LOSH Prediction using Data Mining

Abstract

Only when resources and time of the hospital is managed, the productivity of the Hospital services enhances. Both time and resource consumptions are at its peak when patient is admitted to the hospital. So, they can best be managed at this time of stay. Also, managing the emergency cases as they arrive should also be taken care. These factors can be managed by estimating the future resource requirements of the hospital. The rate at which resources are consumed is to be determined. Hence, if the LOSH (Length Of Stay at the Hospital) of the patient is determined, we can easily manage the resources and emergency admissions. Hence, to derive the stay duration of the patient in the hospital is an important operation. This paper proposes a prediction model that predicts the length of stay of the patient in the hospital and a solution to handle emergency situations when doctor is unavailable.We used basic clustering methods like DBSCAN (Density Based Spatial Clustering Application Network) and KApriori. In addition, we compared the execution time of the both.