Quantum-Enhanced Processing with Tensor-Network Frontends for Privacy-Aware Federated Medical Diagnosis

Abstract

We propose a privacy-aware hybrid framework for federated medical image classification that combines tensor-network representation learning, MPC-secured aggregation, and post-aggregation quantum refinement. The framework is motivated by two practical constraints in privacy-aware federated learning: MPC can introduce substantial communication overhead, and direct quantum processing of high-dimensional medical images is unrealistic with a small number of qubits. To address both constraints within a single architecture, client-side tensor-network frontends, Matrix Product State (MPS), Tree Tensor Network (TTN), and Multi-scale Entanglement Renormalization Ansatz (MERA), compress local inputs into compact latent representations, after which a Quantum-Enhanced Processor (QEP) refines the aggregated latent feature through quantum-state embedding and observable-based readout. Experiments on PneumoniaMNIST show that the effect of the QEP is frontend-dependent rather than uniform across architectures. In the present setting, the TTN+QEP combination exhibits the most balanced overall profile. The results also suggest that the QEP behaves more stably when the qubit count is sufficiently matched to the latent dimension, while noisy conditions degrade performance relative to the noiseless setting. The MPC benchmark further shows that communication cost is governed primarily by the dimension of the protected latent representation. This indicates that tensor-network compression plays a dual role: it enables small-qubit quantum processing on compressed latent features and reduces the communication overhead associated with secure aggregation. Taken together, these results support a co-design perspective in which representation compression, post-aggregation quantum refinement, and privacy-aware deployment should be optimized jointly.

Figures (7)

• Figure 1: Proposed TN+MPC+QEP pipeline. Client-side tensor-network encoders produce latent features, MPC-secured aggregation forms a protected global representation, and the QEP performs post-aggregation refinement before classification.
• Figure 2: Parameterized quantum circuit used in the QEP. Each layer applies single-qubit $R_y$ and $R_z$ rotations followed by nearest-neighbor CNOT entangling gates. For visualization, terminal measurement gates are shown explicitly. In the actual implementation, however, the circuit output is summarized through expectation values of selected Pauli observables rather than bitstring sampling.
• Figure 3: Example samples from PneumoniaMNIST. The top row shows Normal cases and the bottom row shows Pneumonia cases.
• Figure 4: Threshold-optimized test performance of the Classical and Quantum modes across MPS, TTN, and MERA. Boxplots report class-wise Precision, Recall, F1-score, and overall Accuracy for the default QEP setting ($N_q=16$).
• Figure 5: Internal behavior of the QEP across MPS, TTN, and MERA: (A) evolution of $\alpha$, (B) evolution of $q$-std, (C) final-epoch $q$-std distribution, and (D) $\alpha$-$q$-std phase behavior.