AI model CoughSense classifies five respiratory diseases from coughs

By PulseAugur Editorial · [1 sources] · 2026-06-03 04:00

Researchers have developed CoughSense, a system designed to classify respiratory diseases from cough recordings using AI. The system fine-tunes OpenAI's Whisper encoder and employs a dual-encoder cross-attention fusion with balanced contrastive learning to distinguish between five conditions: healthy, COVID-19, asthma, bronchitis, and pneumonia. CoughSense achieved 82.3% balanced accuracy in cross-validation, outperforming other models and demonstrating the effectiveness of its novel active-frame QKV attention pooling technique. AI

IMPACT Enables more nuanced and accessible respiratory health screening through AI-powered cough analysis.

RANK_REASON Academic paper detailing a new AI model and its performance on a specific task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

asthma
bronchitis
Coswara
CoughSense
CoughVID
COVID-19
EfficientNet-B2
ImageNet
OpenAI
OPERA-CT
pneumonia
Virufy
ViT
West China Hospital Pediatric Cough Dataset
Whisper

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

COVERAGE [1]

arXiv cs.LG TIER_1 English(EN) · Nikhil Vincent · 2026-06-03 04:00

CoughSense: Five-Class Respiratory Disease Classification via Whisper Encoder Fine-Tuning and Dual-Encoder Cross-Attention Fusion with Balanced Contrastive Learning

arXiv:2606.02998v1 Announce Type: new Abstract: Automated cough analysis offers a path to low-cost respiratory screening, but most existing work stops at binary COVID-19 detection. A practical tool needs to tell apart several respiratory conditions from one cough recording on a c…

COVERAGE [1]

CoughSense: Five-Class Respiratory Disease Classification via Whisper Encoder Fine-Tuning and Dual-Encoder Cross-Attention Fusion with Balanced Contrastive Learning

RELATED ENTITIES

RELATED TOPICS