Cardiac arrest in newborn babies is a life-threatening medical emergency that requires early detection
and prompt treatment to reduce mortality and long-term health complications. Traditional monitoring
methods used in neonatal intensive care units mainly rely on manual observation and basic clinical
indicators, which often fail to identify subtle and early changes in a baby’s vital signs, leading to
delayed diagnosis and reduced survival chances. To overcome these limitations, the developed model
proposes a machine learning–based system using statistical models for the early detection of cardiac
arrest in newborn babies. The proposed system analyzes key physiological parameters such as heart
rate, respiratory rate, oxygen saturation, and medical history data to identify early warning signs
and potential risk factors. Machine learning techniques including Logistic Regression and Support
Vector Machines are applied to detect hidden patterns that are difficult to recognize using conventional
methods. The system enables automated, accurate, and continuous monitoring in the Cardiac Intensive
Care Unit (CICU), supporting early clinical decision-making and timely medical intervention. By
reducing reliance on manual monitoring, improving prediction accuracy, and minimizing response time,
the proposed approach enhances the quality of neonatal care and increases the chances of survival for
newborn babies through intelligent, data-driven healthcare solutions.
