Unveiling Hidden Patterns in Infant Cry Audio: A Multi-Feature Vision Transformer Approach With Explainable AI

Keyphrases

• Acoustic Features 66%
• Audio 100%
• Classification Accuracy 33%
• Cross-validation Test 33%
• Cry Analysis 33%
• Deep Learning Framework 66%
• Early Detection 33%
• Early Diagnosis 33%
• Explainable AI 100%
• Explainable Artificial Intelligence 33%
• Explanatory Method 33%
• Feature-based 33%
• Gammatone Cepstral Coefficients 33%
• Health Outcomes 33%
• Hidden Patterns 100%
• Hyperbilirubinemia 33%
• Infant Cry Signal 100%
• Jaundice 33%
• Medical Attention 33%
• Mel-spectrogram 33%
• Model Interpretability 33%
• Model Output 33%
• Model Trustworthiness 33%
• Multi-feature 100%
• Neonatal Complications 33%
• Pediatric Medicine 33%
• Respiratory Distress Syndrome 33%
• Sepsis 33%
• Spectrogram 33%
• Transformer-based Deep Learning 33%
• Unveiling 100%
• Vision Transformer 100%
• Vision Transformer Model 33%
• Vomiting 33%
• XAI Methods 33%

Medicine and Dentistry

• Artificial Intelligence 100%
• Cross-Validation 100%
• Health Outcomes 50%
• Hyaline Membrane Disease 50%
• Hyperbilirubinemia 50%
• Jaundice 50%
• Medicine 50%
• Pediatrics 50%
• Sepsis 50%