PROTOTYPE v0.1 — KNN CLASSIFIER
AI Acoustic Biopsy
Real cough audio feature extraction compared against a reference library derived from published COUGHVID & Coswara respiratory research datasets using K-Nearest Neighbor classification.
01
Record
02
Profile
03
Analysis
04
Results
Cough Audio Capture
// WEB AUDIO API · FFT · REAL-TIME SPECTRAL ANALYSIS
LIVE WAVEFORM
0:00
Press the microphone and perform 3 forced coughs.
Recording auto-stops after 8 seconds.
Recording auto-stops after 8 seconds.
How to Cough Correctly
1
Sit upright, mic nearby2
Take a deep breath in3
Cough forcefully 3 times4
Wait for analysisHealth Risk Profile
// COMBINED WITH ACOUSTIC SCORE FOR FINAL TRIAGE
KNN Classifier Running
// COMPARING YOUR COUGH AGAINST REFERENCE LIBRARY
[INIT] Loading audio feature buffer...
Acoustic Classification Complete
// KNN TRIAGE SCORE — RESEARCH-GROUNDED
--
CALCULATING
K-Nearest Neighbor Matches
// YOUR COUGH VS COUGHVID/COSWARA REFERENCE PROFILES
Research basis: Reference feature ranges derived from published studies using the COUGHVID dataset (25,000+ recordings, Orlandic et al. 2021) and Coswara dataset. Features: spectral centroid (healthy > symptomatic), ZCR (restrictive disease = high, obstructive = low), RMS energy, and spectral bandwidth (1300–3200 Hz for coughs).
Extracted Audio Features
// RAW MEASUREMENTS FROM YOUR COUGH RECORDING
⚠️ Important: This is a research prototype pre-screening tool, not a medical diagnosis. The KNN classifier uses published acoustic feature ranges as a reference — it cannot replace clinical evaluation. Always consult a qualified healthcare provider.