Neonatal Pathology Classification from Infant Cry Signals Using a Multi-Branch Graph Neural Network

Keyphrases

• Acoustic Heterogeneity 25%
• Acoustics 50%
• Aggregator 25%
• Attention-based 25%
• Better Performance 25%
• Common Embedding Space 25%
• Convolutional 25%
• Cry Signal 25%
• Deep Learning 25%
• F1 Score 25%
• Gammatone Frequency Cepstral Coefficients 25%
• Graph Convolution 50%
• Graph Convolutional Network 25%
• Graph Neural Network 100%
• Graph-based Deep Learning 25%
• Heterogeneous Features 25%
• Infant Cry Signal 100%
• K-nearest Neighbor (K-NN) 25%
• Macro-average 25%
• Mel Spectrogram 25%
• Mel-frequency Cepstral Coefficients 25%
• Message Passing 25%
• Multi-branch 100%
• Neonatal Disease 25%
• Neonatal Pathology 100%
• Non-invasive Detection 25%
• Pathological Classification 100%
• Precise Identification 25%
• Precision Score 25%
• Precision-recall 25%
• Recall Score 25%
• Respiratory Distress Syndrome 25%
• Sepsis 25%
• Temporal Dependency 25%
• Test Accuracy 25%
• Well-balanced 25%

Computer Science

• Cepstral Coefficient 50%
• Deep Learning Method 50%
• Embedding Space 50%
• Experimental Result 50%
• Graph Convolution 100%
• Graph Convolutional Network 50%
• Graph Neural Network 100%
• k-Nearest Neighbors Algorithm 50%
• Learning Approach 50%
• Mel-Frequency Cepstral Coefficients 50%
• Message Passing 50%