A Glimpse at the Cutting Edge of Cough Research
Cough is finally becoming something that medicine can measure. For the past two decades, long before the onset of COVID-19, researchers driven by the persistent need for effective cough solutions have been working to shed light on the science of coughs. So let us have a dive into some of the most important cough projects.
Acoustic Epidemiology and AI
Acoustic epidemiology analyzes sounds to understand health events, causes, and risk factors. The current technology can already record sounds. These sounds can then be processed by themselves or in correlation with other data.
Dr. Peter Small, Chief Medical Officer at Hyfe Inc., brings his expertise in infectious diseases and global health, with a specific emphasis on cough, to Hyfe, which has a mobile application that detects and tracks coughs in real-time when they happen, where they happen.
Hyfe’s researchers see a vast range of uses and adaptations of Hyfe’s platform for diagnostics and monitoring. They envision a new era of patient care and public health.
Dr. Small comments:
“It’s ironic how much people focus on counting steps while ignoring cough, which is far more consequential. Hyfe is a science-driven company with the technology to make cough count. Particularly now, with increased awareness of cough and the rapid growth of digital health driven by COVID-19, this technology can improve the lives of patients, the care provided by doctors, and the efficiency of health systems.”
Acoustic epidemiology and syndromic surveillance show that tracking sounds could help identify cough variances and anomalies faster than has ever been possible. In collaboration with the University of Navarra’s Zizur Health Center and Montreal University Hospital Center, Hyfe technology is playing a central role in clinical trials with the aim of finding out if tracking coughs can monitor respiratory diseases at community level and even predict future outbreaks.
To broaden the reach and increase the efficiency of clinical trials, Hyfe is collaborating with Obviohealth, a global virtual research organization, to help clinical researchers identify different strains of coughs, respiratory diseases, and also COVID-19.
Tuberculosis is preventable, treatable, and curable, yet it remains one of the deadliest diseases in the world. Simon Grandjean Lapierre, an infectious diseases physician, molecular biologist, and TB researcher, takes the lead on translational research programs focused on Tuberculosis control in Canada and abroad, notably in Madagascar.
His research activities principally include the impact assessment of new innovative technologies and molecular diagnostic tools. Partnering with Hyfe, he researches innovative technologies on digital cough monitoring and acoustic epidemiology and says,
“Cough is a hallmark symptom of tuberculosis, triggering a patients’ entry into TB screening programs and cascade of care. For the first time, we can objectively detect, monitor and classify cough sounds and learn how to use of this data-rich symptom to its full potential.’”
Digital Adherence technology is a low-cost treatment that addresses the problem of incomplete treatment in patients with tuberculosis as it provides acceptable alternative approaches to monitoring TB medication.
Using this technology, project ASCENT, funded by Unitaid, helps TB patients successfully complete their treatment with data-driven support interventions, utilizing smart pillboxes and other innovations. Implemented in July 2019, together with its battalion of partners, it aims to make future scale-up possible so that these digital innovations can be available to all TB patients worldwide.
Adithya Cattamanchi, working out of the University of California San Francisco, employs an innovative research program approach. Cattamanchi seeks to improve the diagnosis and treatment of patients with tuberculosis by identifying novel biomarkers for TB screening and diagnosis and developing point-of-care platforms for biomarker detection, conducting primary studies in the field, measuring the quality of care delivered to patients with TB, and using implementation science-based approaches to develop and test strategies to improve care delivery.
An ongoing study research lead by Prof. dr. Frank Cobelens in the Amsterdam Institute for Global Health and Development investigates a simple and easily applied method for TB screening referred to as Cough Audio Triage for Tuberculosis (CAGE-TB). Cage TB is an automated smartphone-based cough audio classification for rapid tuberculosis triage testing based on automatic cough sound analysis. The study seeks to develop a mobile application capable of on-device (offline) classification of TB in clinical trials.
During the pandemic, researchers developed several algorithms to detect COVID-19 coughs. A project, Cough Against Covid, funded by the Bill and Melinda Gates Foundation at the Wadhwani Institute For Artificial Intelligence in Mumbai, seeks to use AI to decipher if individuals could have the virus based on their cough sounds.
Computer Scientists at RMIT University in Australia have developed an AI model that can hear the effects of COVID-19 in the sound of a forced cough, even when people are asymptomatic. Study lead author Dr. Hao Xu and Co-Author Flora D. Salim used datasets from crowdsourcing platforms such as COVID-19 Sounds App and Coswara. Xu and Salim used these datasets to train the algorithm using contrastive self-supervised learning: a method by which a system works independently to encode what makes two things similar or different.
One study by a team of researchers showed that healthcare professionals can use AI algorithms to detect and diagnose different types of known diseases, including pneumonia, pulmonary edema, asthma, tuberculosis (TB), COVID-19, pertussis, and other respiratory diseases.
On cough managing strategies
In the American Cough Conference 2021, Laurie Slovarp, Ph.D. CCC-SLP, associate professor in the School of Speech, Language, Hearing, and Occupational Sciences at the University of Montana addresses cough by educating patients about coughing, training them to suppress their coughs safely while looking into underlying causes that trigger cough.
Professor Slovarp’s research shows how this approach (Behavioral Cough Suppression therapy) helps patients manage their coughs, prevent their cough, or even trigger their cough less often. A recent paper by Slovarp, Bridget Kathleen Loomis, and Amy Glaspey shows behavioral treatment to be effective in up to 85% of patients with refractory Chronic Cough (RCC).
Artificial Intelligence used as a tool in healthcare shows great promise to mitigate burdens on the health system while also providing the best possible data for patients, efficient diagnosis, and information on the prognosis of the diseases.
Cough monitoring and self-tracking are on the leading edge of the advent of acoustic epidemiology. The generation of more data will go hand in hand with the broader application of cough frequency measurement, producing a virtuous cycle that will lead to further advances in detecting, analyzing, and diagnostic technology.