Your lungs speak a complex language that doctors use to diagnose hidden health conditions
In short: When you breathe, your lungs produce specific acoustic signatures that reveal the state of your airways. While healthy lungs sound smooth and soft, abnormalities like wheezing or crackling act as auditory biomarkers. Modern technology is now learning to decode these sounds to diagnose conditions like asthma or monitor sleep stages.
Respiratory sounds, or lung sounds, are the noises generated by air moving through the respiratory system. In a clinical setting, healthcare providers use a stethoscope to auscultate the lungs, listening for the quality, pitch, and intensity of breath. Normal, healthy breathing is characterized by vesicular sounds—a soft, smooth noise produced when airways are fully open and unobstructed. When these sounds deviate from the norm, they are classified as adventitious, or added, sounds.
These abnormal sounds are categorized by their duration and musicality. Continuous sounds, such as wheezing and rhonchi, last approximately 250 milliseconds and often indicate narrowed airways or blockages. Wheezing is a high-pitched, hissing sound, while rhonchi are low-pitched and snoring-like, often caused by mucus. Conversely, crackles—also known as rales—are noncontinuous, explosive sounds that occur when air opens alveoli. These can be fine or coarse, depending on their pitch and duration. Pleural rub, another distinct sound, occurs when the lung lining grates against itself during breathing.
While these sounds provide vital clues, they are not diagnostic on their own. Clinicians must synthesize this auditory data with patient history and other tests. Advanced research is now pushing beyond the human ear. Machine learning models, such as those using YAMNet embeddings, have achieved high accuracy in detecting asthma by identifying specific auditory biomarkers. Similarly, researchers are using deep audio segmentation to estimate sleep stages by analyzing subtle variations in respiratory cycles. These non-invasive, passive monitoring techniques offer a promising future for scalable, computer-aided diagnostics that move beyond the limitations of traditional, resource-intensive methods.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Lung Sounds
- Also: Respiratory sounds
- Also: Hybrid deep learning and YAMNet features for asthma diagnosis from respiratory sounds.
- Also: Estimating Sleep-Stage Distribution from Respiratory Sounds via Deep Audio Segmentation.
To cite this card, link to https://intellectplus.co.uk/app/health-respiratory-sounds.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Medical procedures · More in Health · Latest · A–Z · All topics