Researchers Help AI Express Uncertainty to Improve Health Monitoring Tech

A team of engineering and health researchers has developed a tool that improves the ability of electronic devices to detect when a human patient is coughing, which has applications in health monitoring. The new tool relies on an advanced artificial intelligence (AI) algorithm that helps the AI better identify uncertainty when faced with unexpected data in real-world situations.

"When AI is being trained to identify the sound of coughing, this is usually done with 'clean' data - there is not a lot of background noise or confusing sounds," says Edgar Lobaton, corresponding author of a paper on the work and an associate professor of electrical and computer engineering at North Carolina State University. "But the real world is full of background noise and confusing sounds. So previous cough detection technologies often struggled with 'false positives' - they would say that someone was coughing even if nobody was coughing.

"We've developed an algorithm that helps us address this problem by allowing an AI to express uncertainty. Rather than having to decide 'Yes, that was a cough' or 'No, that wasn’t a cough,' the AI can also report that it has detected a sound it's not familiar with. In other words, the AI is given a third option: 'I don't know what that was.'"

Cough detection technology is of interest for several potential health monitoring applications.

"For example, there is interest in using wearable health monitoring devices that would detect coughs in people who have asthma, which could trigger a notification about increased risk of an asthma attack," Lobaton says. "There is also interest in using cough detection for COVID monitoring, and so on."

However, previous cough detection technologies had high rates of false positives, with the relevant AI reporting many unfamiliar sounds as coughing. These false positives significantly limited their utility.

"In the near term, our work limits the reporting of false positives by allowing the AI to note when it hears sounds that it can't identify," Lobaton says. "In the longer term, our algorithm should allow us to continually train the AI, by telling it whether the unfamiliar sounds it is hearing are coughs or are unrelated noises. This should allow for much more precise detection over time."

In addition, the researchers tested the new algorithm in computational models and found that the modified cough detection AI can operate effectively using far fewer sound samples per second than previous technologies. For example, previous cough detection tools used approximately 16,000 sound samples per second, while the new AI tool makes use of 750 sound samples per second, with similar sensitivity and fewer false positives.

"Using fewer sound samples is a significant advantage for two reasons," Lobaton says. "First, it means that the electronic device requires less computing power - which allows us to make it smaller and more energy efficient. Second, using fewer sound samples means that the technology will not be recording understandable speech, which addresses privacy concerns."

The researchers are currently in the process of incorporating the new algorithm into a wearable health monitoring device that can be used in real-world testing.

What's more, the researchers say the approach they’ve taken here could be used to address a range of AI applications in which the AI is likely to encounter unexpected input that it was not trained to understand.

"We're looking for research partners who can help us explore other health monitoring challenges that this AI modification could help address in a meaningful way," Lobaton says.

Chen Y, Attri P, Barahona J, Hernandez ML, Carpenter D, Bozkurt A, Lobaton E.
Robust Cough Detection with Out-of-Distribution Detection.
IEEE J Biomed Health Inform. 2023 Apr 5;PP. doi: 10.1109/JBHI.2023.3264783

Most Popular Now

Giving Doctors an AI-Powered Head Start …

Detection of melanoma and a range of other skin diseases will be faster and more accurate with a new artificial intelligence (AI) powered tool that analyses multiple imaging types simultaneously...

AI Agents for Oncology

Clinical decision-making in oncology is challenging and requires the analysis of various data types - from medical imaging and genetic information to patient records and treatment guidelines. To effectively support...

AI Medical Receptionist Modernizing Doct…

A virtual medical receptionist named "Cassie," developed through research at Texas A&M University, is transforming the way patients interact with health care providers. Cassie is a digital-human assistant created by Humanate...

Using Data and AI to Create Better Healt…

Academic medical centers could transform patient care by adopting principles from learning health systems principles, according to researchers from Weill Cornell Medicine and the University of California, San Diego. In...

AI Tool Set to Transform Characterisatio…

A multinational team of researchers, co-led by the Garvan Institute of Medical Research, has developed and tested a new AI tool to better characterise the diversity of individual cells within...

AI Detects Hidden Heart Disease Using Ex…

Mass General Brigham researchers have developed a new AI tool in collaboration with the United States Department of Veterans Affairs (VA) to probe through previously collected CT scans and identify...

Human-AI Collectives Make the Most Accur…

Diagnostic errors are among the most serious problems in everyday medical practice. AI systems - especially large language models (LLMs) like ChatGPT-4, Gemini, or Claude 3 - offer new ways...

Northern Ireland Completes Nationwide Ro…

Go-lives at Western and Southern health and social care trusts mean every pathology service is using the same laboratory information management system; improving efficiency and quality. An ambitious technology project to...

Highland Marketing Announced as Official…

Highland Marketing has been named, for the second year running, the official communications partner for HETT Show 2025, the UK's leading digital health conference and exhibition. Taking place 7-8 October...

MHP-Net: A Revolutionary AI Model for Ac…

Liver cancer is the sixth most common cancer globally and a leading cause of cancer-related deaths. Accurate segmentation of liver tumors is a crucial step for the management of the...

Groundbreaking TACIT Algorithm Offers Ne…

Researchers at VCU Massey Comprehensive Cancer Center have developed a novel algorithm that could provide a revolutionary tool for determining the best options for patients - both in the treatment...

The Many Ways that AI Enters Rheumatolog…

High-resolution computed tomography (HRCT) is the standard to diagnose and assess progression in interstitial lung disease (ILD), a key feature in systemic sclerosis (SSc). But AI-assisted interpretation has the potential...