Bengio corrected paired tests with Holm correction. A first arm with a frozen configuration and the common [0, π] angle encoding left the VQC at 0.563, 0.515, and 0.576 accuracy (majority 0.617, 0.552, 0.652). In a second arm, in which every model was tuned by the same inner cross-validation and the encoding scale was a tuned hyperparameter, VQC accuracy rose to 0.864, 0.616, and 0.747. On Wisconsin the VQC stayed significantly below every matched classical model (differences 0.079–0.097, Holm p≤0.006) but above the majority class; on Golub it was significantly below only the SVM (Holm p=0.016); on Coimbra no comparison was significant. A QSVM with the same feature map exceeded the VQC numerically on every dataset, and three optimizers reached the same training loss in pilot diagnostics, pointing to the encoding and the variational head rather than the optimizer. The Wisconsin arms also differ in sample size (120 versus 569). Results apply to this low-qubit, PCA-compressed, exactly simulated setting and do not indicate a quantum advantage.
Variational quantum classifiers (VQCs) are often reported to perform poorly on biomedical tabular data, but such conclusions depend on the encoding circuit, the training budget, and the comparison protocol. Methods. We compared 4-qubit VQCs, a quantum-kernel SVM (QSVM), a majority-class control, and tuned classical models (on four PCA features and on all features) on Wisconsin breast cancer (n = 569), Coimbra biomarkers (n = 116), Golub leukemia gene expression (n = 72, 7,129 genes), and SEER breast-cancer registry data (n = 4,024; 15% deaths). All models were evaluated on identical blocks of 5-fold × 5-seed stratified cross-validation with preprocessing fitted on training folds, nested tuning, and Nadeau–Bengio-corrected inference with Holm correction. Trained classifiers were also evaluated on three IBM Quantum devices. Results. On the three small datasets, a VQC with the conventional two-repetition ZZ feature map was close to chance (accuracy 0.688, 0.594, 0.509), whereas a product (Z) feature map reached 0.900, 0.797, and 0.658; at matched repetition counts the Z-minus-ZZ gap was +0.12 to +0.21 and, after Holm correction, significant on Wisconsin only. More evaluations (up to 1,500) and six initializations did not close the ZZ deficit on Wisconsin. With the encoding chosen by inner cross-validation, the VQC exceeded the majority control on Wisconsin and Golub and stayed below classical models on Wisconsin and in several Golub contrasts. No statistically significant difference was detected between the encoding-selected VQC and the classical baselines on Coimbra, or between the Z-map QSVM and PCA-4 classical models; Bayesian equivalence analysis supported practical equivalence only for the QSVM on Wisconsin. On SEER the encoding mattered little: the VQC exceeded the majority control in balanced accuracy (0.548 versus 0.500) but ranked patients worse than logistic regression and random forest (ROC-AUC 0.678 versus 0.745 and 0.733). Full-feature classical models were strongest overall. On hardware (class-balanced test subsets, not natural-prevalence performance), accuracy relative to noise-free simulation changed by −0.071 to +0.048 per run (pooled ibm_fez, Wisconsin: −0.024, 95% CI −0.060 to +0.008), using 5.5 QPU-minutes. Conclusions. Under a fixed training budget, encoding choice was associated with large differences in VQC accuracy on small datasets, so single-encoding negative results should not be generalized; because the Z and ZZ maps also differ in entanglement structure, we do not attribute the gap to encoding alone. No quantum advantage was observed.
A potential path forward is Quantum Machine Learning (QML), which aims to leverage quantum computing in conjunction with classical machine learning to enhance computing efficiency and the expressiveness of models. In this paper, two different quantum classifiers - Variational Quantum Classifier (VQC) and Quantum Kernel Support Vector Machine (QSVM) - are compared with three classical classifiers as baseline classifiers - Logistic Regression, Support Vector Machine (SVM), and a Multi-Layer Perceptron (MLP) - on the Breast Cancer Wisconsin dataset. The quantum circuits were created in the PennyLane framework and simulated on a classical backend. However, in terms of accuracy, classical Logistic Regression performed better with an accuracy of 97.8%, classical SVM and QSVM with an accuracy of 95.6% each, although the Quantum VQC achieved a lower accuracy of 88.9% and had a recall of 100% for the benign class, though it correctly identified only 12 of the 17 malignant cases (a malignant-class recall of approximately 70.6%). The drawback of quantum models is the higher training time; however, since the quantum circuit needs to be classically simulated, the quantum SVM took 23.29 seconds compared to less than 0.01 seconds for the classical linear models. These results indicate that for small structured datasets, classifiers based on quantum computing have not yet surpassed well-tuned classical counterparts. In some respects (e.g., benign-class recall), they perform competitively, though not on malignant-class recall, where the VQC in particular performed worse than the classical baselines, which is worth further investigation on real quantum computers.
In recent years, quantum machine learning (QML) has emerged as a promising approach for medical decision-making, with representative methods such as Quantum Support Vector Machines (QSVMs), Quantum Neural Networks (QNNs), and Variational Quantum Circuits (VQCs). Among these, VQCs are distinguished by their expressive capability; however, their training on small biomedical datasets often results in instability and high variance. To address this issue, we propose a heterogeneous ensemble of six shallow VQCs evaluated on the Wisconsin Breast Cancer Diagnostic (WBCD) dataset, which contains 569 diagnostic samples characterized by 30 cytological features and binary class labels differentiating benign and malignant cases. The ensemble introduces diversity through distinct feature maps, circuit architectures, and optimization strategies, using rotation gates (RX, RY, RZ) combined with entangling layers (CNOT, CZ). Model reliability is estimated from out-of-bag errors, with normalized weights calculated as the inverse of the mean squared error, and final decisions obtained via weighted soft voting. The preprocessing pipeline involves median imputation and standardization, followed by principal component analysis (PCA) to compress the 30-dimensional space into three to four principal components for quantum encoding. In our experiments, individual VQCs trained on the full dataset exhibited unstable accuracies ranging from 80% to 96% across three–four qubits, whereas the proposed ensemble achieved consistent performance of 95.71% (3 qubits) and 96.43% (4 qubits), reducing variance and matching strong classical baselines such as SVM and Random Forest. All models were implemented in PennyLane, utilizing the Lightning.qubit simulator for shot-free optimization and the default.mixed device for noise-aware evaluation, demonstrating the potential of weighted quantum ensembles for reliable biomedical classification.
This study looks at how the entangling structure of a variational quantum classifier (VQC) relates to its performance in breast cancer diagnosis, using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. We tested three three-qubit configurations that share the same EfficientSU2 ansatz, COBYLA optimizer, and stratified five-fold cross-validation, but differ in their entangling structure: a single- repetition Ising-type coupling (A), a two-repetition linear Ising-type coupling (B), and a fully connected Heisenberg-type coupling (C). Entanglement was measured with the von Neumann entropy and the Wootters concurrence. From A to C, the mean entropy rose from 0.40 to 0.71, accuracy rose from 92.98% to 93.68%, and F1-score rose from 90.32% to 91.28%. Configuration C was also the most stable across folds. However, the fold-level correlations between entanglement and performance were weak and not significant. Richer topologies also increase circuit expressivity, so the two effects cannot be fully separated. These results show an association, not a proven causal link, between entangling structure and VQC performance.
Introduction: In recent years, high-dimensional clinical and genomic data have gained significant importance for prognosis and personalized medicine in breast cancer. But the use of quantum machine learning (QML) on such data is limited by the availability of few qubits, the computation time of quantum simulation, and dimensionality reduction. This work systematically compares several QML architectures for breast cancer classification in the presence of realistic and simulator constraints. Methods: The experiments were performed on a dataset of METABRIC breast cancer patients (2,509 patients). After handling missing values and one-hot encoding, there were 63 features in the processed data set. The feature space was reduced by Principal Component Analysis (PCA) to 12, 4 and 2 components for the implementations of quantum computers, respectively, with 56.10 ± 0.14%, 26.78 ± 0.57% and 16.23 ± 0.33% of the variance retained. Three QML models were tested: Quantum Neural Networks (QNN), Quantum K-Nearest Neighbors (QKNN), and Quantum Support Vector Machines (QSVM), with the models being simulated. Seven classical classification models were tested: Logistic Regression, SVM with RBF kernel, K-Nearest Neighbors, Random Forest, XGBoost, LightGBM and Multilayer Perceptron, both with PCA-matched and full 63-feature representation. All primary results are reported with 5-fold cross validation. Results: Among the evaluated QML architectures, QKNN using 12 principal components achieved the strongest performance, attaining an accuracy of **75.11% ± 3.76%**, an F1-score of **0.7088 ± 0.0433**, and a ROC-AUC of **0.8148 ± 0.0394**. Even though all of the QML models performed significantly poorly in comparison to classical models trained on the entire 63-feature data set, the latter models were able to achieve about **94% accuracy** with XGBoost, Random Forest, and Logistic Regression. The comparison also showed the effect of information loss due to PCA is significant in predictive performance in both classical and quantum models. Discussion: The results show that for high dimensional breast cancer data, currently available simulator-based QML models can learn meaningful patterns with limited quantum resources, but are not as effective as powerful classical machine learning models when complete feature representations are available. The study does not report any sort of quantum advantage or clinical use, but rather a benchmark of current QML architectures that has been rigorously calculated and repeated, with a focus on the impact of dimensionality reduction, validation approaches, and simulator limitations, as well as outlining challenges that need to be overcome prior to practical implementation on real quantum hardware.
Background: Breast cancer remains a leading cause of cancer-related mortality, and reliable computational decision support is increasingly viewed as a complement to expert pathological assessment rather than a replacement for it. Variational quantum classifiers (VQCs) and Quantum Support Vector Machines (QSVMs) have recently been promoted as candidate models for medical classification, yet most published comparisons rely on internal hold-out validation alone and report only a single point estimate of discrimination, omitting calibration, decision-analytic value, and explainability—three ingredients that any clinically credible model must furnish. Methods: We assembled a complete quantum–classical machine learning pipeline and evaluated it under a deliberately stringent protocol designed to expose, rather than conceal, the limitations of current Noisy Intermediate-Scale Quantum (NISQ)-era models. The analytical hypothesis was conservative and stated in advance; in light of saturated classical baselines on this benchmark, we did not anticipate a quantum advantage in raw discrimination, and we framed the study as a methodological probe rather than as a competition. Using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset (n = 569) for development and an independent Wisconsin Original (WBC) cohort (n = 683) for external validation, we benchmarked five classical learners (XGBoost, LightGBM, CatBoost, RandomForest, RBF-SVM), two quantum models (an eight-qubit VQC implemented in PennyLane and a ZZ-feature-map QSVM implemented in Qiskit), and a stacked hybrid ensemble. The evaluation framework combined Optuna-driven hyperparameter optimisation, internal–external cross-validation, and external validation on the independent WBC cohort. Robustness and interpretability were then probed through circuit depth and embedding rotation ablation, depolarising noise stress tests, learning curve and feature stability analysis, decision curve analysis, and dual SHAP-based explanations covering both a direct tree-based explanation and a quantum surrogate. Reporting followed the TRIPOD + AI guideline. Results: On the internal test partition, RBF-SVM achieved the highest discrimination (AUC = 0.998), with XGBoost, LightGBM, CatBoost, the hybrid ensemble, and the VQC clustering between 0.992 and 0.996; the QSVM with a ZZ-fidelity kernel underperformed substantially (AUC = 0.727). Pairwise tests for correlated ROC curves indicated that most differences among top models were not statistically significant. On the external WBC cohort, model rankings reorganised, as RBF-SVM (AUC = 0.986, 95% CI 0.946–0.997), RandomForest (0.985, 95% CI 0.945–0.996), VQC (0.983, 95% CI 0.942–0.995), and the hybrid ensemble (0.982, 95% CI 0.941–0.995) all retained near-ceiling discrimination with extensively overlapping confidence intervals. Ablation analysis demonstrated that the choice of embedding rotation is decisive—Z-rotation embeddings collapsed VQC performance to chance levels (AUC ≈ 0.50), whereas X- and Y-rotations preserved it. Depolarising noise up to p = 0.10 had a negligible effect on the VQC, and SHAP analyses converged on worst concave points, mean concave points, and worst area as the dominant predictors across both classical and quantum models. Decision curve analysis showed positive net benefit for both classical and hybrid models across the clinically meaningful threshold range, exceeding both the treat-all and treat-none reference strategies throughout. Conclusions: In the present regime, the principal contribution of QML is not raw discrimination—modern classical learners are already at the data ceiling—but the construction of a rigorous, reproducible, externally validated, and interpretable benchmarking framework in which quantum models can be fairly compared with their classical counterparts. Because evaluation was confined to curated benchmark datasets rather than real-world clinical populations, the interpretability and net benefit findings reported here should be read as benchmark-level evidence and not as a demonstration of readiness for clinical deployment.