Hybrid Quantum Machine Learning Assisted Classification of COVID-19 from Computed Tomography Scans

Aug 3, 2026

Upload by Ivy

Practical quantum computing remains in an early stage, and quantum machine learning must operate under strong limits in available feature capacity compared with classical methods. This work presents a hybrid quantum machine learning pipeline for medical image processing using real CT data. Large lung CT scans are classified into COVID-19, CAP, or Normal. Quantum image embedding and hybrid quantum transfer learning are evaluated across multiple circuit and embedding choices.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Related Work

Background

Quantum Transfer Learning Approaches and Circuits

Experimental Setup and Results

Discussion

Conclusion

Hybrid Quantum Machine Learning Assisted Classification of COVID-19 from Computed Tomography Scans arXiv :2310 .02748v1 [ quant-ph] 4 Oct 2023 Leo S¨unkel Darya Martyniuk Julia J. Reichwald Andrei Morariu LMU Munich Fraunhofer FOKUS Smart Reporting GmbH Smart Reporting GmbH leo.suenkel@ifi.lmu.de darya.martyniuk@fokus.fraunhofer.de j.reichwald@smart-reporting.coma.morariu@smart-reporting.com Raja Havish Seggoju Fraunhofer FOKUS [raja.havish.seggoju@fokus.fraunhofer.de](raja.havish.seggoju@fokus.fraunhofer.de) Philipp Altmann LMU Munich [philipp.altmann@ifi.lmu.de](philipp.altmann@ifi.lmu.de) Christoph Roch LMU Munich [christoph.roch@ifi.lmu.de](christoph.roch@ifi.lmu.de) Adrian Paschke Freie Universita¨t Berlin Fraunhofer FOKUS [adrian.paschke@fokus.fraunhofer.de](adrian.paschke@fokus.fraunhofer.de) Abstract—Practical quantum computing (QC) is still in its infancy and problems considered are usually fairly small, especially in quantum machine learning when compared to its classical counterpart. Image processing applications in particular require models that are able to handle a large amount of features, and while classical approaches can easily tackle this, it is a major challenge and a cause for harsh restrictions in contemporary QC. In this paper, we apply a hybrid quantum machine learning approach to a practically relevant problem with real world-data. That is, we apply hybrid quantum transfer learning to an image processing task in the field of medical image processing. More specifically, we classify large CT-scans of the lung into COVID- 19, CAP, or Normal. We discuss quantum image embedding as well as hybrid quantum machine learning and evaluate several approaches to quantum transfer learning with various quantum circuits and embedding techniques. Index Terms—Quantum machine learning, Quantum transfer learning, Quantum medical applications, Quantum image classification I. INTRODUCTION Quantum computing (QC) is an ever-growing field with applications spanning a wide range of domains including finance [1], [2], chemistry [3], [4], simulation [5], machine learning [6], [7] and optimization [8] . While attention for this field is rising rapidly, practical QC is itself in its infancy and the capabilities of current and near-term quantum computers are limited. Thus, the current phase of QC is famously referred to as the noisy intermediate scale quantum era (NISQ-era) [9] . Yet, QC has the potential to solve some computational © 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. This work was partially funded by the BMWK project PlanQK (01MK20005F / 01MK20005I / 01MK20005P) problems faster than classical computers, for example providing an exponential [10] or quadratical [11] speedup. Quantum machine learning (QML) currently is a popular domain amongst researchers in which many hope to find a quantum advantage or speedup. In this paper, we apply a QML assisted approach to a real-world problem from the medical domain, the classification of COVID-19 from CT scans of the lung. However, as these CT scans are images with a large number of features and the capacity of current QC-hardware is fairly limited, we employ a hybrid approach, i.e., an approach that consists of a classical as well as a quantum part, instead of a purely quantum approach. The paper is structured as follows. In Section II we briefly discuss related work. We introduce the necessary background in Section III, i.e., we discuss the topic of classification of medical images in the context of the detection of COVID-19 and give an overview of various quantum embedding methods including image specifi

FAQs

What problem does the document address?

It addresses image-based classification of COVID-19 from lung CT scans under limited current quantum hardware capabilities.

How does the proposed method combine classical and quantum components?

It uses a hybrid approach that includes classical processing together with quantum transfer learning, rather than relying solely on a fully quantum model.

Which classification targets are used for the CT scan task?

The model classifies lung CT scans into COVID-19, CAP, or Normal categories.

Aug 3, 2026

Upload by Ivy