Accurate detection and classification of WBC is essential haematological diagnosis, especially to detect the development of lymphoblastic leukaemia.Class imbalance, limited interpretability, and high computational demands are always prevalent issues in classical machine and deep learning.In this paper, a variational quantum classifer (VQC) based hybrid quantum-classical approach to five-class WBC subtype identification (Basophil, Eosinophil, Lymphocyte, Monocyte, Neutrophil) and four-stage detection of lymphoblastic leukaemia (Benign, Early, Pre-leukaemic, Pro-leukaemic) is proposed.The proposed pipeline combines the use of CNN based feature extraction, PCA dimensionality reduction and amplitude encoding within a 3-qubit quantum system parameterised by multi-layered RY rotations, controlled-Z (CZ) and controlled-phase (CPHASE) gates, with optimisation performed using the AdamW algorithm with early stopping.The approaches used to cope with class imbalance are SMOTE and inverse-frequency weighted cross-entropy loss.Explanation is given using SHAP DeepExplainer and gradient based saliency map.The model classifies WBC into 4 categories with an accuracy of 96.8% on macro-averaging and detects the stage of a disease with an accuracy of 95.4% while having only 25-30 trainable parameters, outperforming classical classifiers such as SVM, Random Forest, standard CNN and MobileNetV2.The results show that quantum inspired methods provide a promising route to scalable, parameter-efficient, and interpretable biomedical image analysis.
Accurate detection and classification of WBC is essential haematological diagnosis, especially to detect the development of lymphoblastic leukaemia.Class imbalance, limited interpretability, and high computational demands are always prevalent issues in classical machine and deep learning.In this paper, a variational quantum classifer (VQC) based hybrid quantum-classical approach to five-class WBC subtype identification (Basophil, Eosinophil, Lymphocyte, Monocyte, Neutrophil) and four-stage detection of lymphoblastic leukaemia (Benign, Early, Pre-leukaemic, Pro-leukaemic) is proposed.The proposed pipeline combines the use of CNN based feature extraction, PCA dimensionality reduction and amplitude encoding within a 3-qubit quantum system parameterised by multi-layered RY rotations, controlled-Z (CZ) and controlled-phase (CPHASE) gates, with optimisation performed using the AdamW algorithm with early stopping.The approaches used to cope with class imbalance are SMOTE and inverse-frequency weighted cross-entropy loss.Explanation is given using SHAP DeepExplainer and gradient based saliency map.The model classifies WBC into 4 categories with an accuracy of 96.8% on macro-averaging and detects the stage of a disease with an accuracy of 95.4% while having only 25-30 trainable parameters, outperforming classical classifiers such as SVM, Random Forest, standard CNN and MobileNetV2.The results show that quantum inspired methods provide a promising route to scalable, parameter-efficient, and interpretable biomedical image analysis.
Abstract Leukemia is a life-threatening blood cancer that requires early and accurate diagnosis to improve patient outcomes. Traditional diagnostic methods based on manual examination of blood smear images are time-consuming and prone to human error. With the advancement of artificial intelligence, automated detection systems using machine learning and deep learning techniques have shown promising results. However, classical models face limitations in handling complex feature spaces efficiently. To address this, this project proposes a hybrid approach titled “Leukemia Detection using Quantum Machine Learning”, which integrates classical image processing with quantum computing techniques. In this work, microscopic blood smear images from a publicly available Acute Lymphoblastic Leukemia dataset are utilized. The images undergo preprocessing steps including resizing, noise removal using Gaussian filtering, and normalization to enhance data quality. A Convolutional Neural Network (CNN) is employed for classical feature extraction, transforming image data into meaningful feature vectors. These features are further reduced using Principal Component Analysis (PCA) to make them suitable for quantum processing. The reduced feature set is encoded into quantum states using both angle encoding and amplitude encoding techniques. Quantum properties such as superposition and entanglement are leveraged through parameterized quantum circuits. Two quantum machine learning models, namely Quantum Support Vector Machine (QSVM) and Quantum Neural Network (QNN), are implemented to perform multi-class classification of leukemia stages, including Benign, Early, Pre, and Pro. The performance of both models is evaluated and compared using metrics such as accuracy, confusion matrix, and classification report. The proposed hybrid classical-quantum framework demonstrates the potential of quantum machine learning in medical image analysis. By combining deep learning with quantum algorithms, the system aims to improve diagnostic accuracy and computational efficiency. This approach not only enhances leukemia detection but also highlights the applicability of quantum computing in real-world healthcare problems, paving the way for future advancements in intelligent medical diagnosis systems.
Accurate classification of blood cells is crucial for early diagnosis and monitoring of hematological disorders. However, manual microscopy is time-consuming, subjective, and prone to inter-observer variability. Deep learning has shown potential in automating this task, but its high computational demands limit its use in resource-constrained settings. This study aims to develop an efficient, hybrid quantum-classical model for blood cell classification using quantum simulation to reduce computational complexity without compromising accuracy. We propose a hybrid framework that integrates a 6-qubit variational quantum circuit (VQC) with a classical EfficientNet-B0 model. The model was trained and tested on the full BloodMNIST dataset, which includes 17,092 microscopic images across eight blood cell types. The proposed model achieved a test accuracy of 96.58% and a validation accuracy of 96.90%. F1-scores ranged from 91.97% to 99.84%, with notable results in neutrophils (98.80%) and lymphocytes (98.36%). These findings confirm the feasibility of combining quantum and classical approaches for medical image classification. The proposed hybrid model shows strong promise for practical use in diagnostic systems, especially in low-resource settings, and offers a pathway for future research in quantum-enhanced medical imaging.
Peripheral blood cell classification is critical for computer-aided haematological diagnosis. In the present study, a hybrid classical-quantum deep learning framework for multi-class classification using the BloodMNIST dataset. Operating directly on native 28×28 images, our pipeline preserves morphological features while minimizing computational complexity. A lightweight convolutional network extracts features that are normalized, reduced via Principal Component Analysis, and encoded into an 8-qubit variational quantum circuit (HybridQNet). The proposed model achieves 96.26% validation and 95.12% test accuracy, demonstrating an efficient, scalable approach for biomedical image analysis.
Abstract Neuro-imaging data is necessary for timely and correct diagnosis of the disease Alzheimer’s (AD). Early clinical intervention is challenging due to the high-dimensional nature of the images, making it computationally difficult to work on representations, sparse annotated datasets and inter-subject variance. Deep convolutional VGG, ResNet, DenseNet, MobileNet and other neural networks (CNNs) have shown good performance. Diagnostic performance, but they are based on a high number of parameters and require highly trained personnel. Data can often times restrict resilience to data-skimy clinical settings. To address these challenges, the current paper presents a hybrid Quantum Machine Learning (QML) architecture for Alzheimer’s disease Detection and classification based on a combination of classical CNN-based feature extraction and Variational quantum classifier. In the suggested method, diseased, relevant, and compact features are extracted using pre-trained CNN backbones. Data from structural MRI scans is later translated into quantum states and computed. With parameterised variational quantum circuits optimised with a hybrid quantum-classical learning loop. The proposed method has been experimentally evaluated on benchmark datasets of MRI scans of Alzheimer’s patients. The hybrid CNNQML model has a mean classification accuracy of about 93% which is similar to deep CNN models, including ResNet and DenseNet (94%), and much more. Beating lightweight models like MobileNet and shallow CNNs by six percentage points. It is worth noting that the hybrid QML model performs better in conditions with limited training data. Performance degradation is observed in accuracy and AUC for all CNN variants, with up to 5% lower performance to deep CNNs. Cross-validation, confidence interval, paired significance test, and statistical validation. Effect size analysis proves the fact that the hybrid model produces statistically significant performance, as well as cases of clinically meaningful improvements over lightweight CNN baselines. Similar to the state-of-the-art deep networks with significantly reduced trainable parameters in the classification stage. These findings show that hybrid QML is a data-efficient and powerful substitute in the diagnosis of Alzheimer’s. On the whole, this is a work that sets a practical way to integrate quantum machine learning into clinical neuro-imaging pipelines and hybrid settings. QML has supplementary deep-learning enhancement to classics in next-generation Alzheimer’s Decision-support systems.
Accurate and computationally efficient classification of brain tumors from MRI scans is critical for deployment in resource-constrained clinical settings, where conventional deep CNN architectures-despite achieving strong accuracy-often require large parameter counts that limit scalability. This work proposes a hybrid quantum-classical deep learning framework that combines a classical convolutional feature extractor with a variational quantum circuit (VQC) for multi-class tumor classification. Spatial features extracted by the CNN backbone will be dimensionality-reduced and encoded into quantum states using angle encoding; a parameterized quantum circuit will then process these features in a higher-dimensional Hilbert space, leveraging superposition and entanglement to capture non-linear feature correlations. Measurement outputs from the quantum layer will be passed to a classical fully connected layer for classification into glioma, meningioma, pituitary tumor, and no-tumor categories. The proposed framework will be evaluated on a publicly available benchmark MRI dataset and benchmarked against state-of-the-art classical CNN architectures in terms of accuracy, precision, recall, F1-score, and parameter efficiency. This study aims to demonstrate that quantum-enhanced layers can achieve competitive classification performance while substantially reducing trainable parameters, offering a practical direction for efficient, quantum-assisted medical image analysis.
Alzheimer's disease (AD) presents significant diagnostic challenges owing to the subtle morphological similarities observed in the early stages, with traditional deep learning approaches often struggling to distinguish between the various stages of disease progression via structural Magnetic Resonance Imaging (MRI) data. Quantum computing offers unique advantages for medical image analysis, leveraging superposition and entanglement capabilities to process high-dimensional feature spaces beyond the limits of classical computation. This study introduces a hybrid quantum-classical neural network architecture (HQC-Net) for accurate four-class Alzheimer's disease classification, which uses quantum processing to detect patterns that are often invisible to classical spatial analysis methods. The proposed framework integrates classical feature extractors, including a custom CNN and modified ResNet18, with six-qubit variational quantum circuits that employ multiaxis rotation encoding (RY→RZ→RX), a quantum Fourier transform for spectral decomposition, and multihead attention for effective quantum-classical feature fusion. Comprehensive evaluations were conducted on the Kaggle dataset (5,121 samples) and the OASIS dataset (20,000 samples), incorporating realistic quantum noise modelling, including depolarising, amplitude damping, and phase damping channels. The modified QResNet18 configuration achieved a test accuracy of 99.67%, with perfect discriminative capability (AUC = 1.0000) on the OASIS dataset. Quantum processing demonstrated superior detection of very mild dementia (99.86% accuracy), which is crucial for early intervention. The proposed approach outperformed existing quantum-enhanced methods by 3.57 percentage points while effectively handling the increased diagnostic complexity associated with four-class classification. This study demonstrates a practical quantum advantage for multiclass neuroimaging classification, achieving superior diagnostic accuracy while maintaining computational efficiency and clinical deployment feasibility under current Noisy Intermediate-Scale Quantum (NISQ) hardware constraints.