HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz
Cardiac pathology classification (CPC) based on the volumetric features of three key heart structures can be extracted from segmented cardiac cine magnetic resonance imaging (CMRI) sequences. Machine learning models have recently become very effective tools for handling these problems. Hybrid quantu...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10416865/ |
_version_ | 1797321066228482048 |
---|---|
author | Doaa A. Shoieb Ahmed Younes Sherin M. Youssef Karma M. Fathalla |
author_facet | Doaa A. Shoieb Ahmed Younes Sherin M. Youssef Karma M. Fathalla |
author_sort | Doaa A. Shoieb |
collection | DOAJ |
description | Cardiac pathology classification (CPC) based on the volumetric features of three key heart structures can be extracted from segmented cardiac cine magnetic resonance imaging (CMRI) sequences. Machine learning models have recently become very effective tools for handling these problems. Hybrid quantum methods can be employed to enhance the capacities of classical machine learning models. Here, we propose a hybrid quantum multiclass cardiac pathologies classification (HQMC-CPC) model. In the proposed model, discriminative features are extracted to characterize the shape and function of the heart, combining the clinical features, patient features, and radiomics features. In the proposed quantum circuits, all variational parameters are trainable, and the enhanced variational quantum circuit is employed for efficient neural network learning. Using only thirty feature values as input, we propose a hybrid quantum multiclass cardiac pathologies classification (HQMC-CPC) model based on the proposed modified hardware efficient ansatz (MHEA). The proposed model achieves promising results in training and testing with the Automatic Cardiac Diagnosis Challenge (ACDC 2017) dataset. Experimental results showed that the proposed HQMC-CPC model is able to classify different cardiac pathologies with an average minimum performance gap of 3.19%. The average maximum improvement in terms of accuracy in cardiac pathology classification is 7.77%. Moreover, the proposed HQMC-CPC speeds up the testing process by around 60% and 40% compared to the classical classifiers and well-established HEA respectively. |
first_indexed | 2024-03-08T04:53:05Z |
format | Article |
id | doaj.art-f864b1b4ce9543fbbee876c88f245b4f |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-08T04:53:05Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-f864b1b4ce9543fbbee876c88f245b4f2024-02-08T00:02:03ZengIEEEIEEE Access2169-35362024-01-0112182951831410.1109/ACCESS.2024.336013910416865HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient AnsatzDoaa A. Shoieb0https://orcid.org/0000-0001-8054-0119Ahmed Younes1https://orcid.org/0000-0002-1594-1589Sherin M. Youssef2Karma M. Fathalla3https://orcid.org/0000-0003-0089-3841Computer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, EgyptDepartment of Mathematics and Computer Science, Faculty of Science, Alexandria University, Alexandria, EgyptComputer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, EgyptComputer Engineering Department, Arab Academy for Science, Technology and Maritime Transport (AASTMT), Alexandria, EgyptCardiac pathology classification (CPC) based on the volumetric features of three key heart structures can be extracted from segmented cardiac cine magnetic resonance imaging (CMRI) sequences. Machine learning models have recently become very effective tools for handling these problems. Hybrid quantum methods can be employed to enhance the capacities of classical machine learning models. Here, we propose a hybrid quantum multiclass cardiac pathologies classification (HQMC-CPC) model. In the proposed model, discriminative features are extracted to characterize the shape and function of the heart, combining the clinical features, patient features, and radiomics features. In the proposed quantum circuits, all variational parameters are trainable, and the enhanced variational quantum circuit is employed for efficient neural network learning. Using only thirty feature values as input, we propose a hybrid quantum multiclass cardiac pathologies classification (HQMC-CPC) model based on the proposed modified hardware efficient ansatz (MHEA). The proposed model achieves promising results in training and testing with the Automatic Cardiac Diagnosis Challenge (ACDC 2017) dataset. Experimental results showed that the proposed HQMC-CPC model is able to classify different cardiac pathologies with an average minimum performance gap of 3.19%. The average maximum improvement in terms of accuracy in cardiac pathology classification is 7.77%. Moreover, the proposed HQMC-CPC speeds up the testing process by around 60% and 40% compared to the classical classifiers and well-established HEA respectively.https://ieeexplore.ieee.org/document/10416865/Ansatzcardiac pathologiesclassificationclinical featuresquantum circuitradiomic features |
spellingShingle | Doaa A. Shoieb Ahmed Younes Sherin M. Youssef Karma M. Fathalla HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz IEEE Access Ansatz cardiac pathologies classification clinical features quantum circuit radiomic features |
title | HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz |
title_full | HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz |
title_fullStr | HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz |
title_full_unstemmed | HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz |
title_short | HQMC-CPC: A Hybrid Quantum Multiclass Cardiac Pathologies Classification Integrating a Modified Hardware Efficient Ansatz |
title_sort | hqmc cpc a hybrid quantum multiclass cardiac pathologies classification integrating a modified hardware efficient ansatz |
topic | Ansatz cardiac pathologies classification clinical features quantum circuit radiomic features |
url | https://ieeexplore.ieee.org/document/10416865/ |
work_keys_str_mv | AT doaaashoieb hqmccpcahybridquantummulticlasscardiacpathologiesclassificationintegratingamodifiedhardwareefficientansatz AT ahmedyounes hqmccpcahybridquantummulticlasscardiacpathologiesclassificationintegratingamodifiedhardwareefficientansatz AT sherinmyoussef hqmccpcahybridquantummulticlasscardiacpathologiesclassificationintegratingamodifiedhardwareefficientansatz AT karmamfathalla hqmccpcahybridquantummulticlasscardiacpathologiesclassificationintegratingamodifiedhardwareefficientansatz |