Mitigating Stethoscope-Induced Shortcuts in Respiratory Sound Classification under Federated Domain Generalization with Causality-Inspired Interventions
A technical paper proposing causality-inspired methods to reduce stethoscope-specific shortcuts in federated respiratory sound classification, improving robustness and generalization.
A technical paper proposing causality-inspired methods to reduce stethoscope-specific shortcuts in federated respiratory sound classification, improving robustness and generalization.
Auto-scored: This is a real but partial technical step toward more reliable medical AI, with no direct risk escalation.