Towards a Utility-Scale Quantum Edge Detection for Real-World Medical Image Data

TL;DR

This paper introduces a two-level decomposition strategy to improve quantum edge detection on noisy quantum devices, enabling practical medical image analysis with high fidelity and reduced circuit complexity.

Contribution

It proposes a novel P×Q decomposition method with optimizations that significantly reduce circuit depth and operations, facilitating high-fidelity quantum edge detection on NISQ hardware.

Findings

Achieved over 62% reduction in circuit depth.

Approximately 93% fewer two-qubit operations.

Maintained fidelity exceeding 95.6% under realistic noise models.

Abstract

We present a two-level decomposition strategy to enhance the quality and performance of Quantum Hadamard Edge Detection (QHED) for practical image analysis on Noisy Intermediate-Scale Quantum (NISQ) devices. A Data-Level Decomposition partitions an input image into P augmented sub-images, each encoded into a separate quantum circuit. Each of these circuits is then further cut via Circuit-Level Decomposition into Q smaller sub-circuits suitable for execution on near-term quantum devices. The two-level P $\times$ Q decomposition, along with optimizations we introduced, achieves over 62\% reductions in circuit depth and approximately 93\% fewer two-qubit operations, while maintaining a fidelity exceeding 95.6\% under realistic IBM noise models for 5-qubit data input sizes. These results demonstrate the feasibility of performing high-fidelity QHED on NISQ hardware and provide lessons and early evidence of distributed utility scale quantum computing, further illustrated by processing raw k-space MRI data with an Inverse Quantum Fourier Transform and a distributed simulation of the modified QHED on large 2D and 3D MRI datasets.