Share

Towards smarter breathing support in overburdened ICUs

December 3, 2025

Lars van de Kamp defended his PhD thesis at the Department of Mechanical Engineering on December 2nd.

/

The pressure on modern healthcare is mounting. Hospitals face rising patient numbers while the available workforce continues to shrink. The consequences are visible every day: exhausted clinicians, longer hospital stays, more medical errors, and higher mortality rates. Nowhere is this pressure felt more urgently than in the intensive care unit. Within his PhD research Lars van de Kamp explored how technological innovation鈥攑articularly in mechanical ventilation鈥攃an help lighten the workload on ICU teams while improving outcomes for patients. The central message is clear from the start: by advancing monitoring, decision-support, and automation in ventilation systems, we move closer to safer, more efficient, and more personalized critical care.

The ICU is a unique clinical environment where every second counts. Patients rely on continuous expert attention, and clinicians must constantly interpret complex, rapidly changing data. In this setting, workforce shortages have an amplified impact, increasing stress for staff and risk for patients. Mechanical ventilation is one of the most demanding responsibilities in the ICU. Clinicians must make frequent and precise adjustments, ensuring the machine supports a patient鈥檚 breathing as their condition evolves. This level of vigilance requires experience, accuracy, and time鈥攔esources that are increasingly scarce.

Untapped potential in ventilator data

Modern ventilators already collect enormous amounts of sensor data, yet most of this information remains unused. The reason is simple: interpreting raw data streams in real time is nearly impossible without technological support. This challenge opens the door to data-driven innovation. By developing artificial intelligence and other advanced algorithms, researchers aim to translate complex data into meaningful insights. These systems help clinicians better understand a patient鈥檚 respiratory condition and adapt ventilator settings more precisely to individual needs. Importantly, the focus is not just on intelligence but on safety and interpretability鈥攅nsuring that clinicians understand and trust the technology supporting their decisions.

Better monitoring as the first step to automation

Improved monitoring capabilities form the foundation for future automation. When ventilators can reliably translate patient data into accurate, understandable feedback, clinicians are empowered to make better decisions with less effort. This reduces workload while enhancing patient comfort and safety. At the same time, progress is being made in the actual control of pressure and airflow delivered by ventilators. Tailoring these parameters to each patient鈥檚 physiology is essential for comfortable and effective respiratory support, and advanced control methods bring us closer to fully individualized care.

/

Closing the loop in mechanical ventilation

In this research Lars van de Kamp demonstrates how clinical data can be transformed into actionable tools that support both caregivers and patients. By integrating novel monitoring and control strategies, mechanical ventilation systems become smarter, more responsive, and easier to manage. This brings us significantly closer to 'closing the loop'鈥攁n automated system that continuously adjusts ventilation based on real-time patient needs. Such technology directly addresses a pressing societal challenge: the high workload in intensive care units. By reducing the effort required to manage ventilators, clinicians gain more time and space to focus on what matters most: delivering high-quality care.

His research ultimately shows that thoughtfully applied technology does more than lighten the burden on ICU staff鈥攊t also leads to safer treatments and better outcomes for patients.

 

Title of PhD thesis: Supervisors: Prof. Nathan van de Wouw, Prof. Tom Oomen and Bram Hunnekens.

Media Contact

Rianne Sanders
(Communications Advisor ME/EE)