Ganesan, Swathi ORCID: https://orcid.org/0000-0002-6278-2090, Karunarathne, Lakmali ORCID: https://orcid.org/0009-0000-7720-7817, Karunarathne, Kavindu, Weerakoddi, Mishani and Somasiri, Nalinda ORCID: https://orcid.org/0000-0001-6311-2251 (2026) Quantum Artificial Intelligence for Medical Data: Comparative Evaluation of Quantum Machine Learning Models. Cureus Journal Of Computer Science.

Text 11082-quantum-artificial-intelligence-for-medical-data-comparative-evaluation-of-quantum-machine-learning-models - Accepted Version Available under License Creative Commons Attribution. Download (239kB)

Abstract

Background
Artificial Intelligence (AI) incorporation within healthcare applications has significantly helped in the improvement of healthcare diagnostics, patient forecasting, and clinical decision-making. The steady growing of medical data in terms of quantity, quality, and dimension poses more significant problems for traditional AI methods in terms of size, time, and computational efficiency. However, classical machine learning models often struggle with high-dimensional, imbalanced, and heterogeneous medical datasets, which limits their scalability and generalizability. The Quantum Artificial Intelligence (QAI) model, which combines quantum computing and machine learning in a complementary way, has emerged as one of the main trends to defeat these problems. This complementary approach is gaining more popularity, with the most advanced work being developed in this area, as it offers a fundamental explanation of current complexities and the underlying principles and their applications.

Methodology
This paper begins with the basic concepts of quantum computing and where it overlaps with AI. It then moves to a comparative approach, discussing various quantum machine learning algorithms such as quantum support vector machines, quantum neural networks, and variational quantum circuits. A comparative evaluation of quantum support vector machine, variational quantum classifier, quantum neural network, quantum K-means, and quantum Boltzmann machine is conducted using the Pima Indians Diabetes dataset to assess model behaviour and performance under standard metrics. Further detail is provided on the theoretical and practical implications of QAI in medical imaging, genomics, electronic health records, and drug discovery. Additionally, this paper examines platforms, tools, and quantum hardware used for QAI research, and discusses unique challenges currently faced in quantum technology.

Results
Unlike prior surveys that primarily provide conceptual overviews, this paper uniquely combines a comprehensive review with practical benchmarking of multiple quantum machine learning models on a benchmark medical dataset. This provides comparative insights into algorithmic performance under real-world constraints, particularly in relation to high-dimensional medical data, genomic data, electronic health records with large number of clinical attributes, radiology images, etc. In this study, dimensionality arises from multivariate clinical features encoded into quantum feature maps and the limitations of current quantum systems.

Conclusion
This paper aims to serve as a starting point for researchers and practitioners interested in the potential of quantum-enhanced AI to transform health data analytics. By identifying current limitations, evaluating the performance of emerging quantum machine learning approaches, and outlining gaps in the literature, the paper highlights future research directions necessary to advance QAI in healthcare. Overall, the findings outline both the potential and practical constraints of quantum models, offering a foundation for future work toward scalable, clinically applicable QAI solutions.

Item Type:	Article
Status:	Published
DOI:	10.7759/s44389-025-00063-x
Subjects:	T Technology > T Technology (General)
School/Department:	London Campus
URI:	https://ray.yorksj.ac.uk/id/eprint/14741

University Staff: Request a correction | RaY Editors: Update this record