Respiratory diseases are major global killers, demanding early diagnosis for effective management. Digital stethoscopes offer promise, but face limitations in storage and transmission. A compressive sensing-based compression algorithm is needed to address these constraints. Meanwhile, fast-reconstruction CS algorithms are sought to balance speed and fidelity. Sound event detection algorithms are crucial for identifying abnormal lung sounds and augmenting diagnostic accuracy. Integrating these technologies can revolutionize respiratory disease management, enhancing patient outcomes.
Globally, respiratory diseases are the leading cause of death, making it essential to develop an automatic respiratory sounds software to speed up diagnosis and reduce physician workload. A recent line of attempts have been proposed to predict accurately, but they have yet been able to provide a satisfactory generalization performance. In this contest, we invited the community to develop more accurate and generalized respiratory sound algorithms. A starter code is provided to standardize the submissions and lower the barrier.
