[Paper Review] Evaluating the Impact of Different Quantum Kernels on the Classification Performance of Support Vector Machine Algorithm: A Medical Dataset Application

Emine Akpinar, Sardar M. N. Islam+1 more|arXiv (Cornell University)|Jul 13, 2024

Artificial Intelligence in Healthcare5 citations

TL;DR

The paper assesses how different quantum feature maps affect QSVM-Kernel classification accuracy and runtime on two medical datasets, identifying feature maps that yield best performance.

ABSTRACT

The support vector machine algorithm with a quantum kernel estimator (QSVM-Kernel), as a leading example of a quantum machine learning technique, has undergone significant advancements. Nevertheless, its integration with classical data presents unique challenges. While quantum computers primarily interact with data in quantum states, embedding classical data into quantum states using feature mapping techniques is essential for leveraging quantum algorithms Despite the recognized importance of feature mapping, its specific impact on data classification outcomes remains largely unexplored. This study addresses this gap by comprehensively assessing the effects of various feature mapping methods on classification results, taking medical data analysis as a case study. In this study, the QSVM-Kernel method was applied to classification problems in two different and publicly available medical datasets, namely, the Wisconsin Breast Cancer (original) and The Cancer Genome Atlas (TCGA) Glioma datasets. In the QSVM-Kernel algorithm, quantum kernel matrices obtained from 9 different quantum feature maps were used. Thus, the effects of these quantum feature maps on the classification results of the QSVM-Kernel algorithm were examined in terms of both classifier performance and total execution time. As a result, in the Wisconsin Breast Cancer (original) and TCGA Glioma datasets, when Rx and Ry rotational gates were used, respectively, as feature maps in the QSVM-Kernel algorithm, the best classification performances were achieved both in terms of classification performance and total execution time. The contributions of this study are that (1) it highlights the significant impact of feature mapping techniques on medical data classification outcomes using the QSVM-Kernel algorithm, and (2) it also guides undertaking research for improved QSVM classification performance.

Motivation & Objective

• Motivate the study by addressing how feature mapping impacts quantum kernel methods when pairing with classical data.
• Evaluate QSVM-Kernel performance using nine different quantum feature maps on medical datasets.
• Compare classification accuracy and total execution time across feature maps.
• Provide guidance on selecting feature mappings to improve QSVM classification outcomes in medical data.

Proposed method

• Apply QSVM-Kernel with quantum kernel matrices derived from nine different quantum feature maps.
• Use two publicly available medical datasets: Wisconsin Breast Cancer (original) and TCGA Glioma.
• Evaluate classification performance and total execution time for each feature map.
• Identify which feature maps (e.g., Rx and Ry rotational gates) yield best results for each dataset.
• Analyze the relationship between feature map choice and computational efficiency.

Experimental results

Research questions

• RQ1 How do different quantum feature maps affect QSVM-Kernel classification performance on medical data?
• RQ2 Which feature maps optimize the trade-off between accuracy and execution time in QSVM-Kernel on the examined datasets?
• RQ3 Do specific gates (e.g., Rx, Ry) consistently yield superior results across datasets?

Key findings

• Rx and Ry rotational gates achieved the best classification performance and execution time on the Wisconsin Breast Cancer dataset and the TCGA Glioma dataset.
• The study demonstrates a significant impact of feature mapping techniques on medical data classification outcomes using QSVM-Kernel.
• Guidance is provided for designing QSVM classifications with improved performance through appropriate feature maps.