Ying Wang

Ying Wang

Auto-generated profile, based on publicly available information

Ying Wang

Macao Polytechnic University, Macau, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 May 2025 in IEEE Access
Verified
Journal Article
Open access
FWFan Wang
Fan Wang
JGJiacheng Gao
Jiacheng Gao
YWYing Wang
Ying Wang
Guoheng HuangGuoheng Huang
Guoheng Huang
XYXiaochen Yuan
Xiaochen Yuan
Respiratory sounds serve as early indicators of lung diseases. The development of computer-aided classification systems has become a key enabler for timely diagnosis and treatment. The technology has improved basic services, particularly in resource-limited urban settings. We proposed an advanced hybrid dual-input model tailored for the intelligent classification of respiratory sounds. In this model, we employed Mel-spectrograms and waveform representations as feature extraction methods, utilizing the strengths of multiple modalities to enhance model performance. The classification framework integrates the Squeezeand- Excitation (SE) attention mechanism into the ResNet architecture to construct the Bi-SEResNet model and adopts the Data-Efficient Image Transformer (DeiT) as the final classification layer. Model performance is evaluated using the SPR-Sound dataset, which includes two classification tasks: a 2-category classification of respiratory sound events into Normal and Abnormal, and a s7-category classification involving Normal, Rhonchi, Wheeze, Stridor, Coarse Crackle, Fine Crackle, and Wheeze & Crackle. Performance was assessed using sensitivity (SE), specificity (SP), average score (AS), and harmonic score (HS) as a composite score. The proposed framework achieved scores of 89.26 and 83.63 for 2-category classification and 7-category classification tasks, respectively.

30 March 2026 in IEEE Transactions on Audio, Speech and Language Processing
Verified
Journal Article
Closed access
YWYing Wang
Ying Wang
FWFan Wang
Fan Wang
Guoheng HuangGuoheng Huang
Guoheng Huang
XZXiaobin Zheng
Xiaobin Zheng
BLBaiying Lei
Baiying Lei
XYXiaochen Yuan
Xiaochen Yuan
Despite the limited availability of pathological samples, automated lung auscultation continues to strive to establish a reliable diagnostic model. Due to this scarcity, existing automated classification methods focus primarily on feature augmentation and transfer learning, while overlooking the challenges of data quality control during the augmentation process and domain adaptation. To address these challenges, we propose a Spectral-Trustworthy Augmentation Harmonizer (STAH) framework comprising three synergistic components. First, SpecDiver (SD) generates a balanced and expanded data distribution via multi-level resampling and quantization augmentation. Based on this expanded spectral domain, TrustworthyAugFilter (TAF) then employs a dual-branch propagation mechanism with confidence-guided filtering to retain beneficial augmented samples while eliminating harmful ones that could destabilize training. Linking the SD and TAF components, HarmonicBridge (HaB) transfers representational knowledge from natural images to the spectral domain through innovative frequency decomposition, recombination, and reconstruction processes. Tests on the SPRSound 2022 dataset show that STAH gets scores of 91.02%, 79.38%, and 69.87% for tasks 1-1, 2-1, and 2-2. In the 2023 version of the dataset, our STAH reaches 82.47%, 80.46%, and 73.02% for these same tasks. We also tested our method using the ICBHI 2017 dataset. It achieves an Average Score of 73.52% for the binary task and 66.11% for the four-class task. The code is available at: https://github.com/wangying1586/STAH_Automated-Lung-Auscultation.

20 October 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
YWYing Wang
Ying Wang
YJYifeng Jiang
Yifeng Jiang
Guoheng HuangGuoheng Huang
Guoheng Huang
BLBaiying Lei
Baiying Lei
XYXiaochen Yuan
Xiaochen Yuan
Automated medical audio diagnosis supports disease detection, especially in resource-limited settings. However, scarce and imbalanced audio data limit model performance, requiring effective data augmentation. Most existing augmentations use fixed policies and ignore acoustic differences among samples, making personalized augmentation difficult.To address this, we propose a FairAug, which is one of the first fairness-aware adaptive augmentation paradigms to formulate augmentation selection as a Markov Decision Process (MDP). It connects the diagnostic model and augmentation agent in five steps: state construction, action modeling, Hybrid Batch-Sample (HBS) execution, reward computation, and network optimization. HBS first collects sample statistics via efficient batch-level exploration, then switches to personalized sample-level control after performance stabilizes.When HBS enters the sample-level phase, fairness under class imbalance requires two things. First, the difficulty scores should be more comparable across classes. Second, the grouping boundaries should adapt to each class. Therefore, we introduce Fair Confidence Measurement (FCM) and Adaptive Decision Boundary (ADB). FCM maps sample features to fixed Equiangular Tight Frame (ETF) anchors and derives difficulty scores from the separation between targets and competitors. ADB updates class-wise boundaries using these scores and operation-level feedback. This allows policy selection to follow relative sample difficulty and action effects.Experiments on five benchmarks and four backbones show gains across respiratory, cardiac, and cough sound tasks. FairAug achieves the best reported scores on all four event-level evaluations of the SPRSound dataset, with an average improvement of 5.49%. Across four backbones on the HF_Lung_V1 dataset, it further improves average recall, accuracy, and macro-F1 over standard fine-tuning by an average of 2.54%, 1.72%, and 2.36%, respectively. FairAug is a universal data augmentation tool for scarce and imbalanced medical audio data with no inference overhead, greatly enhancing the value of limited data.

4 March 2026 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
FWFan Wang
Fan Wang
YWYing Wang
Ying Wang
FPFang Peng
Fang Peng
MZMaoxi Zheng
Maoxi Zheng
XYXiaochen Yuan
Xiaochen Yuan
Classifying respiratory sounds is essential for the early diagnosis of respiratory diseases. However, the computational cost of existing computer-aided respiratory sound analysis approaches limits their practical deployment in real-time clinical environments. This study presents embedded PANNs-based intelligent auscultation with XGBoost (EPIX), an intelligent embedded auscultation instrumentation system that integrates an electronic stethoscope interface, a PANNs-based acoustic feature extraction module, an optimized eXtreme gradient boosting (XGBoost) classifier, and a two-stage decision mechanism for reliable real-time respiratory sound classification and measurement. For hardware implementation, EPIX integrates a Raspberry Pi 5 and a dedicated expansion module, enabling real-time inference for electronic stethoscopes and batch inference through an external USB drive, providing flexible processing capabilities. EPIX achieved an SC score of 0.7618 on the SPRSound dataset, while significantly reducing computational resource requirements. These capabilities make EPIX well-suited for embedded measurement scenarios with limited computational resources. The processing workflow on the device provides consistent measurement performance and enables practical use in portable and bedside intelligent auscultation instruments.

5 September 2026 in ACM Transactions on Knowledge Discovery From Data
Verified
Journal Article
Open access
YSYang Shi
Yang Shi
JWJingchao Wang
Jingchao Wang
LLLiangsi Lu
Liangsi Lu
HHHaiying Huang
Haiying Huang
SXSutong Xiao
Sutong Xiao
BXBoyan Xu
Boyan Xu
Non-Euclidean spaces inherently enable high-fidelity embeddings for hierarchical and cyclical data due to their geometric properties. Existing approaches unify hyperbolic and spherical embeddings within the framework of constant curvature spaces. However, current methods for Lipschitz regularization remain limited to non-positive curvature geometries, such as hyperbolic and Euclidean spaces, and cannot be naturally extended to the general constant curvature setting. In this article, we present a rigorous Lipschitz analysis for constant curvature graph convolutional networks ( \(\kappa\) -GCNs) and enhance their robustness through Lipschitz regularization. We derive upper bounds for the Lipschitz constants across constant curvature spaces, thereby standardizing the Lipschitz limits of the \(\kappa\) -stereographic model. Furthermore, we incorporate these bounds into a regularization framework for \(\kappa\) -GCNs to improve stability and robustness. Experimental results demonstrate that the proposed regularization method often strengthens the robustness of \(\kappa\) -GCNs across various curvature regimes, particularly under Gaussian feature noise.

18 August 2026 in Biomedical Signal Processing and Control
Verified
Journal Article
Closed access
ZLZhixin Liang
Zhixin Liang
JZJunkai Zou
Junkai Zou
MZMingfen Zheng
Mingfen Zheng
Guoheng HuangGuoheng Huang
Guoheng Huang
ZLZeng Lu
Zeng Lu
CSChentian Shen
Chentian Shen