NYC Gazette

Columbia biomedical engineers have made a significant breakthrough in heart research by utilizing artificial intelligence (AI) technology. The innovative tool, named BeatProfiler, promises to revolutionize the study and diagnosis of heart function. Developed by researchers at Columbia Engineering, BeatProfiler aims to streamline the analysis of heart cell function through the integration of AI and machine learning technologies.

"This is truly a transformative tool," expressed project leader Gordana Vunjak-Novakovic, emphasizing the efficiency and accessibility of BeatProfiler. Vunjak-Novakovic, a University Professor at Columbia, highlighted the software's ability to analyze various heart function indicators rapidly and accurately. The groundbreaking nature of BeatProfiler lies in its capacity to differentiate between diseases, assess drug effects on heart function, and significantly expedite the research process.

Lead author Youngbin Kim, a PhD candidate in Vunjak-Novakovic's lab, underscored the unprecedented speed and versatility of BeatProfiler in cardiac research. Kim stated, "Using machine learning, the functional measurements analyzed by BeatProfiler helped us to distinguish between diseased and healthy heart cells with high accuracy and even to classify different cardiac drugs based on how they affect the heart."

The Columbia Engineering team's decision to offer BeatProfiler as open-source software reflects their commitment to advancing heart research collaboratively. By providing the AI tool for free use, the researchers aim to facilitate its widespread adoption across academic, clinical, and commercial laboratories. This approach also enables feedback from users to further enhance the software's capabilities and refine its applications in heart research.

Driven by a clinical imperative to diagnose heart diseases more efficiently, the development of BeatProfiler was a culmination of years of research efforts. The software's evolution was guided by the need to accurately assess cardiac models in real-time, particularly in the context of exploring genetic cardiomyopathies, immune-mediated inflammation, and drug discovery. The team's interdisciplinary collaboration encompassed software development, machine learning, signal processing, and user experience to create a user-friendly interface for biomedical researchers.

Looking ahead, the researchers are focused on expanding BeatProfiler's functionalities to encompass a broader spectrum of heart diseases and drug development applications. By refining the machine-learning algorithm and validating its performance across diverse cardiac models, the team aims to accelerate drug testing processes and enhance the understanding of heart function through AI-driven analysis.

In conclusion, Columbia's BeatProfiler represents a significant advancement in heart research, leveraging AI technology to propel the study of cardiac function into a new era of efficiency and precision.