Decision Support for Healthcare Processes and Services
Permanent URI for this collectionhttps://hdl.handle.net/10125/107480
Browse
Recent Submissions
Item type: Item , Towards Human-AI Interaction in Medical Emergency Call Handling(2024-01-03) Maletzki, Carsten; Elsenbast, Christian; Reuter-Oppermann, MelanieCall-takers in emergency medical dispatch centers typically rely on decision-support systems that help to structure emergency call dialogues and propose appropriate responses. Current research investigates whether such systems should follow a hybrid intelligent approach, which requires their extension with interfaces and mechanisms to enable an interaction between call-takers and artificial intelligence (AI). Yet unclear is how these interfaces and mechanisms should be designed to foster call handling performances while making efficient use of call-taker's often strained mental capacities. This paper moves towards closing this gap by 1) deriving required artifacts for human-AI interaction and 2) proposing an iterative procedure for their design and evaluation. For 1), we apply the guidelines for human-AI interaction and conduct workshops with domain experts. For 2), we argue that performing a full evaluation of the artifacts is too extensive at earlier iterations of the design process, and therefore propose to enact use-case-driven lightweight evaluations instead.Item type: Item , Capacity Balancing Algorithm for Placement of Observation Patients: Assessing the Effect on ED Crowding(2024-01-03) Tilson, Vera; Xiao, Shiya; Pachamanova, Dessislava; Dwyer-Matzky, KeelyWe examine the queueing performance of a real-time capacity-balancing (CB) algorithm for routing patients from the Emergency Department (ED) to an Observation Unit (OU) or to an inpatient ward. We study the significance of relative bed capacities between the ward and the OU. OUs are designed for specific patient types, with shorter expected hospital stays. By redirecting patients and reserving OU capacity for those who benefit most, overall hospital stay durations are reduced. However, high ward utilization may cause longer ED waiting times. To prevent ED crowding, implementing a CB algorithm requires careful analysis of ward and OU long-run utilization.Item type: Item , Optimizing Vaccine Delivery with Drones for Hard-to-Reach Regions(2024-01-03) Campbell, James; Enayati , Shakiba; Kara, Bahar; Peker, Meltem; Li, Haitao; Akenroye, TemiThis research optimizes the use of drones, alongside other transport modes, for delivery of routine childhood vaccines subject to cold chain requirements. We focus on the value of drones to improve vaccine deliveries for hard-to-reach regions. This paper first briefly describes optimization of country-level vaccine distribution from national depots to local health zone distributions centers (DCs) using both large and small drones, along with boats, trucks and planes. Then we focus on research on optimizing local vaccine delivery within one health zone, from the DC to remote aid posts, using small drones, along with walking, boats and trucks. Results using data for the island nation of Vanuatu show that drones can be very useful for vaccine delivery to replace current transportation options, and to resupply health workers with fresh vaccines at remote sites to allow more efficient health worker outreach trips.Item type: Item , From Concept to Commercialization: A Double Co-Design Approach(2024-01-03) Ulapane, Nalika; Forkan, Abdur Rahim Mohammad; Jayaraman, Prem Prakash; Schofield, Penelope; Burbury, Kate; Wickramasinghe, NilminiA Smartphone-based clinical decision support system (CDSS) has been developed to assist the management of perioperative patients in cancer care. The design process followed a systematic approach guided by design science research methodology (DSRM) and the theory of task technology fit. Our previous work discussed our progress up to the Assessment phase of the CDSS. In this paper, we report on the extension of our codesign work towards commercialization of the CDSS partnering with a commercialization partner. This partnership complemented as a second phase of codesign making our design process more rigorous. Hence an extension to design science research was identified—we name it a double codesign approach. This double codesign approach benefits in making artifacts more rigorous and fit for commercialization. In addition to reporting on our double codesign approach, we also present the different data elements we were able to capture through this CDSS. We discuss the possibilities and barriers we encountered and group the data elements into categories that can be generalized to all CDSSs.Item type: Item , Semantic-Level New Information Identification in Electronic Health Records Using Text-Mining Techniques(2024-01-03) Hu, Ya-Han; Tseng, Hsiao-Ting; Huang, Chun-FengElectronic health records (EHRs) are widely used in healthcare systems to store and transmit patients’ health records. They have many advantages, such as saving space, increasing efficiency, and facilitating communication. However, they also have a major drawback: information redundancy. Healthcare professionals often use copy and paste to write clinical notes, which leads to excessive similarity and low diversity in EHRs. This impairs the readability and quality of EHRs and hinders decision making. To address this problem, this study proposes a text-mining approach to identify new information at semantic-level in EHRs. Unlike previous studies that focused on word-level identification, we use concept occurrence and concept similarity score methods to annotate new information at semantic-level and evaluate them with gold standards. The experimental evaluation demonstrates that the method proposed in this study achieves an F1-score ranging from 78.57 to 80.31 under various parameter combinations. The proposed method enables healthcare professionals to read EHRs more efficiently and make more informed decisions.Item type: Item , Explainable AI in healthcare: Factors influencing medical practitioners’ trust calibration in collaborative tasks(2024-01-03) Darvish, Mahdieh; Holst, Jan-Hendrik; Bick, MarkusArtificial intelligence is transforming clinical decision-making processes by using patient data for improved diagnosis and treatment. However, the increasing black box nature of AI systems presents comprehension challenges for users. To ensure the safe and efficient utilisation of these systems, it is essential to establish appropriate levels of trust. Accordingly, this study aims to answer the following research question: What factors influence medical practitioners' trust calibration in their interactions with AI-based clinical decision support systems (CDSSs)? Applying an exploratory approach, the data is collected through semi-structured interviews with medical and AI experts, and is examined through qualitative content analysis. The results indicate that perceived understandability, technical competence and reliability of the system, along with other userand context-related factors, impact physicians’ trust calibration in AI-based CDSSs. As there is limited literature on this specific topic, our findings provide a foundation for future studies aiming to delve deeper into this field.Item type: Item , Explorative Study on the Utilization of Patient Pathways in Finnish Public Healthcare(2024-01-03) Eklund, Amanda; Vesinurm, Märt; Torkki, PaulusIn this study, we explore the perception and usage of patient pathways among professionals in the Finnish public healthcare system, which is currently coping with the need for increased efficiency and productivity. Twenty-three healthcare professionals participated in a survey that combined open-ended and standardized questions. The findings reveal a positive view of patient pathways, with over 90% reporting usage in their work due to the structure and coordination among professionals that the pathways offer. However, some difficulties are highlighted, such as locating updated pathways and experiencing inflexibility in their usage. We recommend focusing on enhancing the accessibility of patient pathways and ensuring that they remain updated for their utilization. Further monitoring of pathway usage is required, and a cooperative approach could help overcome barriers to the pathways' implementation. Centralizing patient pathway information in a single digital environment could prevent the accumulation of underutilized tacit knowledge.Item type: Item , Shifting psychometric bipolar scales data towards the normal distribution(2024-01-03) Lehmann, Rene; Vogt, BodoBipolar Likert scales are commonly used in psychometrics. Improved psychometric profiling can help to reduce costs, optimize resource usage, increase patient welfare and reduce mental health risks. In health economics grant funding depends on quality-adjusted life years (QALY) index values associated with the effect size of a therapeutic intervention. Increasing the statistical power corresponds to increasing effect sizes and, thus, increased grant funding and incentives. Recently, the compositional structure (i.e., the Simplex) of bipolar scales data was revealed. While the isometric log-ratio (ilr) transformation converts compositional data towards the interval scale the central limit theorem of statistics (CLT) postulates that sample means of ilr transformed and means of untransformed item response data, both, are approximately normally distributed. The larger the convergence towards normality the more reliable are the results of procedures based on approximate normal distribution, e.g., correlation analyses and partial least squares path modeling. Via simulation we show that the null-hypothesis of normality is rejected less often when using means of ilr transformed item responses. That is, the ilr transformation causes a shift towards normality. As a result, the statistical power of procedures based on approximate normal distribution increases.Item type: Item , Longitudinal healthcare analytics for early detection and progression of neurological diseases: A clinical decision support system.(2024-01-03) Owusu, Gabriel; Wang, Xuan; Sun, JunNeurological diseases, including Alzheimer's disease (AD), are rising global health challenges. This study presents a two-stage decision support system (DSS) that uses machine learning and neuroimaging for early AD detection and monitoring. The first stage uses deep learning for predicting AD likelihood. The second leverages a 3D convolutional neural network to identify crucial brain regions in AD progression. Notably, the DSS offers a solution to machine learning's "black box" problem using an occlusion map explainability method, enhancing decision transparency. Its design is adaptable to other diseases using imaging data, underscoring its broad healthcare potential. By providing an innovative and interpretable tool for improved disease management, this research helps foster better patient care and outcomes.Item type: Item , Boosting Multi-Professional Collaboration in Palliative Care Through Digital Technologies: A Work System Analysis(2024-01-03) Wöhl, Moritz; Gimpel, HennerThe success of palliative care relies on the collaboration of different professions, which are all part of one multi-professional work system. Varying perspectives and expertise of the team members characterize the joint work, resulting in divergent interpretations and relevancies concerning the treatment of patients and requiring a continuous exchange of knowledge and information. Using the work system (transformation) method, we analyze the current work system in palliative care and propose improvements. Based on extensive on-site observations and discussions with practitioners, we outline the challenges of multi-professional collaboration in palliative care and identify opportunities to tackle them (with digital technologies). The identified “to be” work system should help promote multi-professional collaboration in more contexts than just palliative care, as they aim to foster collaboration within multi-professional teams.Item type: Item , Why Do Acquisitions Negatively Affect Patient Outcomes at Target Hospitals: Quiet Life Hypothesis or Disruptions Caused by Acquisition Integrations?(2024-01-03) Tanriverdi, Huseyin; Yang, Xiaoxuan; Wen, YilinCare quality declines at target hospitals following mergers and acquisitions (M&A) by multihospital health systems (MHSs). The declines have been attributed to the M&A’s consolidation and competitive intensity reduction effects. Insulated from competition, managers may choose to enjoy the “quiet life” instead of taking on difficult tasks to improve care quality. In addition to subjecting this hypothesis to empirical scrutiny, we propose and test a disruption hypothesis: IT M&A integrations and medical service integrations between an MHS and a target could cause disruptions to the target’s care processes and reduce the quality of patient outcomes. We find support for both hypotheses in a sample of 629 M&A transactions conducted by 179 unique MHSs and 579 unique target hospitals in the U.S. hospital industry during 2009-2017. Reduced competitive intensity increases mortality rates. IT M&A integrations increase readmission and mortality rates whereas service integrations increase only readmission rates in target hospitals.Item type: Item , Introduction to the Minitrack on Decision Support for Healthcare Processes and Services(2024-01-03) Reuter-Oppermann, Melanie; Walker, Cameron; Furian, Nikolaus