No abstract

No abstract

Background Unplanned readmission of a hospitalized patient is an indicator of patients’ exposure to risk and an avoidable waste of medical resources. In addition to hospital readmission, intensive care unit (ICU) readmission brings further financial risk, along with morbidity and mortality risks. Identification of high-risk patients who are likely to be readmitted can provide significant benefits for both patients and medical providers. The emergence of machine learning solutions to detect hidden patterns in complex, multi-dimensional datasets provides unparalleled opportunities for developing an efficient discharge decision-making support system for physicians and ICU specialists. Methods and findings We used supervised machine learning approaches for ICU readmission prediction. We used machine learning methods on comprehensive, longitudinal clinical data from the MIMIC-III to predict the ICU readmission of patients within 30 days of their discharge. We incorporate multiple types of features including chart events, demographic, and ICD-9 embeddings. We have utilized recent machine learning techniques such as Recurrent Neural Networks (RNN) with Long Short-Term Memory (LSTM), by this we have been able to incorporate the multivariate features of EHRs and capture sudden fluctuations in chart event features (e.g. glucose and heart rate). We show that our LSTM-based solution can better capture high volatility and unstable status in ICU patients, an important factor in ICU readmission. Our machine learning models identify ICU readmissions at a higher sensitivity rate of 0.742 (95% CI, 0.718–0.766) and an improved Area Under the Curve of 0.791 (95% CI, 0.782–0.800) compared with traditional methods. We perform in-depth deep learning performance analysis, as well as the analysis of each feature contribution to the predictive model. Conclusion Our manuscript highlights the ability of machine learning models to improve our ICU decision-making accuracy and is a real-world example of precision medicine in hospitals. These data-driven solutions hold the potential for substantial clinical impact by augmenting clinical decision-making for physicians and ICU specialists. We anticipate that machine learning models will improve patient counseling, hospital administration, allocation of healthcare resources and ultimately individualized clinical care.

Based on the concept of task/technology fit, a research framework and exploratory case study are presented that assess success factors and impacts of mobile business applications. Preliminary empirical evidence for the applicability of the framework was obtained for a mobile electronic procurement system implemented at a Fortune 100 company. For different user groups, the relationships between the characteristics of technology and tasks, usage, and organizational impacts were analyzed. The results indicate a need for simple but highly functional mobile applications that complement existing information systems. The study provides a basis for further research to improve the design and management of business applications based on emerging technologies.

Mobile information systems hold great promise to support organizational processes. Clear guidelines however, of how to design effective mobile information systems in support of organizational processes have not been developed. Based on earlier research studies that emphasized the importance of a fit between organizational tasks and technology (Goodhue and Thompson 1995), and that developed a systematic fit profile for one particular task−technology combination, namely group support systems to support group tasks (Zigurs and Buckland 1998), this research paper seeks to develop a fit profile for mobile information systems to support managerial tasks. We suggest to determine task−technology fit as a three−way match between managerial tasks (operationalized by non−routineness, interdependence and time−criticality), mobile information systems (operationalized by functionality, user interface, and adaptability), and the mobile use context (operationalized by distraction, quality of network connection, previous experience, and mobility). The analysis shows that use situations characterized by high distraction and poor quality of network connection are particularly challenging for the design of mobile information systems, and that the user interface requires particular attention. The proposed framework provides guidelines for the design of effective mobile information systems and for future research studies. AbstractMobile information systems (IS) hold great promise to support organizational processes. Clear guidelines, however, of how to design effective mobile IS in support of organizational processes have not been developed. Based on earlier research that emphasizes the importance of fit between organizational tasks and technology and that develops fit profiles for specific task-technology combinations, this paper develops a task-technology fit (TTF) profile for mobile IS to support managerial tasks. We suggest a three-way match between dimensions of managerial tasks, mobile IS, and the mobile use context. We find that use situations characterized by high distraction and poor quality of network connection are particularly challenging for the design of mobile IS, and that the user interface requires particular attention. The proposed conceptual model of task-technology fit provides guidelines for the design of effective mobile IS and for future research studies.