An AIoT enabled system for optimizing data retrieval in the intensive care unit evaluated in a randomized crossover...

Area of Science:

• Medical Informatics
• Artificial Intelligence
• Healthcare Technology

Background:

• Intensive care units (ICUs) present information overload challenges for healthcare providers (HCPs), potentially leading to cognitive fatigue and errors.
• Current hospital information systems (HIS) may not optimally support efficient and accurate data collection in critical care settings.

Purpose of the Study:

• To compare the efficiency and accuracy of data collection between an AIoT-enabled ICU command center (CC) and a traditional HIS.
• To evaluate the impact of an AIoT-enabled CC on data collection time and accuracy in the ICU.

Main Methods:

• A randomized crossover pilot trial involving 21 ICU-trained HCPs collecting data from critically ill patients (selected by APACHE II score).
• Data collection was performed using either a CC-HIS sequence or HIS-CC sequence.
• Analysis included data collection time, accuracy, and subgroup differences, with 184 complete data sets analyzed.

Main Results:

• The AIoT-enabled CC significantly reduced data collection time by 41.8% (mean reduction of 6.75 minutes per patient; 8.5 vs. 15.2 minutes, p < 0.0001).
• Data accuracy was improved by 2.07% with the CC (93.8% vs. 95.9%, p = 0.0002).
• The study assessed the effect of increased data volume on time and accuracy.

Conclusions:

• The AIoT-enabled ICU command center substantially enhances both the efficiency and accuracy of data collection in the ICU.
• Implementing AIoT solutions in ICUs can mitigate information overload and improve clinical decision-making.
• This technology offers a promising approach to optimize critical care data management.