Cardiac and respiratory diseases are the primary causes of health problems. If we can automate anomalous heart and lung sound diagnosis, we can improve the early detection of disease and enable the screening of a wider population than possible with manual screening. We propose a lightweight yet powerful model for simultaneous lung and heart sound diagnosis, which is deployable in an embedded low-cost device and is valuable in remote areas or developing countries where Internet access may not be available. We trained and tested the proposed model with the ICBHI and the Yaseen datasets. The experimental results showed that our 11-class prediction model could achieve 99.94% accuracy, 99.84% precision, 99.89% specificity, 99.66% sensitivity, and 99.72% F1 score. We designed a digital stethoscope (around USD 5) and connected it to a low-cost, single-board-computer Raspberry Pi Zero 2W (around USD 20), on which our pretrained model can be smoothly run. This AI-empowered digital stethoscope is beneficial for anyone in the medical field, as it can automatically provide diagnostic results and produce digital audio records for further analysis.
This paper addresses the critical challenge of point cloud down-sampling for digital twin creation, where reducing data volume while preserving geometric fidelity remains an ongoing research problem. We propose a novel down-sampling approach that combines Low-Discrepancy Sequences (LDS) with Hilbert curve ordering to create a method that preserves both global distribution characteristics and local geometric features. Unlike traditional methods that impose uniform density or rely on computationally intensive feature detection, our LDS-Hilbert approach leverages the complementary mathematical properties of Low-Discrepancy Sequences and space-filling curves to achieve balanced sampling that respects the original density distribution while ensuring comprehensive coverage. Through four comprehensive experiments covering parametric surface fitting, mesh reconstruction from basic closed geometries, complex CAD models, and real-world laser scans, we demonstrate that LDS-Hilbert consistently outperforms established methods, including Simple Random Sampling (SRS), Farthest Point Sampling (FPS), and Voxel Grid Filtering (Voxel). Results show parameter recovery improvements often exceeding 50% for parametric models compared to the FPS and Voxel methods, nearly 50% better shape preservation as measured by the Point-to-Mesh Distance (than FPS) and up to 160% as measured by the Viewpoint Feature Histogram Distance (than SRS) on complex real-world scans. The method achieves these improvements without requiring feature-specific calculations, extensive pre-processing, or task-specific training data, making it a practical advance for enhancing digital twin fidelity across diverse application domains.
Feature map pooling in convolutional neural networks (CNNs) serves the dual purpose of reducing spatial dimensions and enhancing feature invariance. Current pooling approaches face a fundamental trade-off: deterministic methods (e.g., MaxPool and AvgPool) lack regularization benefits, while stochastic approaches introduce beneficial randomness but can suffer from sampling biases and may require careful hyperparameter tuning (e.g., S3Pool). To address these limitations, this paper introduces LDS3Pool, a novel pooling method that leverages low-discrepancy sequences (LDSs) for quasi-random spatial sampling. LDS3Pool first linearizes 2D feature maps to 1D sequences using Hilbert space-filling curves to preserve spatial locality, then applies LDS-based sampling to achieve quasi-random downsampling with mathematical guarantees of uniform coverage. This framework provides the regularization benefits of randomness while maintaining comprehensive feature representation, without requiring sensitive hyperparameter tuning. Extensive experiments demonstrate that LDS3Pool consistently outperforms baseline methods across multiple datasets and a range of architectures, from classic models like VGG11 to modern networks like ResNet18, achieving significant accuracy gains with moderate computational overhead. The method’s empirical success is supported by a rigorous theoretical analysis, including a quantitative evaluation of the Hilbert curve’s superior, size-independent locality preservation. In summary, LDS3Pool represents a theoretically sound and empirically effective pooling method that enhances CNN generalization through a principled, quasi-random sampling framework.