Smartphone app detects TB through coughs, but limitations remain
A new smartphone app has shown promise in diagnosing tuberculosis (TB) by analyzing cough sounds. Developed by a team of researchers from the US and Kenya, the app uses artificial intelligence (AI) to analyze recordings of coughs and determine the likelihood of a TB infection. Although the app is not foolproof and failed to detect TB in about 30% of cases, it is still considered a valuable screening tool, particularly in low-income countries where traditional diagnostics methods are costly and labor-intensive.
Acoustic biomarkers: How cough sounds can indicate changes in health
The concept of using cough sounds as acoustic biomarkers to indicate changes in health has been around for several decades. However, recent advancements in AI technology have significantly accelerated research in this field. With AI, larger quantities of cough data can be analyzed faster, making it easier to identify patterns and anomalies associated with various health conditions. Coughs are especially easy for smartphones to capture, which makes them an ideal data source for developing health monitoring apps.
Surge in interest for cough analysis apps driven by COVID-19 pandemic
The COVID-19 pandemic has further fueled interest in cough analysis apps. The acoustics of coughs have become a focus of numerous startups aiming to develop tools for diagnosing COVID-19 and other respiratory diseases. For example, AudibleHealthAI launched a mobile app called AudibleHealth DX, designed to diagnose COVID-19, and is currently awaiting FDA approval. Additionally, ResApp Health, an Australian company, developed an audio-based COVID-19 screening test that correctly identified 92% of positive cases just from cough sounds. The growing interest in cough analysis apps has been driven by the urgent need for non-invasive and easily accessible diagnostic tools in the face of a pandemic.
Cough-tracking apps as valuable health tools and game-changers for clinical trials
While some experts remain skeptical about the accuracy of cough analysis apps as standalone diagnostics, they believe these apps can still prove valuable as screening tools and health monitoring aids. Detecting coughs, regardless of the specific cause, can help identify individuals with potential health issues and enable early intervention. Moreover, cough-tracking apps could revolutionize clinical trials where coughs are essential measurements. These apps provide accurate and objective data on cough frequency, which is often challenging to track using traditional methods such as patient recall. The widespread availability of smartphones makes it easier to capture and analyze cough data, leading to more reliable and precise results.
Developing a massive database of voice and cough sounds for diagnosing various diseases
In an effort to advance the field of acoustic biomarker research, a team of researchers led by Yael Bensoussan is developing a massive database of voice, cough, and respiratory sounds. The aim is to create a publicly available resource that can aid in the development of diagnostic tools for various diseases, including cancers, respiratory illnesses, neurological and mood disorders, and speech disorders. The database will capture a wide range of sounds, enabling researchers to identify specific patterns associated with different health conditions. By making the data publicly available, the researchers hope to foster collaboration and validation of the research, leading to more accurate and reliable diagnostic tools in the future.
As the field of acoustic biomarker research continues to evolve and more data becomes available, the potential for cough and speech analysis apps to revolutionize healthcare diagnosis and screening is promising. While there are still limitations and challenges to overcome, these apps have the potential to greatly improve early detection and intervention in various health conditions. Continued research, validation, and collaboration will be crucial in unlocking the full potential of this technology and improving healthcare outcomes worldwide.
Analyst comment
Positive news: Smartphone app detects TB through coughs, but limitations remain
Short analysis: The development of a smartphone app that uses AI to analyze cough sounds for diagnosing tuberculosis (TB) shows promise, particularly in low-income countries. Despite limitations and a 30% failure rate, it is considered a valuable screening tool. Interest in cough analysis apps has surged due to COVID-19, and they are seen as valuable health tools and game-changers for clinical trials. Efforts to develop a comprehensive database of voice and cough sounds for diagnosing various diseases aim to revolutionize healthcare diagnosis and screening. Continued research and collaboration are needed to unlock the technology’s full potential.