arxiv_ml 80% Match Research Paper Quantum computing researchers,Biomedical engineers,Cardiologists,ML researchers interested in quantum applications 20 hours ago

QuPCG: Quantum Convolutional Neural Network for Detecting Abnormal Patterns in PCG Signals

speech-audio › audio-generation

📄 Abstract

Abstract: Early identification of abnormal physiological patterns is essential for the timely detection of cardiac disease. This work introduces a hybrid quantum-classical convolutional neural network (QCNN) designed to classify S3 and murmur abnormalities in heart sound signals. The approach transforms one-dimensional phonocardiogram (PCG) signals into compact two-dimensional images through a combination of wavelet feature extraction and adaptive threshold compression methods. We compress the cardiac-sound patterns into an 8-pixel image so that only 8 qubits are needed for the quantum stage. Preliminary results on the HLS-CMDS dataset demonstrate 93.33% classification accuracy on the test set and 97.14% on the train set, suggesting that quantum models can efficiently capture temporal-spectral correlations in biomedical signals. To our knowledge, this is the first application of a QCNN algorithm for bioacoustic signal processing. The proposed method represents an early step toward quantum-enhanced diagnostic systems for resource-constrained healthcare environments.

Key Contributions

This paper introduces QuPCG, a hybrid quantum-classical Convolutional Neural Network (QCNN) for detecting abnormal patterns in phonocardiogram (PCG) signals. It transforms 1D PCG signals into compact 2D images, requiring only 8 qubits for the quantum stage, and demonstrates high classification accuracy for cardiac abnormalities.

Business Value

Potential for developing more accurate and efficient diagnostic tools for cardiac diseases, improving patient outcomes and reducing healthcare costs.

Paper Metadata

Innovation Type

Hybrid Model Architecture

Deployment Feasibility

Low to Moderate. Relies on the availability and maturity of quantum computing hardware and software, but the hybrid approach might offer near-term feasibility.

Limitations Addressed

Addresses the need for efficient and accurate early detection of cardiac abnormalities from PCG signals. It explores the potential of quantum computing for bioacoustic signal processing, a novel application area.

Performance Gains

Achieves high classification accuracy for detecting cardiac abnormalities using a novel QCNN approach.

Technical Tags

Quantum Convolutional Neural Network (QCNN)phonocardiogram (PCG) signalscardiac disease detectionabnormal patternswavelet feature extractionadaptive threshold compressionbioacoustic signal processingqubitsHLS-CMDS dataset

Research Topics

Quantum Machine LearningBiomedical Signal ProcessingCardiologyPattern RecognitionHybrid Quantum-Classical Models

Methods & Architectures

Hybrid Quantum-Classical CNN (QCNN)Wavelet feature extractionAdaptive threshold compressionQuantum circuit for classification Quantum Convolutional Neural Network (QCNN)

Applications & Tasks

Medical Diagnostics Biomedical Engineering Quantum Computing Signal ClassificationAnomaly DetectionBiomedical Signal Analysis Detecting S3 and murmur abnormalities in heart soundsCardiac disease diagnosis

Datasets & Benchmarks

Datasets

HLS-CMDS dataset

Benchmarks

93.33% classification accuracy on the test set • 97.14% on the train set

Classification accuracy

Related Fields

Quantum ComputingMachine LearningBiomedical EngineeringSignal ProcessingCardiology

Keywords

Quantum CNNQCNNphonocardiogramPCG signalscardiac diseaseabnormal patternswavelet transformadaptive thresholdingbioacousticsquantum machine learningHLS-CMDS

Academic Context

#Quantum Machine Learning#Biomedical Signal Processing#Cardiology#Pattern Recognition#Hybrid Quantum-Classical Models

Commercial Potential

Potential Products

Quantum-enhanced medical diagnostic devicesAI tools for cardiac auscultation analysis

Target Industries

HealthcareMedical DevicesQuantum Technology

Use Case Examples

Automated screening for heart murmursEarly detection of specific cardiac conditions using heart sounds

Competitive Edge

Presents a novel application of QCNNs to bioacoustic signal processing for cardiac diagnostics, potentially offering advantages in capturing complex temporal-spectral correlations.

Market Opportunity

Significant, driven by the global cardiac care market.

Revenue Models

Licensing technology to medical device manufacturers.

Resource Requirements

Compute Needs

Requires access to quantum computing resources or simulators.

Data Requirements

Phonocardiogram (PCG) signal datasets (e.g., HLS-CMDS).

Deployment Constraints

Maturity and accessibility of quantum hardware.,Integration into clinical workflows.

Scalability

Scalability depends on quantum hardware capabilities and algorithm efficiency.

Regulatory Considerations

Medical device regulations (FDA, CE marking)

Production Readiness

Maturity Level

Early Research / Proof of Concept

Time to Market

5-10 years, contingent on quantum hardware advancements.

View Full Paper Back to Papers