Call for Papers -- Predictive AI Models for Patient Readmission Risk and Resource Allocation

According to medical literature, post-discharge surveillance can prevent or reduce a significant number of readmissions. Anticipating when a readmission might occur and using that information to effectively monitor the patient for problems that could lead to a readmission are crucial components of a well-thought-out monitoring plan. In order to satisfy monitoring needs, this research offers novel techniques to statistically generate a personalized estimate of the time to readmission density distribution. The density function is then used to optimize a post-discharge observation schedule and staffing plan. The quantity of this data varies greatly between patients. A team with a fair resource utilization rate ensures that the project advances overall by allocating time to the right tasks. Resource utilization is especially important if your employees are working on multiple projects because it gives you a more complete picture of their workload.

Artificial intelligence (AI) systems that incorporate these factors can precisely forecast the probability of readmission, allowing medical professionals to take appropriate action and treat high-risk patients promptly. Thus, AI approaches hold great potential for improving patient care and preventive interventions in hospital readmission prediction. While there are obstacles to overcome, such as the availability of data and the interpretability of models, cooperative efforts and developments in AI technology can do so. Healthcare systems could effectively lower readmission rates, optimize resource allocation, and improve patient outcomes by utilizing AI. Planning becomes more difficult when additional hospitals are required to provide support. In order to assist healthcare administrators in organizing their finances and determining the most effective strategy for resource allocation and sharing, there are several data-driven methodologies that offer data-driven indicators. It is insufficient for traditional decision-making techniques to suggest managers implement validated policies. The application of machine learning and data intelligence to predictive modeling can be a helpful decision-support tool for identifying patients are at high risk of readmission. Nevertheless, there are a number of issues with the current systems, including the inability to use unstructured data and the inability to combine data from different hospitals.

Models that are predictive are used by hospital decision-makers to proactively control the discharged patient's risk of readmission. It is apparent that it is difficult to incorporate predictions of the development of risk from time-to-event algorithms into decision-making procedures, but predictions from categorization models are simple to incorporate. The objective of the special issue was to develop and evaluate algorithms that employ machine learning tools to generate features from time-related information autonomously in order to predict readmissions to hospitals. With a real-time awareness of assignments, processes, and resource allocation, you can start scheduling people for the mission with utilization understanding and avoid employee fatigue and overutilization.

Papers are invited that consider, but are not limited to, the following themes:

• Reducing readmissions to hospitals by combining resource allocation with empirical prediction
• A thorough evaluation of risk prediction algorithms for hospital readmission
• Strengthening the control of hospital readmissions through better decision-making
• Maximizing decisions regarding patient readmissions during intensive care unit discharge
• Intelligent systems and machine learning: possibilities in healthcare epidemiology
• A methodology for measuring and mitigating hospital readmission fines
• A conceptual structure for studying the likelihood of readmission
• Applying a Program to Reduce Neurosurgical Readmissions in a Scientific Hospital
• Machine learning for predicting pediatric hospital readmissions
• Optimizing readmission costs for hospitals with latent feature ensemble learning
• Optimizing the flow of hospital critical care units while minimizing patient readmissions

 

Guest Editors:
Prof. Abdul Kadar Muhammad Masum, Department of Software Engineering, Faculty of Science and Information Technology, Daffodil International University, Birulia, Savar, Dhaka, Bangladesh
Dr. Md Zia Uddin, Sustainable Communication Technologies Department, SINTEF Digital, Oslo, Norway
Prof. Md Abdus Samad, Department of Information and Communication Engineering, Yeungnam University, Gyeongsan, South Korea

 

Tentative Timeline:
Deadline for manuscript submissions: December 31, 2024
Expected publication date will be based on Journal decision.