Netherland develops app to warns asthmatic people to keep them safe

Netherland develops app to warns asthmatic people to keep them safe

An app developed by scientists from Maastricht University in Netherlands gives a warning for symptoms of asthma or chronic obstructive pulmonary disease (COPD) using voices from a mobile.

Yes, voice changes show significant promise for detecting asthma and monitoring airway condition. Recent research indicates that AI-driven voice analysis can identify asthma with over 80% accuracy, potentially offering a non-invasive and accessible screening tool.

How Asthma Affects the Voice

Asthma's impact on the respiratory system can subtly alter voice production in several ways:

  • Airway Inflammation and Narrowing: The primary feature of asthma, which changes airflow and the sound of the voice.
  • Changes in Vocal Fold Structure: Inflammation can cause swelling and increased mucus on the vocal folds, affecting how they vibrate.
  • Impact on Vocal Cord Function: The effort required to breathe and speak can alter vocal cord movement and tension.

These changes can manifest as specific, measurable features in a person's voice, known as "vocal biomarkers," that AI algorithms can detect.

How AI Detects Asthma from Voice

Modern studies use machine learning to analyze voice recordings for subtle patterns invisible to the human ear. The process generally involves:

  1. Recording Voice Samples: Patients produce simple sounds like a sustained /??/ vowel sound, phonemes like [i, a, u] at different pitches, or read a standard passage.
  2. Analyzing with AI: Algorithms analyze the recording to identify specific acoustic features (e.g., frequency, pitch, spectral energy).
  3. Classifying the Condition: The model compares the voice pattern to known data from healthy and asthmatic individuals to generate a prediction.

The results from recent studies demonstrate the effectiveness of this approach:

Study

Analysis Method

Key Finding

Li et al. (2025)

Support Vector Machine & Random Forest

Achieved 87% accuracy in identifying asthma patients.

Springer Study (2025)

XGBoost Model

Achieved 85% accuracy, with formant frequencies identified as top predictors.

AI Prediction Study (2025)

Decision Tree & CNN Models

Achieved up to 88% accuracy using different pitch phonemes.

One study also found that vocal biomarkers can correlate with changes in lung function (FEV1) during a methacholine challenge test, suggesting voice analysis could help monitor airway constriction and therapy response.

The Potential and Current Reality

This technology is not meant to replace current diagnostic methods like spirometry but could serve as an accessible screening tool for early detection or remote monitoring. The research is promising, but it's still in development.

We wish them all the best for more progress

By Jamuna Rangachari

 

Life Positive 0 Comments 2026-07-05 24 Views

Discussion (0)

No comments yet. Be the first to comment!

Leave a Comment

You need to login to post a comment.