The latest innovation in medical technology is the AI stethoscope, which goes beyond simply listening to heartbeats.
Researchers from Imperial College London and Imperial College Healthcare NHS Trust have found that these AI stethoscopes can detect heart failure at early stages.
A study published in the BMJ Journals by Tricorder revealed that these AI-enabled stethoscopes can assist doctors in identifying three cardiac conditions in just 15 seconds.
Partially backed by the British Heart Foundation (BHF), the study involved analyzing data from over 1.5 million patients, particularly those showing symptoms of heart failure like shortness of breath, swelling, and fatigue.
Out of 12,725 patients tested with the AI technology, those who underwent the AI stethoscope examination had a twice as high likelihood of being diagnosed with heart failure compared to those who did not use this tool.
Additionally, the AI stethoscope increased the odds of identifying atrial fibrillation, an abnormal heart rhythm that heightens stroke risk, by about 3.5 times for the patients examined.
In fact, patients were also twice as likely to be diagnosed with heart valve disease, where one or more heart valves malfunction.
“Early diagnosis is crucial as it provides access to necessary treatments that can help extend lives,” researchers noted.
A press release from BHF emphasized the importance of early diagnosis for these conditions to potentially receive life-saving treatment.
Dr. Patrick Bectiger, a clinical lecturer at Imperial College, highlighted that the traditional stethoscope design, which dates back to 1816, has remained unchanged for two centuries.
“It’s remarkable that a smart stethoscope can deliver test results within 15 seconds. After that, AI can quickly determine whether someone has heart failure, atrial fibrillation, or heart valve disease,” he stated.
Dr. Sonya Babu Narayan, a cardiologist and clinical director at BHF, pointed out that these heart ailments often get diagnosed at advanced stages when patients seek emergency care.
“Access to prompt diagnosis allows individuals to receive timely treatment, enhancing their chances of longevity,” she added.
The CDC reports that around 6.7 million individuals in the United States are currently living with heart failure, which can manifest through symptoms such as shortness of breath, fatigue, weight gain, and swelling of various body parts.
The risk factors for heart failure include diabetes, obesity, high blood pressure, and heart disease.
How the Device Works
Shaped like a player card, the AI stethoscope records ECG signals from the heart and captures the sound of blood flow through the heart.
This recorded data is processed by AI algorithms trained on health information from tens of thousands of individuals, generating assessments on heart failure risk.
There’s another algorithm specifically aimed at detecting atrial fibrillation, which often remains symptomless.
Risks and Limitations
However, researchers noted that around 70% of healthcare providers utilizing smart stethoscopes stopped after a year, indicating the need for better integration into standard practice.
The study also found that two-thirds of those suspected of having heart failure were later ruled out after further tests like blood samples or heart scans.
“This could potentially lead to unnecessary worry and testing for some patients,” BHF cautioned. They stressed that the AI stethoscopes should primarily be used on those already showing signs of heart issues, rather than for routine assessments of healthy individuals.
In a recent interview, cardiothoracic surgeon Dr. Jeremy London discussed the impact of AI in medicine, suggesting that it often dictates medical protocols but may not always cater to the specific needs of patients.
“While AI paves the way for advancements, ensuring that there’s a human touch in patient care remains crucial,” he remarked.
“AI stethoscopes are best reserved for individuals with current symptoms suggesting heart issues, not for routine checks on healthy people,” reiterated BHF.
He further emphasized that while AI may serve as a guideline, it should not be seen as a catch-all solution, as errors can occur.
Overall, many in the medical field believe AI will eventually provide more advantages than disadvantages in healthcare.
Tricorder’s study was funded through agencies like the National Institutes of Health (NIHR), BHF, and Imperial Health Charities, and was presented at the recent European Society for Cardiology meeting in Madrid.
