Interactive pain nursing intervention system for smart health service

In modern society, the amount of information has significantly increased due to the development of BT-IT convergence technology. This leads to developing information obtaining and searching technologies from much data. Although system integration for medicare has been largely established to accumulate large amounts of information, there is a lack of provision and support of information for nursing activities, using such an established database. In particular, the judgment for pain intervention depends on the experience of individual nurses, leading to usually making subjective decisions. Thus, there is some danger in applying unwanted anesthesia and drug abuse. In this paper, we proposed the interactive pain nursing intervention system for smart health service. The proposed method uses collaborative filtering that extracts some pain strengths, which represent a high relative level, based on similar pain strengths. Pain strength estimation method using collaborative filtering calculates patient similarities through Pearson correlation coefficients in which a neighbor selection method is used based on the pain strength. In general, medical data in patients shows various distributions due to its own characteristics, as sample data demonstrates. Therefore, this is determined as an applicable theory to the sparsity problem. In addition, it is compensated using a default voting method. The medical data evaluated by applying standard data and its accuracy in pain prediction is verified. The test of the proposed method yielded excellent extraction results; it is possible to provide the fundamental data and guideline to nurses for recognizing the pain of patients based on the results of this study. This represents increased patient welfare for smart health services.