MAILab Unveils AI to Detect Congenital Heart Disease in Children: A Life-Saving Breakthrough for Family Health

Artificial intelligence (AI) company MAILab announced today the release of a breakthrough AI algorithm designed to assist in the early detection of congenital heart disease (CHD), one of the leading causes of death among children. In honor of Family Month in May, this release underscores MAILab’s commitment to protecting children’s health through innovation. The algorithm was developed in collaboration with Professor Jeongjin Yoo’s team at the Congenital Heart Disease Center of Seoul Asan Medical Center, based on over 5,000 pediatric echocardiography scans, making it one of the most advanced diagnostic tools of its kind.

The model can accurately detect major CHD defects such as ASD, VSD, and PDA, as well as the more complex tetralogy of Fallot (TOF), using real-time streaming video analysis. Notably, this algorithm was trained and optimized on MAILab’s proprietary AI supercomputer, which was co-developed with NVIDIA and Qualcomm. With a 2 PFLOPS computing capacity, this infrastructure enables the high-speed processing of large-scale medical images and supports deep learning training based on video streaming — a key component of the system's performance.

Photo caption: (From left) Dr. Yeonjin Jang, Chief Medical Officer at MAILab; Dr. Phil Yang, CEO of MAILab; and Professor Jeongjin Yoo, Congenital Heart Disease Center, Seoul Asan Medical Center.

Summary of Core Technologies

MAILab’s CHD AI algorithm achieves an impressive 98.7% diagnostic accuracy, operating not on still images but on live ultrasound video. It includes automated detection and visualization of regions of interest (ROI) and can accurately identify even complex congenital heart conditions such as TOF. Crucially, the algorithm was trained and optimized on MAILab’s 2 PFLOPS AI supercomputer, co-developed with NVIDIA and Qualcomm. This infrastructure plays a vital role in enabling real-time deep learning training using high-resolution video streams, thanks to its ultra-fast processing capabilities.

Why This Technology Saves Lives

CHD affects approximately 1 in every 100 newborns, and early diagnosis can result in over 90% survival rates. However, early detection remains challenging due to complex image interpretation and a shortage of pediatric cardiologists. MAILab’s AI has been designed to assist even non-experts, enabling rapid diagnosis and effective screening even in medically underserved regions.

Accelerating Global Expansion — Discussions with Google and Microsoft

On April 28–29, MAILab held strategic discussions with Google and Microsoft in Silicon Valley to explore joint development in medical AI, global market expansion, and investment partnerships. These meetings are considered a major milestone in MAILab’s integration into the global healthcare ecosystem beyond Korea.

Expanding Research — Unveiling Kawasaki Disease Analysis

In partnership with Professor Jeongjin Yoo’s team, MAILab recently published a machine learning-based analysis on the pathology of Kawasaki disease, another pediatric heart condition, in the international journal Biomedicine (April 29, 2025). The study represents significant progress in the early prediction and prevention of pediatric cardiovascular diseases.

Dr. Phil Yang, CEO of MAILab, stated, “If our technology can help even one child, it will have been worth it. As we celebrate Family Month in May, I hope this inspires reflection on how technology can protect the health and happiness of our families.”

About MAILab

MAILab (www.mailab.co.kr) is a global AI company that delivers solutions across healthcare, industry, logistics, security, and beauty, powered by its proprietary AI engine i2Brain and the world’s largest medical imaging dataset. Headquartered in Seoul, the company operates R&D hubs in the U.S., Canada, China, and India.

Press Contact

MAILab PR Team

Email: sryoo@ckt21.com

Website: www.mailab.co.kr