Real-Time Data Mining Models for Predicting Length of Stay in Intensive Care Units

Abstract

Nowadays the efficiency of costs and resources planning in hospitals embody a critical role in the management of these units. Length Of Stay (LOS) is a good metric when the goal is to decrease costs and to optimize resources. In Intensive Care Units (ICU) optimization assumes even a greater importance derived from the high costs associated to inpatients. This study presents two data mining approaches to predict LOS in an ICU. The first approach considered the admission variables and some other physiologic variables collected during the first 24 hours of inpatient. The second approach considered admission data and supplementary clinical data of the patient (vital signs and laboratory results) collected in real-time. The results achieved in the first approach are very poor (accuracy of 73 %). However, when the prediction is made using the data collected in real-time, the results are very interesting (sensitivity of 96.104%). The models induced in second experiment are sensitive to the patient clinical situation and can predict LOS according to the monitored variables. Models for predicting LOS at admission are not suited to the ICU particularities. Alternatively, they should be induced in real-time, using online-learning and considering the most recent patient condition when the model is induced.

Pervasive Patient Timeline for Intensive Care Units

“…The purpose of the INTCare system is to help intensivists in the decision making process, by providing capabilities such as monitoring patients' conditions, predict clinical events, like organ failure [9], length of stay [10], readmission [11], among others, and issue alert messages when the patients being monitored have vital signs outside of the normal range [6]. In this way it is ensured that the intensivists have an easier job but also the patients have a better and safer care.…”

Real-Time Decision Support Using Data Mining to Predict Blood Pressure Critical Events in Intensive Medicine Patients

“…Online-learning, real-time data processing [14] and system interoperability are others features. Until now, INTCare allows predicting organ failure and patient outcome [7], SEPSIS [15], barotrauma [16], readmissions [5,17] and length of stay [6,18] in realtime and with high sensitivities rates. In the past a first study was performed in order to predict critical events.…”

“…Having in consideration this aspect arises INTCare. INTCare [2] is a Pervasive Intelligent Decision Support System (PIDSS) able to collect and process data in realtime in order to provide new knowledge [3][4][5][6][7] anywhere and anytime. This knowledge is achieved by means of Data Mining (DM) techniques.…”

Step Towards Predicting Patient Length of Stay in Intensive Care Units

“…Since hospitals are faced with very limited resources including beds to receive patients, specially during this time of the COVID-19 pandemic, the Length of Stay (LOS) is a crucial factor for a better planning and management of hospital resources. In this context, the prediction of a patient's LOS will help hospitals achieve many goals, such as better benchmark performance in terms of profitability and patient care efficiency, due to the fact that LOS is an essential measure of healthcare utilization and a determining factor in hospitalization costs [4,5].…”