Abstract Background Cancer is a leading cause of death worldwide. While routine diagnosis of cancer is performed mainly with biopsy sampling, it is suboptimal to accurately characterize tumor heterogeneity. Positron emission tomography (PET)-driven radiomic research has demonstrated promising results when predicting clinical endpoints. This study aimed to investigate the added value of quantum machine learning both in simulator and in real quantum computers utilizing error mitigation techniques to predict clinical endpoints in various PET cancer patients. Methods Previously published PET radiomics datasets including 11C-MET PET glioma, 68GA-PSMA-11 PET prostate and lung 18F-FDG PET with 3-year survival, low-vs-high Gleason risk and 2-year survival as clinical endpoints respectively were utilized in this study. Redundancy reduction with 0.7, 0.8, and 0.9 Spearman rank thresholds (SRT), followed by selecting 8 and 16 features from all cohorts, was performed, resulting in 18 dataset variants. Quantum advantage was estimated by Geometric Difference (GD Q ) score in each dataset variant. Five classic machine learning (CML) and their quantum versions (QML) were trained and tested in simulator environments across the dataset variants. Quantum circuit optimization and error mitigation were performed, followed by training and testing selected QML methods on the 21-qubit IonQ Aria quantum computer. Predictive performances were estimated by test balanced accuracy (BACC) values. Results On average, QML outperformed CML in simulator environments with 16-features (BACC 70% and 69%, respectively), while with 8-features, CML outperformed QML with + 1%. The highest average QML advantage was + 4%. The GD Q scores were ≤ 1.0 in all the 8-feature cases, while they were > 1.0 when QML outperformed CML in 9 out of 11 cases. The test BACC of selected QML methods and datasets in the IonQ device without error mitigation (EM) were 69.94% BACC, while EM increased test BACC to 75.66% (76.77% in noiseless simulators). Conclusions We demonstrated that with error mitigation, quantum advantage can be achieved in real existing quantum computers when predicting clinical endpoints in clinically relevant PET cancer cohorts. Quantum advantage can already be achieved in simulator environments in these cohorts when relying on QML.
Prostate cancer is among the most prevalent malignancies in men, and early, accurate diagnosis is critical for effective treatment. Traditional machine learning techniques have demonstrated success in analyzing multiparametric magnetic resonance imaging (MRI) and clinical biomarkers; however, their scalability and capacity to model complex feature interactions remain limited. This study proposes an Ensemble Quantum Machine Learning (QML) framework to enhance prostate cancer detection using the PROSTATEx Challenge dataset. Radiomic features were extracted from MRI modalities and clinical attributes, then standardized and reduced using principal component analysis (PCA) to match current quantum hardware constraints. Three quantum classifiers—Quantum Support Vector Machine (QSVM), Variational Quantum Classifier (VQC), and Quantum Neural Network (QNN) - were independently trained and integrated using both soft voting and stacked ensemble strategies. Results from stratified 5-fold cross-validation show that the stacked ensemble outperformed individual models and baseline classifiers, achieving an average accuracy of 88.4%, recall of 89.1%, precision of 87.9%, F1-score of 88.5%, and area under the curve - receiver operating characteristic (AUC-ROC) of 0.94 (stack ensemble). These findings highlight the potential of hybrid quantum-classical ensemble learning to improve diagnostic robustness, particularly in reducing false negatives. Furthermore, validation on real quantum hardware demonstrated consistent performance, underscoring the feasibility of QML in near-term medical applications. This work contributes to the growing intersection of quantum computing and clinical AI, offering a scalable and interpretable approach to precision oncology.
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.
The combination of quantum computing with machine learning which forms quantum machine learning (QML) presents promising advances for healthcare purposes with specific applications in predicting cancer outcomes. The research investigates how QML can work together with the quantitative imaging analysis techniques in radiomics to improve prediction methods for lung cancer results. The proposed quantum neural network (QNN) model attempts to predict lung cancer patient 5-year survival status of lung cancer patient by radiomic features extracted from CT scan images. Quantum artificial intelligence showed promise in lung cancer prognosis because its evaluation included a balanced accuracy score of 0.71 and ROC-AUC performance metric of 0.71. The current approach has three major challenges due to data biases, restricted quantum hardware capabilities and insufficient evidence to demonstrate the quantum speedup benefits. In this study QML's clinical potential was noted in this research with operational and developmental pathways moving forward.
Abstract This study presents a comparison between classical machine learning (ML) algorithms and their quantum-enhanced counterparts in classifying scikit’s breast cancer dataset. Specifically, the research focused on evaluating the accuracy and time computational efficiency of quantum-enhanced machine learning (QEML) algorithms against traditional or classical machine learning algorithms. The quantum algorithms employed include the quantum support vector machine (qSVM), variational quantum classifier (VQC) and quantum k-nearest neighbor (qkNN). The qSVM and qkNN algorithms discussed here are really kernel-based hybrid classical-quantum algorithms as opposed to fully quantum algorithms and the VQC is a hybrid classical-quantum algorithm. These algorithms were tested against their classical equivalents: support vector machine (SVM), k-nearest neighbor (kNN), and artificial neural network (ANN). The qSVM algorithm demonstrated an 90% accuracy in distinguishing malignant from benign cancer cases (the classical support vector machine algorithm had a 99% accuracy), while qkNN achieved 90% accuracy (the classical kNN algorithm had an accuracy of 97%), and VQC reported 88.77± 0.26% (averaged over 5 runs) (this was compared with an ANN which had an accuracy of 96%). The accuracies of the classical SVM, kNN, and ANN algorithms were recorded, to provide a basis for comparison. The results indicated that while QEML algorithms perform reasonably well in classification tasks, however they do not consistently outperform classical algorithms. Additionally, the study measured the time efficiency of each algorithm. The ANN required 33.8 seconds, kNN required 0.21 seconds, and SVM required 1.98 seconds to perform the classification tasks. In contrast, the quantum algorithms were significantly slower, with the VQC taking 632.53 seconds, qkNN requiring 495.34 seconds, and qSVM requiring 538.05 seconds (This is not fixed per run but give an impression of the time scales to run the algorithms). We emphasize that this simulation runtime does not reflect actual quantum hardware performance, where gate execution and circuit depth constraints differ significantly. To mitigate this ambiguity, we distinguish between simulation runtime and estimated quantum runtime where relevant. This work potentially highlights the cases where it may not offer any advantage to use quantum machine learning algorithms. The Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) analysis are also considered in this work where the values measured for the SVM, kNN and ANN were 99%, 99% and 99% respectively and for the qSVM, qkNN and VQC were 87%, 86% and 86± 0.2% (averaged over 5 runs) respectively. Overall, this research contributes valuable insights into the comparative performance of classical and quantum ML algorithms in a specific application, underscoring the potential and current limitations of quantum approaches in machine learning tasks. Hyperparameter optimization was done for all the algorithms, both quantum and classical, with Optuna. This was primarily because the QEML algorithms had more parameters and required considerable computational resources to optimize. This could have influenced the performance outcomes and is the basis for future work. The development of key comparison criteria is critical in the development of quantum machine learning, and this work offers some building blocks for this.
The rapid growth of computer vision and increasingly complex image recognition tasks has exposed fundamental computational limitations of classical machine learning models, motivating the exploration of quantum computing as an emerging new paradigm. This paper presents a comprehensive benchmarking study of classical and quantum machine learning models for image recognition on the MNIST handwritten digit dataset, evaluating both traditional models, a Classical Support Vector Machine (CSVM) and a Quantum Support Vector Machine (QSVM), and deep neural network models, a Classical Convolutional Neural Network (CCNN) and a Quantum Convolutional Neural Network (QCNN), across four performance dimensions: classification accuracy, computational runtime, parameter count, and memory requirements. Experiments are conducted as functions of both feature dimensionality and sample size, and across CPU and GPU execution environments, providing a controlled, multidimensional comparison to address gaps in prior work. For the SVM-based models, QSVM consistently outperforms CSVM in accuracy, reaching ∼ 0.90 versus ∼ 0.85 at 1,000 samples, with a higher computational cost. A feature count of 10 qubits and a sample size in the range of 200 -- 500 emerge as practical operating points that balance accuracy and runtime. For the neural network models, CCNN and QCNN achieve comparable classification accuracy, both exceeding 0.96 at 64 features and 60,000 samples, yet QCNN offers superior parameter and memory efficiency at higher feature counts, while incurring higher runtime. Across both model families, quantum models consistently outperform classical models by greater margins in accuracy as feature dimensionality or sample size increases.