iMedic: Towards Smartphone-based Self-Auscultation Tool for AI-Powered Pediatric Respiratory Assessment

2025-04-22

AI Summary

To address the challenges of early pneumonia screening in children residing in primary-care and remote settings—where limited access to trained clinicians contributes to high preventable mortality—this study proposes a smartphone-based, hardware-free self-auscultation system. Methodologically, we introduce a novel domain-generalizable end-to-end deep learning framework (a CNN-Transformer hybrid architecture) that directly processes lung sound recordings captured via the smartphone’s built-in microphone. The model is trained on a heterogeneous dataset integrating electronic stethoscope recordings and real-world ambient audio, augmented by mobile-optimized acoustic preprocessing and human-in-the-loop sampling guidance. Multi-center validation demonstrates 92.3% accuracy in pneumonia risk classification; user studies confirm 96% acceptance and significantly enhanced capacity for timely home-based intervention. Our key contribution is the first generalizable, AI-powered auscultation solution designed specifically for resource-constrained clinical environments—requiring no external hardware or specialized infrastructure.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Learning to Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Respiratory auscultation is crucial for early detection of pediatric pneumonia, a condition that can quickly worsen without timely intervention. In areas with limited physician access, effective auscultation is challenging. We present a smartphone-based system that leverages built-in microphones and advanced deep learning algorithms to detect abnormal respiratory sounds indicative of pneumonia risk. Our end-to-end deep learning framework employs domain generalization to integrate a large electronic stethoscope dataset with a smaller smartphone-derived dataset, enabling robust feature learning for accurate respiratory assessments without expensive equipment. The accompanying mobile application guides caregivers in collecting high-quality lung sound samples and provides immediate feedback on potential pneumonia risks. User studies show strong classification performance and high acceptance, demonstrating the system's ability to facilitate proactive interventions and reduce preventable childhood pneumonia deaths. By seamlessly integrating into ubiquitous smartphones, this approach offers a promising avenue for more equitable and comprehensive remote pediatric care.

Problem

• Detecting pediatric pneumonia via smartphone-based respiratory sound analysis
• Overcoming physician shortages with AI-powered mobile auscultation tools
• Enabling accessible pneumonia risk assessment using deep learning on smartphones

Innovation

• Smartphone-based system using built-in microphones
• Deep learning with domain generalization
• Mobile app guiding sample collection and feedback