Quantum optimization for computed tomography (CT) reconstruction is constrained by the number of binary variables required for image representation, making direct whole-image quantum reconstruction difficult for large or structurally complex objects. We propose a hybrid region-of-interest (ROI) refinement framework in which a coarse global image is first reconstructed by quantum tomographic reconstruction (QTR) and quantum compressed sensing tomographic reconstruction (QCSTR), filtered backprojection (FBP), or simultaneous algebraic reconstruction technique (SART), and quantum optimization is then applied only to the selected ROI through a residual projection-image formulation. This strategy reduces the effective QUBO size while preserving high-fidelity reconstruction in the target region. Experiments on three discrete phantom samples show that both QTR/QCSTR+QTR/QCSTR and SART+QTR/QCSTR achieve accurate ROI reconstruction for moderate-size cases under a reduced-angle setting. For the largest and most complex sample, the quality of the coarse global estimate becomes critical, and the best result is obtained when a stable classical coarse reconstruction is combined with second-stage ROI-only QTR/QCSTR. Among the tested pipelines, SART+QTR/QCSTR achieves the lowest average ROI RMSE and MAE. The results indicate that the practical advantage of quantum-assisted CT reconstruction lies in reserving quantum optimization for local refinement while using classical reconstruction to stabilize the global background.
Quantum optimization for computed tomography (CT) reconstruction is constrained by the number of binary variables required for image representation, making direct whole-image quantum reconstruction difficult for large or structurally complex objects. We propose a hybrid region-of-interest (ROI) refinement framework in which a coarse global image is first reconstructed by quantum tomographic reconstruction (QTR) and quantum compressed sensing tomographic reconstruction (QCSTR), filtered backprojection (FBP), or simultaneous algebraic reconstruction technique (SART), and quantum optimization is then applied only to the selected ROI through a residual projection-image formulation. This strategy reduces the effective QUBO size while preserving high-fidelity reconstruction in the target region. Experiments on three discrete phantom samples show that both QTR/QCSTR+QTR/QCSTR and SART+QTR/QCSTR achieve accurate ROI reconstruction for moderate-size cases under a reduced-angle setting. For the largest and most complex sample, the quality of the coarse global estimate becomes critical, and the best result is obtained when a stable classical coarse reconstruction is combined with second-stage ROI-only QTR/QCSTR. Among the tested pipelines, SART+QTR/QCSTR achieves the lowest average ROI RMSE and MAE. The results indicate that the practical advantage of quantum-assisted CT reconstruction lies in reserving quantum optimization for local refinement while using classical reconstruction to stabilize the global background.
We propose a quantum-assisted reconstruction framework for high-resolution tomographic imaging that reduces qubit/variable requirements by combining sinogram downsampling and region-wise iterative refinement. Previous quantum optimization–based CT reconstruction methods require solving QUBO (Quadratic Unconstrained Binary Optimization) problems over full-resolution image grids, which limits scalability under current hardware constraints. Our method addresses this by combining sinogram downscaling with region-wise iterative refinement, allowing reconstruction to begin from a reduced-resolution sinogram and image, and then to be progressively upscaled and optimized region by region. Each region is independently transformed into a compact QUBO problem and solved via a D-Wave hybrid quantum-classical solver while keeping the surrounding image fixed. The framework supports both full-view and sparse-view sinograms and is compatible with diverse acquisition geometries, including parallel-beam, fan-beam, cone-beam, synchrotron tomography, and electron tomography (ET). Adaptive region selection (including overlapping or boundary-covering subregions) enables targeted refinement of residual-error under a limited qubit/variable budget. Experimental validation on binary and integer-valued Shepp-Logan phantoms, together with an additional chest CT slice experiment, demonstrates accurate reconstructions under both dense-view and sparse-view angular sampling using reduced qubit/variable budgets. We observed that nearest-neighbor interpolation may cause edge artifacts that hinder convergence, which can be mitigated by smoother interpolation and Gaussian filtering. Sparse-view experiments indicate that the refinement pipeline can remain effective under angular undersampling in our discrete setting, suggesting potential relevance to low-dose imaging scenarios, while a dose–quality trade-off analysis is left for future work. These findings highlight the practicality of QUBO-based, quantum-assisted refinement for high-resolution tomographic reconstruction under current hardware constraints and provide a pathway for scaling as hybrid quantum optimization capabilities evolve.
Quantum-optimization-based computed tomography (CT) reconstruction has demonstrated quantum supremacy in reconstruction accuracy over conventional classical methods under sparse-view and corrupted-sinogram conditions. However, scaling this approach remains challenging because larger images require increasingly many binary variables and interactions in the underlying quadratic unconstrained binary optimization (QUBO) model. Even with iterative region-of-interest (ROI) refinement, a 60 by 60 ROI with three X-ray mass attenuation coefficients requires 10,800 binary variables under our encoding. As an initial step toward larger quantum CT images, we investigate simulated annealing (SA) followed by spin-error mitigation for optimisation (SEMO), without altering the CT-QUBO objective. We evaluate three 54-view ROI reconstruction instances and nine-view reconstruction with detector-row corruption, comparing SA and SEMO with selected hybrid quantum-classical solver results. For the ROI cases, SEMO reduced the numbers of incorrectly reconstructed pixels in low-budget SA outputs from 13 to 1, 38 to 0, and 54 to 15. With 20 percent detector-row corruption, three of four selected SA and SEMO pairs achieved exact reconstruction. Applying SEMO to one hybrid-solver output reduced its ROI error from 78 to 1 pixel. Improvements were instance-dependent: in one corrupted-sinogram case, QUBO energy decreased while image-domain error increased slightly; in one ROI case, the refined image retained 15 incorrect pixels despite having lower QUBO energy than ground truth. At the evaluated problem sizes, SA and SEMO attained reconstruction accuracy comparable to or better than selected unmitigated hybrid-solver results, providing an initial solution-quality benchmark for future larger-scale, parallelized quantum CT image reconstruction.
Computed tomography (CT) is a non-destructive technique for observing internal images and has proven highly valuable in medical diagnostics. Recent advances in quantum computing have begun to influence tomographic reconstruction techniques. The quantum tomographic reconstruction algorithm is less affected by artifacts or noise than classical algorithms by using the square function of the difference between pixels obtained by projecting CT images in quantum superposition states and pixels obtained from experimental data. In particular, by using quantum linear systems, a fast quadratic unconstrained binary optimization (QUBO) model formulation for quantum tomographic reconstruction is possible. In this paper, we formulate the QUBO model for quantum compressed sensing tomographic reconstruction, which is a linear combination of the QUBO model for quantum tomographic reconstruction and the QUBO model for total variation in quantum superposition-state CT images. In our experiments, we used sinograms obtained by using the Radon transform of Shepp-Logan images and body CT images. We evaluate the performance of the new algorithm by reconstructing CT images using a hybrid solver with the QUBO model computed from each sinogram. The new algorithm was able to obtain a solution within 5 projection images for 30 × 30 image samples and within 6 projection images for 60 × 60 image samples, reconstructing error-free CT images. We anticipate that quantum compressed sensing tomographic reconstruction algorithms could significantly reduce the total radiation dose when quantum computing performance advances.
Medical imaging systems generate large volumes of high-resolution images, making efficient storage, transmission, and real-time clinical analysis challenging. This work proposes a Region of Interest (ROI)-Aware Quantum Autoencoder (ROI-QAE) framework to achieve high compression efficiency while preserving diagnostic image quality. The ROI-Aware Quantum Autoencoder (ROI-QAE) is a four-stage pipeline for compressing computed tomography (CT), magnetic resonance imaging (MRI), and COVID-19 chest X-ray images within a single unified implementation. The first stage uses a Vision Transformer (ViT) with eight-head multi-head self-attention over 32×32-pixel patches to generate a binary region-of-interest (ROI) mask, requiring no labelled annotations. The second stage applies an ROI-aware adaptive JPEG engine. The third stage introduces a Quantum Autoencoder (QAE) built in PennyLane that operates on 8×8-pixel patches. A three-layer classical encoder maps each 64-dimensional patch to four qubit angle values, a three-layer Strongly Entangling Layers parameterized quantum circuit (PQC) on a lightning Qubit device processes these angles, and a three-layer classical decoder reconstructs 64-pixel values. The fourth stage applies lossless Huffman entropy coding. Three modality-specific QAE models are trained on 60,000 patches each (300 training images × 200 patches per image) using the Adam optimizer (initial learning rate 5×10⁻³, weight decay 1×10⁻⁵) with a cosine-annealed learning-rate schedule and gradient-norm clipping. Evaluated on 50 held-out test images per modality (drawn from pools of 120 CT, 120 MRI, and 100 X-ray test images), the framework yields mean PSNR of 23.70 dB, 24.73 dB, and 29.18 dB; SSIM of 0.7969, 0.7256, and 0.7396; and space savings of 81.6%, 81.2%, and 80.5% for CT, MRI, and X-ray respectively. Huffman compression ratios reach 1.7304 (CT), 1.3544 (MRI), and 1.1470 (X-Ray). The framework demonstrates that parameterized quantum circuits with as few as 36 quantum parameters can be embedded into a clinically aware compression pipeline and trained on commodity cloud hardware.
Quadratic unconstrained binary optimization (QUBO)-based quantum computed tomography (CT) casts reconstruction as a binary quadratic problem for quantum annealing and hybrid quantum--classical solvers. For grayscale CT, however, image encoding is constrained by the binary-variable budget: fixed global bit-plane encodings increase QUBO size and coupling complexity as gray-level precision improves, whereas low-bit encodings introduce quantization error. We propose a QUBO-based grayscale CT reconstruction framework that combines dynamic interval encoding with prior-balanced optimization. Each refinement round encodes active pixels only within local gray-level intervals around the current estimate, and a boundary-hit-guided update rule adaptively switches between search expansion and local refinement. To improve optimization stability, the method balances projection-domain data consistency and an edge-preserving quadratic prior before forming the final QUBO. Sparse-view and limited-angle fan-beam CT experiments show that the proposed method recovers structures and gray-level distributions more faithfully than the evaluated analytic, iterative, variational, and representation-based baselines. Expressivity analysis and ablation studies further indicate that the improvement mainly arises from effective gray-level representation through dynamic local encoding and more stable data-fidelity--prior coupling. Experiments on the D-Wave hybrid binary quadratic model (BQM) solver further demonstrate that the formulation is executable on a hardware-backed hybrid quantum--classical backend.
The emergence of quantum annealers has catalyzed the development of quantum optimization algorithms across various fields. The formulation of a quadratic unconstrained binary optimization (QUBO) model for quantum linear systems suggests that quantum optimization algorithms can play an important role in science and engineering. Recently, a QUBO model was developed by applying a quantum linear system to computed tomography (CT) image reconstruction to clearly show the internal structure of the sample. To date, ultra-clear CT images have not been reconstructed using quantum optimization algorithms owing to limitations in logical qubits. However, the quantum CT image reconstruction algorithm using a quantum linear system provides confidence in the accuracy of the CT image by leveraging the difference between the global and theoretical minimum energies. In this paper, we propose an algorithm that can use two or more QUBO models for CT image reconstruction. The proposed algorithm formulates several QUBO models from CT image reconstruction and verifies them using the hybrid solver of the D-Wave system. The proposed algorithm suggests that quantum-optimized CT algorithms can make greater progress in the future.