Multiple Sensors, A.I. Power Digital Stethoscope

(Gerd Altmann, Pixabay)

29 Oct. 2018. Researchers in Rochester, New York are developing a system combining functions of a conventional stethoscope and an electrocardiogram or ECG to better check on implanted heart pumps. The multi-functional digital stethoscope, now a working prototype, is a project of engineers at Rochester Institute of Technology and medical researchers at University of Rochester Medical Center.

The team from University of Rochester’s Cardiovascular Engineering Lab led by cardiologist Karl Schwarz, and RIT biomedical and mechanical engineers Steven Day and Jason Kolodziej respectively, are seeking better technologies to monitor heart failure patients with implanted heart pumps, known as left-ventricle assist devices. Heart failure is a condition where the heart cannot pump enough blood to meet the body’s needs, a disorder affecting some 5.7 million people in the U.S., according to National Heart, Lung, and Blood Institute. Most cases of heart failure affect both the left and right sides of the heart, although in some cases only one side is affected.

A left-ventricle assist device is a battery-powered pump that helps the left ventricle, the heart’s main pumping chamber, send blood through the aorta to the rest of the body. The device is implanted in people with heart failure as a temporary measure until a heart can be found for transplant, or in some cases as a long-term solution to prolong and improve the quality of a patient’s life. While various imaging and diagnostic systems, such as echocardiograms and magnetic resonance imaging, or MRIs, are used to monitor these implanted heart pumps, often the first signs of problems with the device are found with acoustic indicators heard by a physician through a stethoscope.

The team designed its advanced digital stethoscope prototype to detect sounds indicating those problems, which now require a number of separate systems. “This stethoscope that we designed integrates multiple sensors into one,” says Kolodziej in an RIT statement. “It is not necessarily a new sensor that measures something never measured before; it is something that in a hospital setting would be done by multiple machines. Most cardiologists aren’t wheeling around this data collection hardware.”

The new device has a microphone to record audio signals from the patient’s heart and heart pump, much like a conventional stethoscope. But it also has ECG leads to record electrical signals from the heart. Moreover, after these audio and ECG signals are captured, they’re analyzed with machine-learning algorithms, a type of artificial intelligence, to provide actionable information to the patient’s physician. The team 3-D printed its advanced digital stethoscope prototype in RIT’s Construct lab, the campus’s maker-space.

The device’s algorithms aim to emulate the physician’s judgment and decision-making processes, based on the signals captured by the device. “It is almost unconscious, like people listening to engines, and hear something off that could mean a problem,” notes Day. “But if you ask what sounds different, they may not be able to explain it or articulate it, but they just know. This is about getting the device to do that, to hear sounds and then to make that determination of what the problem could be.”

More from Science & Enterprise:

• Univ. Spin-Off Developing A.I.-Boosted Heart Monitor
• Non-Invasive Imaging Technology Probes Deeper Into Tissue
• Tissue-Engineered Heart Chamber Model Developed
• Computer-Designed Heart Valves Operate Over Long Term
• Trial Shows Patch Feasible for Heart Rhythm Detection

*     *     *