|
Every second counts in cardiac arrest, and this project set out to create a device that could act in those critical moments. The purpose was to develop an intelligent system capable of detecting and responding to cardiac arrest in real time. The hypothesis was that combining continuous ECG monitoring with artificial intelligence would allow earlier detection and faster intervention than current approaches, increasing survival rates.
CROSHA was designed as a wearable vest that continuously recorded ECG signals and analysed them using CardioNet-1D, a custom-built convolutional neural network. This model was trained and validated on labelled ECG datasets to recognise patterns linked to cardiac arrest. When abnormal rhythms were detected, the device was programmed to deliver an immediate corrective shock while simultaneously sending emergency alerts to family members through the companion CROSHA app.
Data testing was carried out using open-source ECG datasets to evaluate accuracy and reliability. Results showed that CardioNet-1D achieved high levels of precision in identifying life-threatening rhythms, confirming the potential of this approach for real-world application.
In conclusion, this project demonstrated the feasibility of integrating artificial intelligence, continuous monitoring and automated intervention into a single life-saving system. CROSHA provides rapid detection, immediate treatment and instant communication, addressing the critical time window in cardiac arrest cases. By reducing delays between collapse and defibrillation, the system has the potential to save lives and provide reassurance for patients and families. A patent application has been submitted, and the system is currently patent pending.
|