Leveraging IoTs and Machine Learning for Patient Diagnosis and Ventilation Management in the Intensive Care Unit

Area of Science:

• Healthcare Technology
• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Clinical decision support systems (CDSS) are crucial for enhancing clinician decision-making in future healthcare.
• Intensive care units (ICUs) generate vast amounts of streaming physiologic data from medical devices.
• The COVID-19 pandemic highlighted the need for rapid detection of respiratory conditions like acute respiratory distress syndrome (ARDS).

Purpose of the Study:

• To design and develop a research-focused CDSS for ICU patient management.
• To leverage Internet of Things (IoT) devices for real-time physiologic data collection.
• To create machine learning (ML) models for automated clinical event recognition and diagnosis.

Main Methods:

• Developed a CDSS integrating IoT devices for streaming data from ventilators and other medical equipment.
• Created ML models to analyze physiologic data for harmful ventilator therapy detection and ARDS identification.
• Aggregated ML models into a mobile application for real-time provider alerts.

Main Results:

• Demonstrated the capability of CDSS to analyze physiologic data for clinical event recognition.
• Showcased the potential for automated diagnosis of conditions like ARDS using ML models.
• Successfully integrated data analysis and alerting into a mobile application.

Conclusions:

• CDSS, enhanced by IoT and ML, can significantly improve ICU patient management.
• Real-time analysis of physiologic data enables early detection of critical conditions like ARDS.
• This research paves the way for advanced, AI-driven CDSS in hospital settings.