Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Multi-VQC: A Novel QML Approach for Enhancing Healthcare Classification

Antonio Tudisco, Deborah Volpe, Giovanna Turvani

TL;DR

This work tackles imbalanced healthcare classification by leveraging quantum neural networks built from sequential variational quantum circuits (Multi-VQC). It evaluates the approach on three clinical datasets (Heart Failure, Diabetes, Prostate Cancer) using PCA-based dimensionality reduction and class-weighted training. Results show Quantum Multi-VQC generally outperforms classical baselines, with gains strongest on harder or more imbalanced tasks, but risk of overfitting with many VQCs and features. The work suggests a practical hybrid quantum-classical workflow and points to future exploration of encoder/topology variations and broader validation.

Abstract

Accurate and reliable diagnosis of diseases is crucial in enabling timely medical treatment and enhancing patient survival rates. In recent years, Machine Learning has revolutionized diagnostic practices by creating classification models capable of identifying diseases. However, these classification problems often suffer from significant class imbalances, which can inhibit the effectiveness of traditional models. Therefore, the interest in Quantum models has arisen, driven by the captivating promise of overcoming the limitations of the classical counterpart thanks to their ability to express complex patterns by mapping data in a higher-dimensional computational space.

Multi-VQC: A Novel QML Approach for Enhancing Healthcare Classification

TL;DR

This work tackles imbalanced healthcare classification by leveraging quantum neural networks built from sequential variational quantum circuits (Multi-VQC). It evaluates the approach on three clinical datasets (Heart Failure, Diabetes, Prostate Cancer) using PCA-based dimensionality reduction and class-weighted training. Results show Quantum Multi-VQC generally outperforms classical baselines, with gains strongest on harder or more imbalanced tasks, but risk of overfitting with many VQCs and features. The work suggests a practical hybrid quantum-classical workflow and points to future exploration of encoder/topology variations and broader validation.

Abstract

Accurate and reliable diagnosis of diseases is crucial in enabling timely medical treatment and enhancing patient survival rates. In recent years, Machine Learning has revolutionized diagnostic practices by creating classification models capable of identifying diseases. However, these classification problems often suffer from significant class imbalances, which can inhibit the effectiveness of traditional models. Therefore, the interest in Quantum models has arisen, driven by the captivating promise of overcoming the limitations of the classical counterpart thanks to their ability to express complex patterns by mapping data in a higher-dimensional computational space.

Paper Structure

This paper contains 13 sections, 3 equations, 8 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Motivations and general idea
  3. Implementation
  4. Results
  5. Settings
  6. Figures of Merit
  7. Datasets
  8. Performed Tests
  9. Heart-Failure Dataset
  10. Diabetes Dataset
  11. Prostate Cancer Dataset
  12. Discussion
  13. Conclusion

Figures (8)

  • Figure 1: Representation of a Variational Quantum Circuit.
  • Figure 2: Representation of a Multi-VQC model.
  • Figure 3: Basic Entangling Layer.
  • Figure 4: Strongly Entangling Layer.
  • Figure 5: The cumulative explained variance for each component in the three considered datasets --- Heart Failure, Diabetes, and Prostate Cancer -- , applying Principal Component Analysis.
  • ...and 3 more figures