Due to the overwhelming number of patients and amount of data, physicians risk overlooking and neglecting important information from the patients’ vitals. It is an ongoing struggle and a huge task to monitor patients’ status in a hospital setting, and infant patients bring additional difficulties. Unlike older patients, infants are unable to effectively communicate their feelings of pain or uncomfortableness during medical episodes and treatment. The lack of communication leads to the possibility of a great number of severe or sudden medical episodes going unnoticed. It is the responsibility of the patient care team, led by the physician, and they must be able to keep up to date on the infant’s current medical data. This data is oft collected at intervals of several hours, which allows for the possibility of drastic changes in the infant’s vitals within these periods that may be overlooked. For example, in infants who experience SIDS (Sudden Infant Death Syndrome) first suffer through apnea, low heart rate, or low oxygen concentration in the blood. These three occurrences could be used to warn when SIDS will occur. The patient care team needs to collect data more often, even continuously, to discover these incidents prior to the SIDS. Once data is collected, the physician needs to interpret said data, and make appropriate diagnoses and predictions in a timely manner. Thus, the barriers for infant patients are lack of patient communication, infrequent data collection, and time-consuming diagnosing and prediction.
Wearable devices could be the key to monitoring these infant patients better. Such devices would continuously measure and record vital data, specifically breathing rates, heart rates, oxygen concentration in the blood, and body temperature. This would be cost effective and would free up a physician or a nurse to complete more important work. This constantly updated information would drastically decrease the possibility of a medical episode going unnoticed due to gaps in data collection. After the data is collected, the device could store the data where it would be accessible to machine learning or artificial intelligence algorithms that can predict if a dangerous medical episode is likely to occur. For example, it is known that some tachycardia in infants can be attributed to a rise in body temperature (Hanna, C M, and D S Greenes. “How Much Tachycardia in Infants Can Be Attributed to Fever?” Annals of Emergency Medicine., U.S. National Library of Medicine, June 2004). The algorithms in this device could learn how much of the heart rate is not caused by the body temperature and is an indication of something more serious. By having such algorithms, this would eliminate the need for a physician to immediately read and interpret the data. The algorithm could then notify and send reports to physicians and let them know the possibility or probability of a dangerous medical incident. It would still fall to the physicians to determine the severity of an incident and to them to prescribe an appropriate treatment.

 

DECISION SUPPORT & HOSPITAL MONITORING

Author: James Wimberley

Coauthor(s): James D Wimberley

Status: Project Concept