AI Framework Developed to Classify Lung Conditions

Researchers created a model that accurately differentiates between asthma and COPD using lung sound recordings.

Updated on Oct. 10, 2026 in Asthma

Bold flat-color editorial illustration featuring a geometric abstract bronchial structure, symbolizing advanced medical AI diagnostic analysis.
Researchers have developed the AVMD-ST framework, an AI-based system that uses acoustic analysis of lung sounds to differentiate between asthma and COPD with 95.75% accuracy. AI Illustration. Upload story photo >

Live Poll

Would you trust an AI-based diagnostic tool to screen for respiratory conditions in your area?

Scientists have introduced the AVMD-ST framework, an artificial intelligence system designed to improve the diagnosis of chronic lung conditions. The model demonstrated a 95.75% accuracy rate in classifying asthma and COPD through acoustic analysis.

Why it matters

Distinguishing between asthma and COPD is clinically challenging due to their shared physiological characteristics. This tool offers a more precise method for identifying specific respiratory issues using non-invasive sound data.

The study analyzed 200 participants, including 100 healthy individuals, 50 asthma patients, and 50 COPD patients. The model achieved 95.75% classification accuracy using lung sound recordings processed through FIR-conditioned VMD.

The players

AIIMS Raipur

This public medical research university and hospital served as a key site for collecting the lung sound recordings used in the study.

The details

The framework utilizes Finite Impulse Response (FIR) filtering to reduce background noise, combined with Variational Mode Decomposition (VMD) to isolate distinct respiratory sounds like wheezes and crackles. These sounds are then converted into Mel-spectrograms for processing by the AI's audio transformer model.

Timeline

  1. The research findings regarding the AVMD-ST framework were published on October 10, 2026.

The Big Picture

The study utilized the ICBHI respiratory sound database to validate its classification framework, following a pattern set by previous diagnostic AI research.

This diagnostic tool could eventually allow doctors to provide faster and more accurate treatment plans by correctly identifying lung conditions through simple audio recordings. It aims to reduce the confusion often caused by the overlapping symptoms of asthma and COPD.

The takeaway

Improving diagnostic precision for chronic respiratory issues helps prevent the mismanagement of symptoms that can arise from misdiagnosis. Patients experiencing chronic breathing difficulties should continue to work closely with pulmonologists to interpret their specific clinical test results.

Further reading

Learn more about the latest research in this field at the Asthma section.

Source note: This article includes information reported by Nature.

Live Poll

Would you trust an AI-based diagnostic tool to screen for respiratory conditions in your area?