BPG is committed to discovery and dissemination of knowledge
Systematic Reviews
Copyright ©The Author(s) 2025.
World J Radiol. Sep 28, 2025; 17(9): 109116
Published online Sep 28, 2025. doi: 10.4329/wjr.v17.i9.109116
Table 1 Summary of the database information
Dataset name
Subject sample
Age range
Sampling rate
Recording environment
Snoring-detection[22]1000 audio clips (500 snoring + 500 non- snoring)Not specified16 kHzVarious background environments
ICSD[23]Infant crying and snoring recordings0–2 years (infants)VariousIndoor environments
PSG-audio corpus[24]212 subjects23–85 years48 kHzSleep laboratories
SSBPR dataset[25]20 patients for body position recognition26–57 years32 kHzHospital environment
Table 2 The performance of snoring sounds detection based on different works
Ref.
Image type
Model
Main results
Hong et al[56]Log-Mel spectrogramVision Transformer-based deep learning modelSen: 89.8%, Spe: 91.3%, Acc: 95.9%
Romero et al[58]Bottleneck featuresDeep autoencoder, auditory modelF1: 94.75%
Liu et al[59]Time-domain waveform, spectrogram, Mel-spectrogramMobileNetV2 CNNAcc: 95.00%
Ye et al[57]Spectrogram, Mel-spectrogram, CWTCNN, multi-channel spectrogramAcc: 94.18%
Lim et al[44]Time-domain waveform, spectrogram, Mel-spectrogramRNNAcc: 98.9%
Jiang et al[60]Time-domain waveform, spectrum, spectrogram, Mel-spectrogram, CQT-spectrogramCNNs-DNNs, CNNs-LSTMs-DNNsAcc: 95.00%
Li et al[61]Spectrogram 1D CNN, 2D CNN (visibility graph)Acc: 89.3%, Sen: 89.7%, Spe: 88.5%
Xie et al[62]SpectrogramCNN, RNNAcc: 95.3%, Sen: 92.2%, Spe: 97.7%
González-martínez et al[63]Harmonic spectrogramCNNAUC: 0.89
Table 3 The performance of snoring sounds classification of obstructive sleep apnea-hypopnea syndrome patients
Ref.
Image type
Model
Classification
Classification results
Song et al[55]Mel-spectrogramXGBoost, CNN, ResNetOSAHS snoring vs simple snoreAcc: 83.44%, Sen: 85.27%
Ding et al[46] Mel-spectrogramVGG19 + LSTMSimple snoring vs OSAHS snoringAcc: 85.21%
Cheng et al[65]MFCC, Fbanks, LPCLSTMApnea vs normal snoring, Acc: 95.3%
Li et al[66]Spectrogram, Mel-spectrogramCNNOSAHS detectionAcc: 92.5%, Sen: 93.9%, Spc: 91.2%
Serrano et al[67]Mel-spectrogramVGGish + bi-LSTMApnea vs non-apneaAcc: 95%