Machine Learning (ML) approaches have achieved high performance by discovering underlying data patterns and structures. Diabetes is a life-threatening disease that affects vital organs such as the eyes (retinopathy), nerves (neuropathy), heart and kidneys (nephropathy). Although various ML models have been proposed for diabetes prediction, most struggle to effectively handle large, multi-feature datasets. Quantum Machine Learning (QML), which integrates quantum computing principles with machine learning, offers a promising solution for complex classification tasks. However, effective feature extraction remains challenging on current NISQ (Noisy Intermediate-Scale Quantum) devices due to hardware limitations. In this study, a Differentiable Quantum Circuit with Classical PCA Feature Compression is proposed for medical disease classification using a SMOTE-ENN resampled dataset and a penalized binary cross-entropy cost function to enhance quantum learning. The efficient ansatz designed in this work plays a crucial role in determining the model’s learning capability and feature extraction performance. To evaluate the robustness and generalization capability of the proposed model, three categories of medical datasets are considered. The primary focus is on diabetes classification using four diabetes-related datasets, while additional experiments are conducted on liver disease and chronic kidney disease datasets to assess the adaptability of the proposed framework across multiple healthcare domains. Six datasets from UCI, Mendeley and IEEE repositories are used: PIMA Indian Diabetes, Type-2 Diabetes, a private RTML Insulin dataset, Indian Liver Patient Dataset (ILPD), Early-Stage Diabetes Risk Prediction Dataset (ESDRPD) and Chronic Kidney Disease (CKD). The proposed quantum model is simulated using the PennyLane quantum simulator and compared against preprocessed classical classifiers. Experimental results demonstrate remarkable efficiency and rapid convergence within 10--20 epochs, achieving high F1-scores of 96.15% (PIMA), 97.50% (Type-2), 98.04% (RTML), 89.47% (ILPD), 98.66% (ESDRPD) and 97.84% (CKD). These findings demonstrate that the proposed quantum model efficiently extracts relevant features and effectively handles medical disease classification tasks, highlighting its strong potential for practical healthcare applications.
Machine Learning (ML) approaches have achieved high performance by discovering underlying data patterns and structures. Diabetes is a life-threatening disease that affects vital organs such as the eyes (retinopathy), nerves (neuropathy), heart and kidneys (nephropathy). Although various ML models have been proposed for diabetes prediction, most struggle to effectively handle large, multi-feature datasets. Quantum Machine Learning (QML), which integrates quantum computing principles with machine learning, offers a promising solution for complex classification tasks. However, effective feature extraction remains challenging on current NISQ (Noisy Intermediate-Scale Quantum) devices due to hardware limitations. In this study, a Differentiable Quantum Circuit with Classical PCA Feature Compression is proposed for medical disease classification using a SMOTE-ENN resampled dataset and a penalized binary cross-entropy cost function to enhance quantum learning. The efficient ansatz designed in this work plays a crucial role in determining the model’s learning capability and feature extraction performance. To evaluate the robustness and generalization capability of the proposed model, three categories of medical datasets are considered. The primary focus is on diabetes classification using four diabetes-related datasets, while additional experiments are conducted on liver disease and chronic kidney disease datasets to assess the adaptability of the proposed framework across multiple healthcare domains. Six datasets from UCI, Mendeley and IEEE repositories are used: PIMA Indian Diabetes, Type-2 Diabetes, a private RTML Insulin dataset, Indian Liver Patient Dataset (ILPD), Early-Stage Diabetes Risk Prediction Dataset (ESDRPD) and Chronic Kidney Disease (CKD). The proposed quantum model is simulated using the PennyLane quantum simulator and compared against preprocessed classical classifiers. Experimental results demonstrate remarkable efficiency and rapid convergence within 10--20 epochs, achieving high F1-scores of 96.15% (PIMA), 97.50% (Type-2), 98.04% (RTML), 89.47% (ILPD), 98.66% (ESDRPD) and 97.84% (CKD). These findings demonstrate that the proposed quantum model efficiently extracts relevant features and effectively handles medical disease classification tasks, highlighting its strong potential for practical healthcare applications.
Early diagnosis of chronic diseases is paramount to improving patient survival rates, optimizing treatment trajectories, and mitigating escalating healthcare expenditures. While the recent explosion of heterogeneous healthcare data presents robust opportunities for predictive modeling, it simultaneously introduces severe computational bottlenecks regarding data dimensionality and non-linear complexity. This paper proposes a novel, high-performance hybrid prediction framework that seamlessly fuses classical machine learning topologies with Variational Quantum Circuits (VQCs) within a Quantum Machine Learning (QML) paradigm. The proposed architecture targets the multi-classification and risk profiling of five high-prevalence chronic conditions: diabetes, cardiovascular disease, breast cancer, chronic liver disease, and chronic kidney disease This project introduces “Quantum Health,” a Quantum Machine Learning (QML)-based multi disease prediction system that integrates classical and quantum computing to enhance diagnostic performance. It employs five quantum-enhanced models: Quantum Random Forest for breast cancer, Quantum Neural Network for diabetes, Quantum k-Nearest Neighbors (QkNN) for heart disease, Quantum AdaBoost for liver disease, and Quantum Naive Bayes for kidney disease. The models are evaluated using both classical methods and PennyLane simulations, enabling direct comparison of performance. The system is deployed as a Flask-based web application featuring a user-friendly dashboard, disease-specific prediction modules, prediction history with CSV export, and an analytics dashboard comparing classical and quantum accuracies. Each module provides predictions, confidence scores, and feature importance insights. Unlike monolithic classical models, our approach incorporates a parameterized quantum layer that maps low-dimensional classical medical vectors into high-dimensional Hilbert spaces, capturing intricate, non-linear correlations that evade classical kernels. Experimental evaluations executed on simulated quantum backends demonstrate that the proposed Quantum Hybrid Model drastically outperforms conventional baselines, achieving an overall classification accuracy of 96.8% and an Area Under the Curve (AUC) of 98.4%. The findings establish that integrating quantum circuit state vectors with ensemble classical architectures provides a scalable, computationally efficient, and highly generalizable paradigm for next-generation clinical decision support systems.
A chronic metabolic disease, diabetes mellitus has become a serious global public health concern. Using the Indian Pima Diabetes Dataset, this study compares Classical Support Vector Machines (CSVM) with Quantum Support Vector Machines (QSVM) for diabetes prediction. To address the challenges associated with high-dimensional data and feature redundancy, before classification, a feature reduction method called Principal Component Analysis (PCA) is used. Experimental results demonstrate that PCA effectively reduces computational complexity while preserving essential information required for classification. Furthermore, the study investigates the potential advantages of quantum-enhanced learning in healthcare prediction tasks by comparing the predictive capabilities and computational characteristics of QSVM against its classical counterpart. The findings provide insights into the applicability of quantum machine learning techniques for medical diagnosis and highlight the opportunities and challenges associated with implementing quantum algorithms on real-world healthcare datasets.
Diabetes remains one of the common chronic diseases around the world and finding a good early prediction from clinical and lifestyle data is very important for timely help. Classical machine learning models such as regression, decision trees and random forests have been used for this purpose but the complicated non‑linear connections among clinical features might need richer ways of representation. Quantum Machine Learning (QML) gives an approach using quantum feature encoding and variational circuits to find patterns that are expensive to express with classical methods. In this study we show a diabetes prediction pipeline that compares baseline models—logistic regression, decision tree and random forest—with two quantum classifiers built in Qiskit: a Quantum Support Vector Classifier (QSVC) and a Variational Quantum Classifier (VQC). Using the Pima Indians Diabetes dataset we test all models on accuracy, precision, recall and F1‑score. The results show that the classical Random Forest classifier reaches the accuracy (0.73) beating both quantum classifiers. The Quantum Support Vector Classifier (QSVC 0.66) and the Variational Quantum Classifier (VQC 0.65) are still competitive. Do not outdo the classical baselines. These results agree with reports that in the present Noisy Intermediate‑Scale Quantum (NISQ) era quantum models usually match rather than surpass well‑tuned classical approaches on standard health‑care data and that a real quantum advantage will need larger data, deeper circuits and testing, on real quantum hardware.
This paper presents a Quantum versus classical implemented of Machine learning (ML) algorithm applied to a diabetes dataset. Diabetes is a Sixth deadliest disease in the world and approximately 10 million new cases are registered every year worldwide. Using novel Quantum computing (QC) along with Quantum Machine Learning (QML) techniques in the healthcare system to improve and accelerate the computing of existing ML models that allows the different approach to understanding the complex patterns of the disease. The proposed system tackles a binary classification problem of patients with diabetes into two different classes: diabetes patients with acute diseases and diabetes patients without acute diseases. Our study compares classical and quantum algorithms, namely Decision Tree, Random Forest, Extreme Boosting Gradient and Adaboost, Qboost, Voting Model 1, Voting Model 2, Qboost Plus, New model 1 and New Model 2 along with an ensemble method which creates a strong classifier from a committee of weak classifiers. The results we achieved using the validation metrics of the New Model 1 showed an overall precision of 69%, a recall of 69%, an F1-Score of 69%, a specificity of 69% and an accuracy of 69% on our diabetes dataset, with an increase of the computation speed by 55 times in comparison of the classical system. Our study has proved that QC improves the computational speed and its inclusion in medical applications will deliver faster results to physicians and caregivers.
Heart disease is one of the major global causes of death, requiring early and precise detection for efficient treatment.With advancements in quantum computing, Variational Quantum Circuits (VQCs) have gained recognition as a valuable technique for improving learning models in medical diagnosis.This study further explores the application of VQC-based models for heart disease prediction, leveraging quantum computing to enhance classification accuracy.A hybrid quantum-classical model has been developed, which combines classical preprocessing for feature extraction from patient datasets with a VQC-based classifier for medical predictions.The quantum circuit using variational parameters is optimized to reduce the loss function and facilitate effective learning from complex medical data.This proposed approach is tested on benchmark heart disease datasets and evaluated against classical Machine Learning (ML) models, including Support Vector Machines (SVMs), Random Forests, and Neural Networks.The experimental results show that VQC-based models achieve high accuracy and outperform classical models, notably on high-dimensional, non-linearly separable data.The quantum advantage of VQCs lies in their ability to encode and process data in a huge Hilbert space.However, challenges such as noise in near-term quantum devices and scalability issues remain key.This research underscores the efficacy of VQCbased quantum machine learning in medical diagnosis, paving the way for future studies on quantum-enhanced healthcare solutions.Further advancements in quantum hardware and hybrid learning techniques could allow quantum-assisted heart disease prediction to become an effective clinical tool.
Abstract Background Diabetes, the fastest growing health emergency, has created several life-threatening challenges to public health globally. It is a metabolic disorder and triggers many other chronic diseases such as heart attack, diabetic nephropathy, brain strokes, etc. The prime objective of this work is to develop a prognosis tool based on the PIMA Indian Diabetes dataset that will help medical practitioners in reducing the lethality associated with diabetes. Methods Based on the features present in the dataset, two prediction models have been proposed by employing deep learning (DL) and quantum machine learning (QML) techniques. The accuracy has been used to evaluate the prediction capability of these developed models. The outlier rejection, filling missing values, and normalization have been used to uplift the discriminatory performance of these models. Also, the performance of these models has been compared against state-of-the-art models. Results The performance measures such as precision, accuracy, recall, F1 score, specificity, balanced accuracy, false detection rate, missed detection rate, and diagnostic odds ratio have been achieved as 0.90, 0.95, 0.95, 0.93, 0.95, 0.95, 0.03, 0.02, and 399.00 for DL model respectively, However for QML, these measures have been computed as 0.74, 0.86, 0.85, 0.79, 0.86, 0.86, 0.11, 0.05, and 35.89 respectively. Conclusion The proposed DL model has a high diabetes prediction accuracy as compared with the developed QML and existing state-of-the-art models. It also uplifts the performance by 1.06% compared to reported work. However, the performance of the QML model has been found as satisfactory and comparable with existing literature.