The Research Roundup

New cough science publications vetted and collected in one place‍

1. Could AI evaluate pulmonary function using only cough sounds?

Key Takeaway: The results are promising! On a small sample size, the investigated AI model showed good performance in basic pulmonary function parameters - FEV1, FEV1%, FEV1/FVC, FVC, and FVC%.

Why It Matters: The standard method for evaluating pulmonary function, spirometry, is complex and hard for patients to self-test accurately. AI analysis of cough sounds could replace or complement this with non-invasive monitoring for faster, higher-quality insights into disease progression.

Read Publication‍

2. Is mucus transpiration key to understanding environmental impact on cough hypersensitivity?

Key Takeaway:  Dry and polluted air exacerbates osmotic pressure differences across mucus that are pivotal in transpiration. This potentially leads to increased ATP levels, inflammation, acidity, and changes in the permeability of mucus water - a ‘cascade of events’ that is believed to contribute to cough hypersensitivity.

Why it Matters: By better understanding and addressing the root causes of mucus transpiration – reaction to dry and polluted air – more effective and targeted therapies might be developed, offering relief for chronic coughers.

Read Publication

3. Can nighttime audio of nonverbal vocalizations improve assessment of asthma patients?

Key Takeaway: Combining overnight audio recordings of cough and other nonverbal vocalizations with spirometry data resulted in a 30% improvement in FEV1 forecasting.

Why It Matters: Additional audio data, collected non-intrusively via a bedside smartphone, could lead to a more comprehensive view of an asthma patient’s condition, paving the way for more tailored and effective asthma management strategies.

Read Publication

Plus:

• Neural networks can improve the ability to remove everyday background noise from cough sound recordings, rendering them more useful for diagnosis of respiratory conditions (More).
• Diagnosis based on cough sound in primary care has potential, but more comprehensive data is required (More).
• A publicly accessible cough counting dataset of multimodal biosignals is used to train a model that can operate entirely on-device (“edge-AI”) (More).

‍

Deep Dive with Researchers

Exploring the Possibilities of Longitudinal Cough Data

‍Q&A with Dr Dominic Sykes

Could esophageal dysmotility be a cause of cough hypersensitivity syndrome?

“This work is important for a multitude of different patients who suffer from chronic cough… [and] we hope to delineate a common pathology of esophageal dysmotility in these conditions and present a simple way to optimize the management of such patients.”

We spoke with Dr Sykes to know more about his current clinical trial that is examining the effect of azithromycin on cough in patients with a multitude of chronic respiratory diseases. This is the first trial ever to integrate continuous longitudinal cough data with esophageal manometry.

Read the full Q&A here.

‍

Cough Science Events Ahead

6th Cough Forum: Cough Science enters precision medicine

It is a new era in cough research, driven by AI advancements and the ability to monitor patients' cough frequency over time in their daily lives.

Join the Cough Science Forum for the latest breakthroughs and learn how cough data can predict clinical outcomes, facilitate early diagnosis, assess treatment response, and be a vital signal for disease progression.

Previous highlights include:

• Digital cough monitoring predicts clinical outcomes in hospitalized COVID-19: a predictive acoustic biomarker
• Continuous cough monitoring in a bronchiectasis treatment program
• Using cough sounds to improve the allocation of chest radiography
• Behavioral cough suppression therapy

🗓️ Next Cough Science Forum: February 14th

🕜 Time: 3.30PM UTC / 10:30 AM EST

Register Now

‍