“ Predicting possible complications and delivering best supportive care is extremely important to improve an overall ICU survival rate. NTT DATA Group developed a smart ICU system, called ehCOS SmartICU, which collects and analyzes several vital signs of ICU patients simultaneously by using Artificial Intelligence (AI). ehCOS SmartICU sends doctors and nurses an early alert of which patients are at imminent risk. ”
The Intensive Care Unit (ICU) is a specialized facility dedicated to seriously ill patients who require intensive monitoring, nursing care and complex respiratory support. Medical studies have reported that early medical intervention before the occurrence of complications leads to a more effective recovery. To enable this, NTT DATA and the everis Group (“everis” hereafter), an NTT DATA affiliate in Spain, developed ehCOS SmartICU. This solution predicts the risks of complications occurring to the patient based on the accumulated chronological data of the patient’s vital signs and reports the risk scores and information necessary for doctors or nurses to diagnose. Virgen del Rocío University Hospital in Spain is the first hospital to verify the effectiveness of the system in assisting clinical decision-making. Officials at the hospital believe that detecting risks of complications just two hours prior to the occurrence will facilitate more appropriate treatment, shortening patient recovery time. Specific complications that this solution is targeting are : (1) septic shock, (2) episode of rapid drop in blood pressure, and (3) hypoxemia, all of which may be fatal to ICU patients. AI is being used to build a system that predicts risks for these complications. An AI prediction model is being built based on public ICU data called MIMIC II*1. The patients’ data will then be processed by distributed streams on NTT DATA’s OSS big data infrastructure to enable a real-time prediction of risks for complications. *1 MIMIC II is a database on ICU patient treatments. It anonymizes data accumulated at the Beth Israel Deaconess Medical Center in the US.