arXiv subjects

Enze Zhou

Publications and source records attributed to Enze Zhou.

3 recordsLinked to original sources

arXiv2025

EZhouNet:A framework based on graph neural network and anchor interval for the respiratory sound event detection

Auscultation is a key method for early diagnosis of respiratory and pulmonary diseases, relying on skilled healthcare professionals. However, the process is often subjective, with variability between experts. As a result, numerous deep learning-based automatic classification methods have emerged, most of which focus on respiratory sound classification. In contrast, research on respiratory sound event detection remains limited. Existing sound event detection methods typically rely on frame-level predictions followed by post-processing to generate event-level outputs, making interval boundaries challenging to learn directly. Furthermore, many approaches can only handle fixed-length audio, limiting their applicability to variable-length respiratory sounds. Additionally, the impact of respiratory sound location information on detection performance has not been extensively explored. To address these issues, we propose a graph neural network-based framework with anchor intervals, capable of handling variable-length audio and providing more precise temporal localization for abnormal respiratory sound events. Our method improves both the flexibility and applicability of respiratory sound detection. Experiments on the SPRSound 2024 and HF Lung V1 datasets demonstrate the effectiveness of the proposed approach, and incorporating respiratory position information enhances the discrimination between abnormal sounds. The reference implementation is available at https://github.com/chumingqian/EzhouNet.

cs.SD↗

arXiv2025

STEPC: A Multi-energy Nonuniform Response Calibration Framework for Photon-Counting Micro-CT in Multi-material Imaging

Photon-counting computed tomography has demonstrated significant advancements in recent years; however, micro photon-counting CT (Micro-PCCT) systems are still limited by pixel-wise detector response nonuniformity, which degrades measurement uniformity across detector pixels and commonly produces ring artifacts in reconstructed images. Existing calibration methods exhibit limited generalizability in complex multi-material scenarios, such as contrast-enhanced imaging. This study introduces a Signal-to-Nonuniformity Error Polynomial Calibration (STEPC) framework based on measurement nonuniformity error modeling to address this issue. STEPC first fits multi-energy projections using a 2D polynomial surface to generate ideal references, then applies a nonlinear multi-energy polynomial model to predict and correct pixel-wise nonuniformity errors. The model is calibrated using homogeneous slab phantoms of different materials, including PMMA, aluminum, and iodinated contrast agents, enabling correction for both non-contrast and contrast-enhanced imaging. Experiments were performed on a custom Micro-PCCT system with phantoms and mouse. Correction performance of STEPC was evaluated using the mean local standard deviation (MLSD) in the projection domain and the ring artifact deviation (RAD) on the reconstructed images. Compared with existing methods, STEPC achieved an average MLSD reduction of at least 21.58% and reduced RAD by at least 14.18%, consistently yielding the best performance in both non-contrast and contrast-enhanced scenarios. Furthermore, STEPC can be readily extended to compensate for beam hardening effects within the same calibration framework. Quantitative material decomposition results indicate that the proposed method preserves measurement accuracy across different basis materials...

physics.med-ph↗

arXiv2025

CycleGuardian: A Framework for Automatic RespiratorySound classification Based on Improved Deep clustering and Contrastive Learning

Auscultation plays a pivotal role in early respiratory and pulmonary disease diagnosis. Despite the emergence of deep learning-based methods for automatic respiratory sound classification post-Covid-19, limited datasets impede performance enhancement. Distinguishing between normal and abnormal respiratory sounds poses challenges due to the coexistence of normal respiratory components and noise components in both types. Moreover, different abnormal respiratory sounds exhibit similar anomalous features, hindering their differentiation. Besides, existing state-of-the-art models suffer from excessive parameter size, impeding deployment on resource-constrained mobile platforms. To address these issues, we design a lightweight network CycleGuardian and propose a framework based on an improved deep clustering and contrastive learning. We first generate a hybrid spectrogram for feature diversity and grouping spectrograms to facilitating intermittent abnormal sound capture.Then, CycleGuardian integrates a deep clustering module with a similarity-constrained clustering component to improve the ability to capture abnormal features and a contrastive learning module with group mixing for enhanced abnormal feature discernment. Multi-objective optimization enhances overall performance during training. In experiments we use the ICBHI2017 dataset, following the official split method and without any pre-trained weights, our method achieves Sp: 82.06 $\%$, Se: 44.47$\%$, and Score: 63.26$\%$ with a network model size of 38M, comparing to the current model, our method leads by nearly 7$\%$, achieving the current best performances. Additionally, we deploy the network on Android devices, showcasing a comprehensive intelligent respiratory sound auscultation system.

cs.SD↗