Social network analysis for medical narcotics in South Korea: focusing on patients and healthcare organizations.

BACKGROUND: Medical narcotics must be administered under medical supervision because of their potential for misuse and abuse, leading to more dangerous and addictive substances. The control of medical narcotics requires close monitoring to ensure that they remain safe and effective. This study proposes a methodology that can effectively identify the overprescription of medical narcotics in hospitals and patients. METHODS: Social network analysis (SNA) was applied to prescription networks for medical narcotics. Prescription data were obtained from the Narcotics Information Management System in South Korea, which contains all data on narcotic usage nationwide. Two-mode networks comprising hospitals and patients were constructed based on prescription data from 2019 to 2021 for the three most significant narcotics: appetite suppressants, zolpidem, and propofol. Two-mode networks were then converted into one-mode networks for hospitals. Network structures and characteristics were analyzed to identify hospitals suspected of overprescribing. RESULTS: The SNA identified hospitals that overprescribed medical narcotics. Patients suspected of experiencing narcotic addiction seek treatment in such hospitals. The structure of the network was different for the three narcotics. While appetite suppressants and propofol networks had a more centralized structure, zolpidem networks showed a less centralized but more fragmented structure. During the analysis, two types of hospitals caught our attention: one with a high degree, meaning that potential abusers have frequently visited the hospital, and the other with a high weighted degree, meaning that the hospital may overprescribe. For appetite suppressants, these two types of hospitals matched 84.6%, compared with 30.0% for propofol. In all three narcotics, clinics accounted for the largest share of the network. Patients using appetite suppressants were most likely to visit multiple locations, whereas those using zolpidem and propofol tended to form communities around their neighborhoods. CONCLUSIONS: The significance of this study lies in its analysis of nationwide narcotic use reports and the differences observed across different types of narcotics. The social network structure between hospitals and patients varies depending on the composition of the medical narcotics. Therefore, these characteristics should be considered when controlling medication with narcotics. The results of this study provide guidelines for controlling narcotic use in other countries.

Narcotics information management system in South Korea: system development and innovation.

BACKGROUND: As the misuse and abuse of medical narcotics are increasing in South Korea, an information system for the integrated information management of medical narcotic drugs across the nation is needed. This paper presents the development process of the Narcotics Information Management System (NIMS) for the monitoring of medical narcotics usage and the results of its implementation. METHODS: As the NIMS enforces that all narcotics handlers digitally report all information on handling medical narcotic drugs, the functional requirements of the NIMS have been identified in accordance with the Narcotics Control Act. In addition to the functional requirements, the non-functional requirements of the NIMS have been elicited by major narcotics handlers and their associations. The non-functional requirements include privacy, availability, connectivity, interoperability, and data integrity. The system design with entity-relationship diagrams and its implementation processes have been presented. RESULTS: The NIMS encompasses all narcotic handlers, which comprise exporting, importing, and pharmaceutical companies; wholesalers; hospitals and clinics; and pharmacies, collecting over 120 million cases annually. It enables transparent monitoring throughout the life cycle, from manufacturing, sales, purchase, and disposal of narcotics. As a result, the number of prescriptions for medical narcotics has been reduced by 9.2%. CONCLUSIONS: To the best of our knowledge, the NIMS is the world's first system to manage all information on the total life cycle of medical narcotics, including imports, production, distribution, use, and disposal of drugs. This system has enabled the safety management and monitoring of medical narcotic drugs. Additionally, it provides consistent and transparent information to physicians and patients, leading to the autonomous safety management of narcotics. The successful development of the NIMS can provide guidelines for implementing a narcotics management system in other countries.

Evaluation of e-learners' concentration using recurrent neural networks.

Recently, interest in e-learning has increased rapidly owing to the lockdowns imposed by COVID-19. A major disadvantage of e-learning is the difficulty in maintaining concentration because of the limited interaction between teachers and students. The objective of this paper is to develop a methodology to predict e-learners' concentration by applying recurrent neural network models to eye gaze and facial landmark data extracted from e-learners' video data. One hundred eighty-four video data of ninety-two e-learners were obtained, and their frame data were extracted using the OpenFace 2.0 toolkit. Recurrent neural networks, long short-term memory, and gated recurrent units were utilized to predict the concentration of e-learners. A set of comparative experiments was conducted. As a result, gated recurrent units exhibited the best performance. The main contribution of this paper is to present a methodology to predict e-learners' concentration in a natural e-learning environment.

Fast Outlier Detection Using a Grid-Based Algorithm.

As one of data mining techniques, outlier detection aims to discover outlying observations that deviate substantially from the reminder of the data. Recently, the Local Outlier Factor (LOF) algorithm has been successfully applied to outlier detection. However, due to the computational complexity of the LOF algorithm, its application to large data with high dimension has been limited. The aim of this paper is to propose grid-based algorithm that reduces the computation time required by the LOF algorithm to determine the k-nearest neighbors. The algorithm divides the data spaces in to a smaller number of regions, called as a "grid", and calculates the LOF value of each grid. To examine the effectiveness of the proposed method, several experiments incorporating different parameters were conducted. The proposed method demonstrated a significant computation time reduction with predictable and acceptable trade-off errors. Then, the proposed methodology was successfully applied to real database transaction logs of Korea Atomic Energy Research Institute. As a result, we show that for a very large dataset, the grid-LOF can be considered as an acceptable approximation for the original LOF. Moreover, it can also be effectively used for real-time outlier detection.