Deep learning has transformed radiology, but the computational cost of training and deploying modern models is outpacing what classical hardware can support. Quantum computing has been proposed as one response to this bottleneck: an n-qubit system can represent $2^n$ states simultaneously, although practical speedups require efficient algorithms, data encoding, and measurement strategies rather than superposition alone. This review introduces radiologists to five core quantum concepts, qubits, superposition, entanglement, quantum gates, and quantum circuits, and to the superconducting, trapped-ion, and photonic hardware platforms that define the current noisy intermediate-scale quantum (NISQ) era. We then summarize the quantum machine learning architectures most often applied to medical imaging (quantum support vector machines, quantum neural networks, variational quantum circuits, and quantum convolutional neural networks) and quantum image representations such as the flexible representation of quantum images (FRQI) and the novel enhanced quantum representation (NEQR). Reported applications include brain tumor classification and segmentation on MRI, pneumonia detection on chest radiographs, skin and breast lesion classification, CT and PET reconstruction, and ultrasound denoising. Nearly all published systems are hybrid quantum-classical pipelines evaluated on small, downsampled benchmark datasets or simulators, and clinical deployment is not yet feasible. We conclude with current limitations and translational priorities for radiology.
Deep learning has transformed radiology, but the computational cost of training and deploying modern models is outpacing what classical hardware can support. Quantum computing has been proposed as one response to this bottleneck: an n-qubit system can represent 2n states simultaneously, although practical speedups require efficient algorithms, data encoding, and measurement strategies rather than superposition alone. This review introduces radiologists to five core quantum concepts, qubits, superposition, entanglement, quantum gates, and quantum circuits, and to the superconducting, trapped-ion, and photonic hardware platforms that define the current noisy intermediate-scale quantum (NISQ) era. We then summarize the quantum machine learning architectures most often applied to medical imaging (quantum support vector machines, quantum neural networks, variational quantum circuits, and quantum convolutional neural networks) and quantum image representations such as the flexible representation of quantum images (FRQI) and the novel enhanced quantum representation (NEQR). Reported applications include brain tumor classification and segmentation on MRI, pneumonia detection on chest radiographs, skin and breast lesion classification, CT and PET reconstruction, and ultrasound denoising. Nearly all published systems are hybrid quantum-classical pipelines evaluated on small, downsampled benchmark datasets or simulators, and clinical deployment is not yet feasible. We conclude with current limitations and translational priorities for radiology.
Quantum Machine Learning (QML) is an emerging, powerful paradigm at the intersection of quantum computing and artificial intelligence, with the potential to enhance medical image analysis. We compared their performance with classical counterparts, while critically examining the gap between theoretical quantum advantages and the practical constraints of current Noisy Intermediate-Scale Quantum (NISQ) implementations. This review explored the current landscape of QML-based methodologies applied to key medical imaging modalities - Magnetic Resonance Imaging (MRI), Computed Tomography (CT), X-rays, and histopathology. We systematically analyzed how quantum-enhanced models were employed for critical tasks such as image classification, segmentation, and reconstruction, and we compared their performance with that of classical counterparts. The review delved into the architectural innovations that enabled these quantum approaches, including hybrid quantum-classical frameworks, variational quantum circuits, and quantum convolutional neural networks. Particular emphasis was given to the clinical relevance of QML applications, from early diagnostics to real-time surgery planning, and how quantum advantage might influence data efficiency, interpretability, along computational scalability. Finally, we highlighted current challenges in data encoding, noise resilience, and hardware limitations, and discussed promising future directions that could enable the practical deployment of QML in healthcare settings. This comprehensive review aimed to serve as a foundational reference for researchers and practitioners looking to harness quantum computing in medical imaging and precision medicine.
Abstract Quantum computing is set to revolutionize medical imaging and diagnostics by enhancing speed, accuracy, and the ability to process complex datasets, leading to improved patient outcomes. This review paper provides an extensive comparative study of how quantum-mechanical principles like superposition, entanglement, and interference can overcome the inherent limitations of classical binary computing in the medical field. We explore specific applications in MRI and CT scan reconstruction, the mathematical advantages of the Quantum Fourier Transform (QFT), the integration of Quantum Machine Learning (QML) for automated pathology detection, and the transformative potential of hybrid quantum-classical systems. The paper also addresses current hardware limitations such as qubit decoherence, evaluates materials science innovations in superconducting circuits, and provides a strategic roadmap for the future of quantum-enhanced healthcare. Keywords: Quantum Computing, Medical Imaging, MRI Reconstruction, Qubits, Quantum Machine Learning, Diagnostics, Materials Science Introduction The field of medical imaging has undergone significant transformations over the last few decades, progressing from simple analog X-ray plates to complex digital 3D reconstructions and functional metabolic imaging. However, as the medical community moves toward "Precision Medicine," the demand for ultra-high-resolution data and real-time diagnostic feedback is growing at an exponential rate. Classical computing systems, governed by Moore’s Law, are reaching a physical and algorithmic bottleneck. The sheer volume of data generated by modern 7-Tesla MRI, Dual-Energy CT scans, and high-resolution PET-CT systems is becoming increasingly difficult to process with traditional von Neumann architectures. In modern clinical settings, a single high-resolution volumetric scan can generate several gigabytes of raw data. This data requires intensive signal processing—often taking minutes or even hours—to reconstruct into a format that a radiologist can interpret. In emergency medicine, specifically for stroke or trauma patients, the latency of classical reconstruction algorithms can be the difference between recovery and permanent disability. Quantum computing introduces a fundamental paradigm shift. By leveraging quantum bits (qubits), which utilize the subatomic properties of superposition and entanglement, quantum computers can theoretically perform massive parallel computations that are mathematically impossible for classical binary systems. This paper provides a comprehensive review of these quantum paradigms, comparing their efficiency to classical methods and outlining the path toward clinical implementation. Theoretical Framework: The Physics of Quantum Advantage Classical computers operate on bits, representing a deterministic state of either 0 or 1. In medical imaging, this means algorithms must process pixels or voxels sequentially. Quantum computing utilizes the unique properties of quantum mechanics to alter the fundamental complexity classes of imaging tasks. 2.1 Superposition and Parallelism Unlike a bit, a qubit can exist in a state |⟩=|0⟩+|1⟩, where and are complex probability amplitudes such that ||2+||2=1. This allows n qubits to represent 2n states simultaneously. For a medical image consisting of 10241024 voxels, a quantum system can map the entire state space into a vastly smaller number of physical qubits (approximately 20 qubits for a million voxels). This allows for "Global Optimization," where the computer evaluates all possible image configurations at once to find the one with the least noise. 2.2 Entanglement and Non-Locality Entanglement allows qubits to be correlated in a way that exceeds classical physics. In image processing, this property is being researched to create "Quantum Sensors." These sensors use entangled photons or atoms to detect minute magnetic field variations in MRI far exceeding the sensitivity of classical induction coils. This could lead to MRI machines that do not require massive, expensive superconducting magnets, making them portable and more accessible. 2.3 Quantum Interference and Phase Estimation Quantum interference is used to bias the probability of measuring the correct answer in an algorithm. In image denoising and deconvolution, interference patterns can be engineered to amplify the signal (the anatomical structure) while canceling out the noise (electronic artifacts or motion blur). This process, known as Quantum Phase Estimation (QPE), is a core component of many imaging algorithms that provide a "Quantum Speedup." Applications in Medical Modalities 3.1 Accelerated MRI Reconstruction via Quantum Fourier Transform (QFT) MRI relies on the Fast Fourier Transform (FFT) to convert raw signals from the frequency domain (k-space) into the spatial domain. The classical FFT has a complexity of O(NlogN). In contrast, the Quantum Fourier Transform (QFT) operates with a complexity of O(log2N). This exponential speedup has several critical clinical implications: Real-time Guided Surgery: Surgeons currently rely on static images. QFT could allow for "Live MRI," providing updated feedback during neurosurgery as brain tissue shifts. Pediatric and Geriatric Care: Reducing scan times from 45 minutes to 45 seconds would eliminate the need for anesthesia in children who cannot remain still. Low-Field Imaging: Higher computational power can compensate for lower-quality raw signals, allowing for high-quality images from smaller, less expensive MRI units in rural areas. 3.2 CT and PET Optimization: The Radon Transform CT scans utilize the Radon transform to reconstruct 3D volumes from 2D projections. Classical iterative reconstruction (IR) is computationally expensive and often results in "staircase artifacts." Quantum algorithms, such as the HHL (Harrow-Hassidim-Lloyd) algorithm for linear systems of equations, can solve the massive matrices required for CT reconstruction with logarithmic scaling. This allows for higher resolution while simultaneously reducing the radiation dose required for the patient. 3.3 Comparative Analysis: A Detailed Quantitative Review Table 1: Comparative Analysis: Classical vs. Quantum Computing in Medical Imaging Feature Classical Computing (Binary) Quantum Computing (Qubit) Clinical Impact & Consequences Basic Data Unit Bits (0 or 1) Qubits (Superposition) Exponentially higher data density and state representation. Logic Gates Boolean Logic (AND, OR, NOT) Unitary Gates (Hadamard, CNOT) Ability to explore multiple logical paths simultaneously. Processing Power Sequential or Vectorized (CPU/GPU) Massively Parallel (Global) Enables real-time diagnostic feedback during surgical procedures. MRI Scaling O (N log N) Complexity O (log2 N) Complexity Theoretical 100x speedup in image reconstruction times. Algorithmic Aim Gradient Descent (Local Minima) Quantum Annealing (Global Minima) Superior noise reduction and image clarity by finding the "true" global optimum. Storage & Memory Linear Growth Exponential Growth (2n) Efficient management of high-dimensional datasets like whole-genome sequencing. Radiation Safety Higher Dose (to maintain SNR) Lower Dose (Q-enhanced SNR) Significant reduction in long-term radiation exposure for chronic patients. Biomarker Detection Limited by pixel resolution/noise Enhanced via Quantum Interference Early-stage detection of micro-metastases and neurodegenerative markers. Hardware State Mature (Silicon CMOS) Emerging (NISQ Era) Shift toward hybrid architectures in the immediate clinical future. The transition from classical to quantum paradigms represents a shift from linear deterministic processing to exponential probabilistic optimization. While classical systems are robust and reliable for standard data management, quantum systems address the "computational bottleneck" inherent in high-resolution volumetric imaging (7T MRI, 4D-CT). The primary clinical advantage identified is the acceleration of the Radon and Fourier transforms, which reduces patient wait times and allows for lower-dose diagnostic protocols without sacrificing image quality. Integration with Artificial Intelligence (AI) and Machine Learning The synergy between Quantum Science and Artificial Intelligence Quantum Machine Learning (QML) is a primary focus for current research at institutions like eminent University. 4.1 Quantum Neural Networks (QNN) and Feature Spaces In classical AI, deep learning models often struggle with the "curse of dimensionality" when analyzing 4D imaging datasets. QNNs utilize "Quantum Feature Maps" to project data into a high-dimensional Hilbert space where classes of data (e.g., malignant vs. benign) become linearly separable. This leads to significantly higher sensitivity in automated mammography and lung cancer screening. 4.2 Automated Segmentation and Pathology Detection Automated segmentation involves the computer outlining the boundaries of an organ or a tumor. Classical AI requires thousands of labeled images to learn these boundaries. Preliminary studies suggest that QML algorithms can achieve higher accuracy with 70% less training data because the quantum kernels are inherently better at recognizing complex geometric patterns in biological tissue. Materials Science and Hardware Implementation The transition from theory to clinical practice is a challenge of Materials Science. The qubits must be isolated from all external interference to prevent "decoherence." Superconducting Qubits: Currently favored by IBM and Google, these use Josephson junctions made of superconducting materials like Niobium. They require dilution refrigerators to reach temperatures colder than deep space (15 mK). Trapped Ion Qubits: Utilize individual atoms (li
Abstract The convergence of quantum computing and medical image analysis has emerged as a promising frontier for advancing diagnostic intelligence beyond the limits of classical machine learning. Recent developments in quantum hardware and quantum machine learning (QML) algorithms have stimulated growing interest in their potential to enhance complex medical imaging tasks, including disease classification, image segmentation, and clinical decision support (Wei et al., 2023; Ullah and Garcia-Zapirain, 2024). This study presents a comprehensive review and synthesis of state-of-the-art QML methodologies applied to medical imaging, systematically examining both theoretical foundations and empirical implementations. We analyze quantum-based and hybrid quantum–classical models employed across modalities such as MRI, CT, X-ray, and histopathological imaging, with particular attention to classification accuracy, computational efficiency, and diagnostic robustness (Maheshwari et al., 2022; ElBedoui et al., 2025). Our findings reveal clear trends indicating that hybrid quantum–classical approaches currently dominate practical applications, consistently outperforming purely classical baselines in select low-dimensional and domain-specific diagnostic tasks. Performance gains are most evident in feature-rich yet data-constrained scenarios, where quantum kernels and variational circuits demonstrate enhanced representational capacity (Ajlouni et al., 2023; Landman et al., 2022). However, these advantages remain context-dependent and sensitive to noise, circuit depth, and encoding strategies. Overall, QML holds significant promise for improving medical image-based diagnosis, but its widespread clinical adoption is constrained by hardware scalability, limited qubit coherence, and unresolved data encoding challenges. Addressing these limitations will be critical for translating QML from experimental feasibility to routine medical practice (Marengo and Santamato, 2025; Rawas and AlSaeed, 2026).
Deep learning techniques have significantly enhanced medical image classification for disease diagnosis. However, traditional deep learning models face several challenges, such as dependence on large annotated datasets, complex architectural layers, and millions of trainable parameters, which can increase computational load and cause overfitting. Although quantum computing offers potential for parallel processing and improved feature representation, its practical use is limited by the current capabilities of Noisy Intermediate-Scale Quantum (NISQ) hardware. To overcome these issues, Hybrid Quantum–Classical Convolutional Neural Networks (HQCNNs) have been developed, using quantum properties like superposition and entanglement to improve feature extraction and classification of medical images. This review offers a thorough analysis of 50 peer-reviewed studies published from 2021 to 2026 and introduces a taxonomy of HQCNN architectures. It classifies existing approaches into three categories: Quantum-to-Classical, Classical-to-Quantum, and Hybrid Quantum Transfer Learning. The studies reviewed encompass a diverse range of medical imaging applications, including binary and multi-class disease classification using X-ray, CT, MRI, ultrasound, dermoscopic, retinal, and mammographic images. Additionally, the review compares HQCNNs with their classical CNN counterparts, exploring how factors like quantum circuit design, dataset size, class imbalance, task complexity, evaluation protocols, and hardware limitations influence classification performance. It concludes with an analysis of the key limitations of current HQCNN models, discusses challenges in deploying NISQ devices practically, and proposes future research directions to bridge the gap between experimental HQCNN frameworks and reliable clinical diagnostic systems.
ABSTRACT Classical deep learning achieves high accuracy in image analysis tasks but require large volume of data for generalisation. Medical image datasets are often small and expensive to annotate. Arguably, quantum machine learning (QML) has the following benefits over classical machine learning (ML). (i) Quantum superposition and entanglement allow quantum machines to excel over the computational competence of classical computers. (ii) By parallel processing, QML solves problems faster. (iii) QML produces promising results with limited‐size image datasets and limited‐parameter circuits. The practical advantages of QML over classical methods are still emerging and not consistent across all application domains. Ongoing research in quantum hardware and algorithms are expected to bridge this gap. In this review, we provide an outline of quantum neural networks, quantum convolution neural networks and various hybrid models characterised by continuous learning. We also explore human brain‐inspired quantum neuromorphic computing using quantum spiking neural networks, characterised by learning from discrete neuromorphic spikes. We discuss quantum circuits, used for different medical image applications, from the perspectives of the circuit topology, the numbers of input and measurement qubits and rotation and entanglement gates. Furthermore, we conducted a systematic review of the literature on QML‐based medical image applications, datasets and benchmarks, and the analysis of the research gap separately indicating possible improvement opportunities.
Purpose This study investigates the integration of quantum-enhanced machine learning (QML) into medical image analysis, focusing on classification and feature extraction tasks. It aims to evaluate whether hybrid quantum-classical models can overcome the limitations of traditional machine learning in handling high-dimensional, clinically relevant imaging data. Design/methodology/approach Hybrid architectures combining classical preprocessing with quantum components, specifically variational quantum circuits (VQCs) and quantum kernel methods, were developed and implemented in simulated quantum environments using Qiskit and PennyLane on Google Colab. Publicly available datasets (ChestX-ray14, Brain Tumor MRI and CT Hemorrhage) were preprocessed and encoded into quantum states. Model performance was evaluated against classical baselines (MLP and CNN) using accuracy, precision, recall, F1-score, convergence speed and trainable parameters. Findings Quantum-enhanced models achieved competitive or superior performance compared to classical models with significantly fewer trainable parameters and faster convergence. The Quantum Kernel SVM demonstrated the highest precision (89.1%) and F1-score (88.5%) using only 8 parameters, while VQC models converged up to 2× faster than classical baselines. These results highlight the potential of QML to improve efficiency, robustness and scalability in diagnostic imaging workflows, particularly in resource-constrained or data-limited settings. Research limitations/implications The study relies on simulated quantum environments, which do not capture hardware noise or qubit limitations. Future work will extend to real quantum devices, multimodal data integration and interpretability strategies. Practical implications Quantum kernel methods and lightweight VQC architectures enable faster, resource-efficient diagnostic modeling, making them suitable for clinical environments with limited computational capacity or time-sensitive decision-making needs. Social implications Improving the efficiency and accessibility of medical image analysis has the potential to support faster diagnosis and enhance healthcare delivery, particularly in underserved or resource-limited regions. Originality/value This study provides a practical evaluation of hybrid quantum-classical models applied to multiple real medical imaging datasets in a reproducible simulated environment. It offers an early-stage framework demonstrating how QML techniques can be integrated into existing machine learning pipelines for medical image analysis.