New AI-Powered Mobile App Predicts Cardiovascular Events with High Accuracy

Before Health Intelligence, Ltd. emerged from stealth with the announcement of Antshrike™, an mHealth app that uses artificial intelligence to transform wearable device data into precise predictive cardiovascular health warnings. The iOS application employs proprietary AI algorithms to identify early warning signs of critical cardiovascular events such as heart attacks or strokes with high predictive accuracy for at-risk individuals.

The app processes biometric time-series data collected by commercially available Apple smartphones and smart watches, calculating individual risk of critical events in real time and predicting when and how changes in a patient's health states may occur. By notifying patients of imminent danger in advance, Antshrike enables timely intervention by healthcare providers, facilitating personalized medical intervention and treatment while improving medical outcomes.

According to Before Health Intelligence CEO and co-founder Ekaterina Mitricheva, MD, PhD, "With cardiovascular disease as the number one cause of death worldwide, there is an urgent need to predict and prevent heart attacks and strokes." The app passively monitors health status without interrupting daily routines, social interactions, and work habits while predicting the danger and timing of cardiovascular events with more than 85% accuracy in a personalized manner.

Antshrike uses proven science based on decades of research, analyzing more than 130 billion data points along with open source CDC data from tens of thousands of individuals. The application assures HIPAA-compliant patient privacy by preprocessing individual patient data on a patient's phone with anonymized preprocessed values that cannot be traced back to any specific person. It requires no manual input of data or reliance on expensive hospital equipment and procedures.

The app utilizes two layers of AI—static and dynamic—to calculate an individual's risk of critical events in real-time. The static layer processes data entered by an individual to determine an initial risk score, with the machine learning algorithm trained on over 500,000 CDC datapoints from approximately 23,000 individuals. The dynamic layer collects and analyzes biometric time-series from smart watches in real-time, using the proprietary AI algorithm to identify early warning signs of critical state transitions before delivering a personalized, time-sensitive risk score for upcoming major ischemic cardiovascular events. The freemium app is expected to be available in the U.S. in 2025. More information about the product can be found at https://www.before-intelligence.com/contact.