quantum-histopathology-cancer-detection

name	quantum-histopathology-cancer-detection
description	Quantum algorithms for histopathologic cancer detection on real hardware (NISQ). Covers DG-CSWAP and DG-DST circuits, NISQ mitigation pipeline, and practical QPU validation strategies.
source	arXiv:2606.21752
created	2026-06-24T00:00:00.000Z
tags	["quantum-computing","medical-imaging","cancer-detection","histopathology","nisq","noise-mitigation"]

Quantum Histopathologic Cancer Detection on Real Hardware

Overview

Methodology from paper "Configurable Algorithms for Histopathologic Cancer Detection on Quantum Hardware" (arXiv:2606.21752). First quantum hardware implementation study with noise mitigation for histopathologic image classification.

Key Innovation

Two quantum circuits for multi-directional edge response computation in histopathologic images:

DG-CSWAP (Dual-Gradient CSWAP)

• Computes multi-directional edge responses in a single execution
• Uses per-pixel local Ry encoding
• Requires 12-qubit global state preparation baseline

DG-DST (Destructive Swap Circuit)

• Hardware-efficient destructive swap circuit
• Natively matched to QPU gate sets at substantially lower circuit complexity
• Proven algebraically equivalent to DG-CSWAP
• Enables two-circuit QPU validation strategy

NISQ Mitigation Pipeline

Three-stage noise mitigation pipeline that reduces single-pixel hardware MSE by ~8x:

1. Readout Error Correction - Calibrates measurement bias across qubits
2. Bias Subtraction - Removes systematic offset from quantum measurements
3. Slope Regression - Recovers true amplitude scaling from noisy observations

Performance Results

• Inter-platform Pearson r ~0.93-0.94 across all local-simulator pairs
• Validated on five quantum processors via Amazon Braket
• 79.80% accuracy on PatchCamelyon dataset (single ResNet-50)
• Prior QFT-based baseline: 85.55% (three-model ensemble)
• Lite configuration: 17x preprocessing speedup at 2.59% accuracy cost

Practical Implementation Patterns

When to Use

• Building quantum algorithms for medical image classification
• Need hardware-efficient circuits for NISQ devices
• Implementing noise mitigation for quantum ML pipelines
• Comparing quantum vs classical approaches for histopathology

Circuit Design

1. Encode pixel data using local Ry rotations (per-pixel encoding)
2. Apply CSWAP or destructive swap for edge detection
3. Use shot-based measurements rather than full state vector simulation
4. Validate equivalence between algorithm variants on real hardware

Noise Mitigation Workflow

Raw QPU Output → Readout Correction → Bias Subtraction → Slope Regression → Cleaned Predictions

Validation Strategy

1. Run both DG-CSWAP and DG-DST on same input
2. Verify algebraic equivalence holds under noise
3. Cross-validate across multiple QPU platforms
4. Compare against classical baseline (ResNet-50)

Activation Keywords

quantum histopathology, cancer detection, DG-CSWAP, DG-DST, NISQ mitigation, quantum hardware, PatchCamelyon, medical imaging quantum, edge detection quantum

Related Papers

• 2606.22551 - QMT for hybrid QNN training stability
• 2606.21570 - Correlation Aware Quantum Feature Map for VQC