This 14-Year-Old Built an AI App That Detects Heart Disease in Seven Seconds
Most 14-year-olds spend their summers playing video games, hanging out with friends, or trying to get through summer reading lists. Siddarth Nandyala spent his building an artificial intelligence-powered medical app that can detect heart disease in under seven seconds. He did not have a team of engineers or a university lab. He had a computer, a goal, and almost eight months of focused work.
The app is called Circadian AI, and here is what it actually does. You hold a smartphone near your chest, and the app records the sounds your heart makes, similar to how a doctor uses a stethoscope to listen to your heartbeat. The app uses enhanced noise-cancellation, the same kind of technology that blocks out background sound in high-end headphones, to make sure it captures a clean recording even in a noisy hospital waiting room. Those recorded sounds are then sent to a machine-learning model, a type of artificial intelligence that has been trained on thousands of real heartbeats to recognize patterns that signal a problem. Within seven seconds, the app produces a report about the health of the user's heart.
What Siddarth built is specifically designed to catch problems early, and that distinction matters enormously. Heart disease is the leading cause of death worldwide, responsible for nearly 20 million deaths in 2021 alone. One of the most dangerous things about cardiovascular disease, which is any condition that affects the heart and blood vessels, is that it is often asymptomatic in its early stages. Asymptomatic means it produces no noticeable symptoms, so a person can feel completely fine while their heart is quietly developing a serious problem. By the time most people find out something is wrong, it is often during a heart attack or stroke, which is far too late for easy treatment. Circadian AI is designed to be a pre-screening tool, meaning it flags potential problems before they become emergencies, giving doctors a head start.
The app can detect arrhythmias, which are irregular heartbeats, early signs of heart failure, coronary artery disease, and heart valve abnormalities. Siddarth tested it on roughly 15,000 patients in the United States and around 3,500 in India. During one trial at a government hospital in India, the app identified about ten patients with cardiovascular diseases, and doctors later confirmed those diagnoses using traditional medical equipment like electrocardiograms. Across all testing, the app achieved over 96 percent accuracy, which is a genuinely impressive number for any medical screening tool.
One of the most important things about Circadian AI is where it could be used. Access to medical care is not equal around the world. In many low- and middle-income countries, expensive equipment like cardiac MRI machines or specialist cardiologists simply are not available to most people. Studies show that 80 percent of deaths from cardiovascular disease happen in these lower-income nations, partly because early detection is out of reach for so many. A smartphone app that costs nothing to run and can be used by trained health workers at local clinics could change that reality for millions of people.
Jameel Ahmed, a heart specialist at Louisiana State University, called it especially impressive that a 14-year-old built something with the potential to affect millions of lives. He also noted one real limitation worth understanding: the quality of the microphone on the phone being used could affect how accurately the app hears the heart. Doctors have spent years learning to interpret subtle differences in heartbeat sounds, and even experts can sometimes be uncertain. Siddarth's app is not meant to replace a doctor but to act as a first-pass screening tool that can point people toward care they might never have sought on their own.
Siddarth is no stranger to ambitious projects. At 12, he designed a prosthetic arm that cost just $150 to build in India, compared to traditional models priced at up to $30,000. He founded a startup called STEM IT that sells science and engineering kits to students. He invented a fall-detection armband for elderly people that outperformed the Apple Watch in accuracy. He has received awards from a local chamber of commerce, a Certificate of Recognition from the U.S. House of Representatives, and a congratulatory letter from a former president. Now, at 14, he is heading to the University of Texas at Dallas to study computer science.
Circadian AI is currently going through the process of receiving regulatory approval, the official medical certification required before a tool can be used widely. Once approved, Siddarth plans to roll it out across primary health care centers in the U.S., India, and other countries. His goal has stayed the same from the very beginning: use technology to reach the people who need help most. As he puts it, even one life detected is one life saved.
Source: Smithsonian Magazine