SEOUL, South Korea, Sept. 18, 2023 /PRNewswire/ -- VUNO Inc., South Korean medical AI company announced that its AI-based cardiac arrest risk management system, VUNO Med^®-DeepCARS™, has proven its clinical validity for the first time through a prospective study utilizing Real World Data(RWD).

VUNO announced that the multicenter clinical study paper, which confirms the clinical validity of VUNO Med^®-DeepCARS™ through its initial prospective research, has been published in "Critical Care," a leading international journal in intensive care medicine.

VUNO Med^®-DeepCARS™ is a medical AI device that analyzes four vital signs: respiratory rate, body temperature, blood pressure, and heart rate to predict cardiac arrest within 24 hours in general ward patients. It's Korea's first cardiac arrest prediction AI medical device and is recognized as a breakthrough device by the Korean Ministry of Food and Drug Safety(MFDS).

In the study published in Critical Care, the VUNO research team conducted a multicenter study using patient data from general wards at four tertiary medical institutions: Seoul National University Hospital, Bundang Seoul National University Hospital, Inha University Hospital, and Dong-A University Hospital, each with varying sizes, locations, and medical environments.

A total of 55,083 patients admitted to general wards in all participating institutions over three months, comparing the in-hospital cardiac arrest (IHCA) prediction accuracy, unexpected intensive care unit transfer (UIT) prediction accuracy, and false alarm rate of VUNO Med^®-DeepCARS™ with existing high-risk patient prediction systems like NEWS (National Early Warning Score).

The study found that the predictive performance of VUNO Med^®-DeepCARS™ had an AUROC score of 0.869, outperforming traditional methods (NEWS 0.767, MEWS 0.756). The number of alarms per 1,000 beds was more than halved compared to the same sensitivity level, proving higher trustworthiness in the alarms leading to actual medical interventions. Furthermore, its effectiveness was apparent across varying patient ages, genders, and times of occurrence.

In this study, VUNO Med^®-DeepCARS™ displayed exceptional predictive abilities and adaptability in identifying high-risk patients, even in its first prospective research. It consistently performed at the same level as in previous retrospective studies published in renowned international journals.

Yeha Lee, CEO of VUNO said, "The significance of this research lies in its successful execution of the first multicenter prospective study with many variables and challenges, thereby reaffirming the consistent predictive performance and versatility of VUNO Med^®-DeepCARS™" and he said, adding "We will continue to enhance the product's credibility in clinical settings by conducting additional prospective research. We are committed to introducing it to more medical settings as essential care to ensure patient safety."