[Paper Review] Heart Disease Detection using Quantum Computing and Partitioned Random Forest Methods

Hanif Heidari, Gerhard Hellstern+1 more|arXiv (Cornell University)|Aug 17, 2022

Artificial Intelligence in Healthcare5 citations

TL;DR

This paper proposes a hybrid quantum random forest (HQRF) model using 2–4 qubits to improve early heart disease detection, integrating quantum computing with partitioned random forests to enhance accuracy and robustness. It achieves AUC scores of 96.43% (Cleveland) and 97.78% (Statlog), outperforming prior hybrid quantum neural networks (HQNN) in outlier resilience and efficiency across small and large datasets.

ABSTRACT

Heart disease morbidity and mortality rates are increasing, which has a negative impact on public health and the global economy. Early detection of heart disease reduces the incidence of heart mortality and morbidity. Recent research has utilized quantum computing methods to predict heart disease with more than 5 qubits and are computationally intensive. Despite the higher number of qubits, earlier work reports a lower accuracy in predicting heart disease, have not considered the outlier effects, and requires more computation time and memory for heart disease prediction. To overcome these limitations, we propose hybrid random forest quantum neural network (HQRF) using a few qubits (two to four) and considered the effects of outlier in the dataset. Two open-source datasets, Cleveland and Statlog, are used in this study to apply quantum networks. The proposed algorithm has been applied on two open-source datasets and utilized two different types of testing strategies such as 10-fold cross validation and 70-30 train/test ratio. We compared the performance of our proposed methodology with our earlier algorithm called hybrid quantum neural network (HQNN) proposed in the literature for heart disease prediction. HQNN and HQRF outperform in 10-fold cross validation and 70/30 train/test split ratio, respectively. The results show that HQNN requires a large training dataset while HQRF is more appropriate for both large and small training dataset. According to the experimental results, the proposed HQRF is not sensitive to the outlier data compared to HQNN. Compared to earlier works, the proposed HQRF achieved a maximum area under the curve (AUC) of 96.43% and 97.78% in predicting heart diseases using Cleveland and Statlog datasets, respectively with HQNN. The proposed HQRF is highly efficient in detecting heart disease at an early stage and will speed up clinical diagnosis.

Motivation & Objective

• To address limitations in existing quantum-based heart disease prediction models, including high qubit requirements, poor outlier handling, and high computational cost.
• To develop a more efficient and robust quantum machine learning model suitable for both small and large training datasets.
• To improve prediction accuracy and AUC by integrating quantum computing with a partitioned random forest approach.
• To reduce dependency on large training data while maintaining high performance, enabling earlier clinical diagnosis.
• To evaluate the model’s robustness against data outliers compared to prior hybrid quantum neural networks.

Proposed method

• The proposed HQRF model combines quantum circuits with a partitioned random forest ensemble to enhance feature learning and classification.
• Quantum circuits are implemented using 2–4 qubits, minimizing resource demands while maintaining high performance.
• The dataset is partitioned into subsets, each processed by a quantum-enhanced decision tree within a random forest framework.
• Outlier effects are mitigated through the ensemble nature of the random forest, which reduces sensitivity to extreme values.
• Two evaluation strategies are used: 10-fold cross-validation and 70-30 train/test split, ensuring robust performance assessment.
• The model is trained and tested on two open-source datasets: Cleveland and Statlog, representing diverse clinical data distributions.

Experimental results

Research questions

• RQ1 Can a quantum machine learning model with fewer than 5 qubits achieve higher accuracy than existing quantum-based models in heart disease detection?
• RQ2 How does the proposed HQRF model perform in the presence of data outliers compared to the HQNN model?
• RQ3 Does the HQRF model maintain high performance across both small and large training datasets?
• RQ4 What is the impact of using a partitioned random forest architecture on model generalization and computational efficiency?
• RQ5 How does the HQRF model compare to HQNN in terms of AUC and training efficiency?

Key findings

• The HQRF model achieved an AUC of 96.43% on the Cleveland dataset, outperforming previous quantum models.
• On the Statlog dataset, the HQRF achieved an AUC of 97.78%, demonstrating superior performance in cross-dataset evaluation.
• HQRF showed significantly reduced sensitivity to outlier data compared to the HQNN model, enhancing robustness.
• HQRF maintained high performance on both small and large training datasets, unlike HQNN, which required large datasets for optimal results.
• The model achieved high accuracy with only 2–4 qubits, reducing computational cost and memory usage compared to prior quantum models.
• In 10-fold cross-validation, HQRF outperformed HQNN, while HQNN performed better in 70-30 train/test splits, indicating dataset-size dependency.