cs.CVNov 12, 2026

Quantum-inspired orthogonal adapters for certifiably robust medical image classification

Authors: Olga Tsurkan, Laia Domingo, Victoria Orlova, Asel Sagingalieva, Iris Zalaudek, Raffaele Dante Caposiena Caro

Organizations: EigenOrbit Sàrl, Luxembourg · Dermatology Clinic, Hospital Maggiore of Trieste, University of Trieste, Trieste, Italy

Abstract

Medical image classifiers require not only high accuracy but also reliable predictions under small input perturbations. Certified robustness provides mathematical guarantees of prediction stability, but preserving these guarantees when adapting pretrained models to new tasks remains challenging. We propose a fine-tuning approach based on quantum-inspired orthogonal butterfly layers that preserves the Lipschitz bounds of pretrained certified networks. Inspired by quantum circuits of reconfigurable beam splitters, these layers preserve orthogonality by construction using only O(nlog⁡n)\mathcal{O}(n\log n) trainable parameters. Their structured design enables efficient classical implementation and offers a path toward future quantum implementations with logarithmic circuit depth under suitable connectivity assumptions. Controlled experiments on MNIST and MedMNIST demonstrate improved certified robustness compared with conventional alternatives and, in the DermaMNIST transfer, greater resistance to the tested attacks than low-rank adaptation (LoRA). We then apply it to trichoscopic image classification for alopecia diagnosis and alopecia areata activity assessment. By adapting a certified ImageNet-pretrained network, butterfly fine-tuning achieves approximately twice the certified accuracy of conventional fine-tuning and LoRA, with comparable classification accuracy on the same backbone and data split and similar empirical resistance to the tested adversarial attacks. Larger robustness guarantees, including certificates covering certain image-processing transformations, can be obtained at the cost of reduced accuracy. These results highlight the potential of quantum-inspired butterfly layers for parameter-efficient adaptation of medical image classifiers while preserving the mathematical guarantees underlying certified robustness.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. LLM-Guided Initialization for Accelerated Hybrid Quantum-Classical Medical Image Classification

    Jul 29, 2026Riza Alaudin Syah, Irwan Alnarus Kautsar, Haza Nuzly Bin Abdull HamedNeural Network InitializationHybrid Quantum-Classical ML

  2. PFedQFT: Personalized Federated Quantum-Inspired Fine-Tuning for Medical Large Language Models

    Oct 19, 2026Qihan Huang, Lingxia Zhang, Guye Zhang +2Federated Fine-TuningFederated Learning

  3. Useful nonrobust features are ubiquitous in biomedical images

    Apr 24, 2026Coenraad Mouton, Randle Rabe, Niklas C. Koser +4HealthcareNeural Network Robustness