Hybrid Quantum Neural Networks (HQNNs) have recently emerged as a promising paradigm for near-term quantum machine learning. However, their practical performance strongly depends on design choices such as classical-to-quantum data encoding, quantum circuit architecture, measurement strategy and shots. In this paper, we present a comprehensive design space exploration of HQNNs for Chronic Kidney Disease (CKD) diagnosis. Using a carefully curated and preprocessed clinical dataset, we benchmark 625 different HQNN models obtained by combining five encoding schemes, five entanglement architectures, five measurement strategies, and five different shot settings. To ensure fair and robust evaluation, all models are trained using 10-fold stratified cross-validation and assessed on a test set using a comprehensive set of metrics, including accuracy, area under the curve (AUC), F1-score, and a composite performance score. Our results reveal strong and non-trivial interactions between encoding choices and circuit architectures, showing that high performance does not necessarily require large parameter counts or complex circuits. In particular, we find that compact architectures combined with appropriate encodings (e.g., IQP with Ring entanglement) can achieve the best trade-off between accuracy, robustness, and efficiency. Beyond absolute performance analysis, we also provide actionable insights into how different design dimensions influence learning behavior in HQNNs.
Hybrid Quantum Neural Networks (HQNNs) have recently emerged as a promising paradigm for near-term quantum machine learning. However, their practical performance strongly depends on design choices such as classical-to-quantum data encoding, quantum circuit architecture, measurement strategy and shots. In this paper, we present a comprehensive design space exploration of HQNNs for Chronic Kidney Disease (CKD) diagnosis. Using a carefully curated and preprocessed clinical dataset, we benchmark 625 different HQNN models obtained by combining five encoding schemes, five entanglement architectures, five measurement strategies, and five different shot settings. To ensure fair and robust evaluation, all models are trained using 10-fold stratified cross-validation and assessed on a test set using a comprehensive set of metrics, including accuracy, area under the curve (AUC), F1-score, and a composite performance score. Our results reveal strong and non-trivial interactions between encoding choices and circuit architectures, showing that high performance does not necessarily require large parameter counts or complex circuits. In particular, we find that compact architectures combined with appropriate encodings (e.g., IQP with Ring entanglement) can achieve the best trade-off between accuracy, robustness, and efficiency. Beyond absolute performance analysis, we also provide actionable insights into how different design dimensions influence learning behavior in HQNNs.
Abstract Chronic Kidney Disease (CKD) is a progressive and potentially fatal condition that is often diagnosed at advanced stages due to its asymptomatic nature in early progression. Accurate early prediction is challenging because of limited, heterogeneous clinical data and complex nonlinear relationships among biomarkers. This paper proposes HCF-QNet, a hybrid quantum–classical learning framework for robust CKD prediction. The proposed approach integrates three complementary components: optimized feature selection, contrastive representation learning, and variational quantum classification. Initially, a Grey Wolf Optimizer (GWO) is employed to select the most informative clinical features, reducing redundancy and dimensionality. The selected features are then transformed into a compact latent space using a supervised contrastive encoder, which enhances intra class compactness and inter class separability. These embeddings are subsequently processed by an 8-qubit Variational Quantum Neural Network (VQNN) with ring-based entanglement, enabling expressive nonlinear modeling beyond classical classifiers. A two-stage training strategy is adopted, where the encoder is contrastively pretrained and frozen during quantum fine-tuning to ensure stable optimization. The model is evaluated on the UCI CKD dataset using five fold cross-validation. Experimental results demonstrate an average accuracy of 99.75%, achieving better predictive performance than most existing classical machine learning and deep learning approaches. The results highlight the potential of hybrid quantum–classical models for accurate medical diagnosis on small and heterogeneous clinical datasets.
Quantum machine learning has emerged as a promising application domain for near-term quantum hardware, particularly through hybrid quantum-classical models that leverage both classical and quantum processing. Although numerous hybrid architectures have been proposed and demonstrated successfully on benchmark tasks, a significant open question remains regarding the specific contribution of quantum components to the overall performance of these models. In this work, we aim to shed light on the impact of quantum processing within hybrid quantum-classical neural network architectures through a rigorous statistical study under ideal, noise-free simulation. We systematically assess a selection of commonly used HQNN architectures on three binary medical classification tasks covering medical signal data as well as planar and volumetric images. In doing so, we examine the influence attributable to classical and quantum aspects such as encoding schemes, entanglement, and circuit size. Within this experimental setting, we find that in best-case scenarios, hybrid models show performance comparable to their classical counterparts. However, in most cases, performance metrics deteriorate when the investigated QNN components replace matched classical processing layers. Our multi-modal analysis provides empirical insights into the contributions of the tested quantum components and advocates for cautious claims and design choices for hybrid models in near-term applications.
Hybrid quantum-classical neural networks (HQNNs) are emerging as a practical approach for quantum machine learning in the noisy intermediate-scale quantum (NISQ) era, as they combine classical learning components with parameterized quantum circuits in an end-to-end trainable framework. However, their performance and efficiency depend strongly on architectural choices such as data encoding, circuit structure, measurement design, and the coupling between classical and quantum modules. This makes manual design increasingly difficult, especially when hardware limitations and resource constraints must also be taken into account. In this paper, we study the foundations of HQNNs and neural architecture search (NAS), discuss how NAS extends to quantum and hybrid settings, and demonstrate FLOPs-aware search (where FLOPs serve as a proxy for computational complexity), as an important hardware-aware direction for building HQNNs that are not only accurate but also computationally efficient and practically deployable.
In this study, we develop and compare quantum and classical machine learning-based chronic kidney disease prediction models. We used the "Chronic_Kidney_Disease Data Set" of the UCI Machine Learning Repository. We performed data preprocessing and applied feature engineering techniques to select the best features. We developed two quantum machine learning-based models and two classical machine learning-based models. We used a hybrid classical-quantum environment for building quantum machine learning models. Finally, we compared the performances of all four models. We found that the Quantum Support Vector Machine performs best among the quantum models. The model’s accuracy was 95% with a k-fold cross-validation score of 94.5% and an ROC-AUC score of 0.987. Among the classical models, the Support Vector Machine showed the highest performance with an accuracy of 92.5%, a k-fold cross-validation score of 93.9%, and a ROC-AUC score of 0.974. Overall, the Quantum Support Vector Machine outperformed all other developed models in terms of accuracy and validation scores. If the quantum models can be executed in a quantum computer instead of a hybrid environment, the models will exhibit higher accuracy and faster execution time. As the models are precisely predicting chronic kidney disease with high accuracy, we believe this study will act as an inspiring framework in the less-investigated field of quantum machine learning-based chronic kidney disease prediction.
Chronic Kidney Disease (CKD) is a serious global health problem and if not detected early, can lead to severe complications including hypertension, anaemia and renal failure. Accurate and quick prediction of chronic kidney disease (CKD) remains a crucial research area in health care analytics. While classical machine learning approaches have strong prediction performance, the growing complexity of data and the nonlinear relationships between features demands more sophisticated computational models. In this research, a hybrid system called the QML-CKDP that combines classical machine learning techniques with quantum support vector machines (QSVM) is proposed to improve chronic kidney disease (CKD) forecasting. The system is equipped with an extensive data preparation pipeline that includes hierarchical encoding, missing value imputation and min-max normalization. To improve prediction accuracy and reduce dimensionality, feature selection is performed using various methods, including correlation analysis, chi-square test, variance threshold, recursive feature elimination (RFE), sequential forward selection (SFS), lasso regression and ridge regression. To evaluate the system, a variety of classical machine learning classifiers are used such as logistic regression, random forest, AdaBoost, gradient boosting, extreme gradient boosting, Naive Bayes, decision trees, and classical support vector machines (SVMs). These are complemented with quantum support vector machine (QSVM) models that use quantum kernel estimation with variational quantum circuits. Does. The QSVM model shows high predictive performance and good representation of nonlinear features in highdimensional Hilbert spaces, and is likely to be scalable to large biomedical data sets. However, current limitations of quantum technology impact on implementation speed compared to classical SVM models.
Quantum Machine Learning represents a promising approach to address complex computational challenges beyond classical computing capabilities. In this study, we investigate its application to healthcare, focusing specifically on the classification of cardiovascular diseases. Classical Machine Learning methods often struggle with high-dimensional data, resulting in limited accuracy and reduced generalization, especially in real-world medical scenarios. To overcome these issues, we propose a Hybrid Quantum-Neural Network that combines classical Autoencoders for dimensionality reduction with Quantum Neural Networks, facilitating effective data processing compatible with current NISQ hardware. Evaluated on the Cleveland dataset, our HQNN achieves state-of-the-art classification accuracy up to 90.98%, outperforming purely classical counterparts by capturing complex feature relationships in the quantum domain. We further demonstrate the effectiveness of the proposed HQNN model in scenarios characterized by limited training data and realistic noisy quantum environments, performing robustness analyses and simulations using IBM’s quantum hardware framework. These findings highlight the practical advantages of our Hybrid Quantum-Neural Network, underscoring its suitability and immediate applicability for real-world healthcare scenarios. The code is made publicly available.