Workshop Description
Quantum machine learning for medical imaging is not yet clinically deployable at scale. A systematic review published in April 2025 (arXiv 2504.13910) found that hybrid quantum-classical models outperform purely classical baselines only in low-dimensional and domain-specific diagnostic tasks: carefully selected benchmark datasets, not the high-volume, high-complexity imaging workloads that NHS radiology departments actually process. Only 16 of 72 published QML healthcare studies considered realistic operating conditions with actual quantum hardware or noisy simulations. The gap between research publication and clinical use is wider for QML than for any other AI modality because hardware noise, limited qubit coherence, and unresolved data encoding challenges compound the normal clinical validation burden.
That is not an argument against engaging with QML. It is an argument for clinical informatics teams developing the technical literacy to distinguish genuine early-stage capability from premature commercial positioning. The vendor market for QML in healthcare is active, with companies claiming quantum advantage in MRI feature extraction, CT scan classification, and histopathology analysis. Some of those claims will be sound; most will not be reproducible at clinical scale. This workshop builds the evaluation framework: how to read a QML benchmark paper, what validation evidence is required before a procurement decision, how to apply NICE evidence standards to quantum AI submissions, and how to interpret hardware roadmap claims from quantum computing vendors in terms of specific clinical use case timelines.
What participants cover
- Current state of QML in clinical imaging: what 72 peer-reviewed studies show versus what vendors claim
- Hybrid quantum-classical architectures: where classical pre-processing ends and quantum circuits begin
- Hardware noise and coherence time: the technical constraints that limit current clinical QML viability
- Clinical validation frameworks: applying NICE evidence standards and MHRA/FDA SaMD pathways to quantum AI submissions
- The 8 questions every clinical technology assessment team should ask a QML vendor before procurement
- Hardware roadmaps for medical AI: evidence-based timeline for when specific imaging use cases may become viable