Decision Support for Healthcare Processes and Services
Permanent URI for this collection
Browse
Recent Submissions
Item Towards a Hospital-wide Simulation Framework(2025-01-07) Reuter-Oppermann, MelanieIn a hospital, there are many dependencies between different resources, processes, and departments. When changes are made to part, it is often very difficult to estimate the consequences for the rest of the hospital and it is unclear how far they stretch. While there is a tendency towards more integrated planning in hospitals, analysing the effects and benefits, especially for the whole hospital, is difficult. The aim of our research project is to design and build a simulation framework for a complete hospital including all departments, processes, and resources that allows for a hospital-wide analysis. By using a container-based approach, the integration of various simulation models for different departments and their interactions is possible. A simulation-wide event calendar and clock align the events and makes sure that the interaction between the individual sub-models is well-coordinated and stable. A first case study targets Tauranga Hospital in New Zealand.Item Supporting ICU Admission Scheduling under Uncertainty(2025-01-07) Witteborg, Arne; Borgstedt, Rainer; Rehberg, Sebastian; Wetzel, Daniel; Römer, MichaelAdmission scheduling in an intensive care unit (ICU) poses a complex challenge that requires balancing the optimization of resources for the post-operative care of patients while managing the risk of capacity overload. This process is further complicated by uncertainties such as the unknown arrival of emergencies and the varying length-of-stay of each patient. We model this as a stochastic optimization problem with a planning horizon of one week, using Monte-Carlo simulation to approximate the uncertain and scenario-dependent evolution of ICU occupation. Specifically, the model aims at minimizing the risk of exceeding critical occupancy levels, represented as chance constraints, while efficiently utilizing available resources. The application of this model on real-world data highlights the potential for optimizing resource utilization without increasing the risk of surpassing critical occupancy thresholds by providing admission schedules that are robust to different outcomes of the uncertainties.Item Discovering Careless Response Behavior in Psychometric Data(2025-01-07) Lehmann, Rene; Bengart, Paul; Vogt, BodoIn psychological healthcare considering bipolar Likert scales data as compositional data can enhance statistical validity. Applying an isometric log-ratio transformation yields interval scaled real-valued data. It increases the normal approximation of item response means, reduces statistical biases and enhances the statistical power of the Pearson correlation test and two-sample t-tests (paired and unpaired) affecting linear regression, partial least squares path modeling and moderator analysis. Mental overload, missing attention, faking or social desirability can corrupt a test person's answers in a psychometric survey. As a result, the corresponding questionnaire data are useless affecting subsequent analyses and interpretations. Aiming to detect careless response behavior as statistical outliers we compare the well-known Mahalanobis-distance to a multivariate projection pursuit method. Performing outlier detections with traditional and with isometric log-ratio transformed data we point out the superiority of the compositional data interpretation of psychometric bipolar scales data.Item Resilient Healthcare Systems through AI-Powered Cloud Computing: Perspectives on Resource Allocation(2025-01-07) Bokaba, Tebogo; Ndayizigamiye , Patrick; Idemudia, Efosa; Alexandre Junior, ArmindoThis study explores the impact of AI-powered cloud computing on the resilience of healthcare systems, focusing on the resources building block of the World Health Organization. The study adopts the interpretivism philosophy coupled with an inductive approach, and a combination of purposive and snowball sampling, to gather insights from 11 participants through interviews. The findings reveal that AI-powered cloud computing significantly improves resource distribution through advanced data analytics and predictive tools, facilitating smarter decision-making and optimizing operations. The study concludes that AI-powered cloud computing is essential for flexible, efficient service delivery and developing adaptable healthcare models. These insights are invaluable for policymakers and technology developers and highlight the need for integrating AI technologies into healthcare to foster their resilient systems. This research contributes to the discourse on healthcare resilience, underscoring the synergistic potential of technology and healthcare.Item Clinical Handover at Hospital Night Shift: Review of Challenges and Impacts on Patient Safety and Staff Experiences(2025-01-07) Madanian, Samaneh; Phyo Lwin, Phyo; Kwong, Charlie; Johnson, Kenneth; Techatassanasoontorn, Angsana; Singh, Harminder; Wong , WilliamThe aim of safe patient care in hospitals after hours at night has become increasingly elusive due to healthcare workers' exhaustion from long work hours, challenges of communication and coordination, and confusion from day to night handover. Among these challenges, clinical handover or the transfer of patient information and responsibility between healthcare workers has been among the most difficult issues to address. To develop a deeper understanding of challenges and implications of clinical handover, this systematic literature review examines clinical handovers during night shifts, highlighting critical issues such as inconsistent information and a lack of standard procedures, which leads to communication gaps and errors. Night shifts also increase stress and health problems for healthcare workers. Future studies should investigate the long-term effects of night shift handovers, the impact of training, and how technology can assist in improving these processes. Enhancing these areas will significantly boost patient safety and maintain workers' well-being.Item From Manual to Automated: A Multi-case Study of Utilizing Robotic Process Automation and Intelligent Automation in Healthcare Operational Processes(2025-01-07) Kaitosalmi, Jani; Ratia, Milla; Torkki, PaulusHealthcare organizations are applying digitalization to the constantly increasing amounts of different administrative processes. Automation of digital workflow processes using low-code solutions, such as Robotic Process Automation (RPA) and Intelligent Automation (IA), has emerged to increase process efficiency in healthcare. Although automation has the potential to offer significant efficiency benefits, it is still unclear what factors should be considered when implementing automation solutions. This research investigates the impact of automation in two Finnish healthcare districts through a comprehensive multi-case study. Findings reveal that while RPA and IA can significantly enhance process efficiency and reduce repetitive tasks, technical faults, process readiness, and unmet user expectations hinder their successful implementation. The study underscores the need to manage expectations, ensure process readiness, and foster open communication to realize the full potential of automation in healthcare.Item Deep-learning-based Detection of Food Hypersensitivity from Confocal Laser Endomicroscopy Images of the Gastro-intestinal Tract(2025-01-07) Hasan, Md Abid; Li, Frédéric; Tetzlaff-Lelleck, Vivian; Schmelter, Franziska; Ahlemann, Greta Marie; Jablonski, Lennart; Huang, Xinyu; Sina, Christian; Grzegorzek, MarcinFood hypersensitivity (FHS) is a relatively common pathological condition characterised by adverse reactions to specific foods, and that currently has limited diagnostic methods. Diagnostic tools using gastro-intestinal confocal laser endomicroscopy (CLE) images have recently been proposed for the assessment of FHS, but their interpretation can be challenging even for trained physicians. We propose to alleviate this problem by training machine learning models on CLE images of the gastro-intestinal tract for the binary classification problem of recognising images that show an adverse reaction to food intake. More specifically, the performances of four state-of-the-art image classification models (VGG16, Inception-v3, Xception, MobileNet-v2) are compared on one dataset acquired from 38 patients with proven FHS. Additionally, the models decisions are interpreted using the Grad-CAM technique. Our study shows that although all four models achieve satisfying classification performances, they learn very different features in terms of interpretability from the clinical perspective.Item Introduction to the Minitrack on Decision Support for Healthcare Processes and Services(2025-01-07) Furian, Nikolaus; Walker, Cameron; Reuter-Oppermann, Melanie