HOME Consortium Research
HOME Consortium projects focus on improving healthcare delivery and quality of care using research and tools that exist across various disciplines.
Additional workshops are designed to demonstrate and explore how academic and industry leaders can work together to improve healthcare delivery. These connections provide an opportunity for participants to network, form new collaborations, and influence the direction of the Center.
HOME Consortium Project List
Machine Learning Algorithm to Predict Intervertebral Disc Degeneration (IDD), University of Missouri Orthopedics
Lower back pain is the second leading cause of disability across the world, resulting in an annual economic impact of $100 – 200 Billion. Intervertebral disc (IVD), a vital component of the spinal cord, is known to be a common cause of chronic low back pain. Our goal is to provide a tool to aide in earlier detection of disc degeneration through use of predictive analytics. Models will be developed to predict risk (based on several factors such as age, gender, occupation and lifestyle habits) for disc degeneration as early as a patient’s first visit to a primary care provider for back pain thereby allowing for early monitoring and pro-active treatment for patients.
Predictive and Prescriptive Analytics for Inventory Management of Blood, Taiwan Blood Services Foundation (TBSF)
The demand uncertainty of blood combined with its very short shelf life has led to a significant wastage of the total blood collected from the donors. On the other hand, there is severe shortage of blood along the blood supply chain due to the very limited donor pool. This study proposes a two-phased approach, the first phase uses predictive analytic techniques to provide accurate blood demand forecast to maintain the balance between supply and demand, while the second phase proposes effective inventory policies for blood centers and hospitals to minimize outdating and shortage.
A Pre-Post Economic Cost Analysis Cost: A Pilot Study, University of Missouri Healthcare & A Multinational Healthcare Company
A tumor board, consisting of a group of physicians, fellows, residents and other healthcare providers with varying backgrounds, meet frequently and review cancer cases and share expertise. The objective of the board is to identify the best cancer treatment option given the patient-specific conditions. Despite the widespread advantage of this inter-disciplinary board gathering, scheduling such a meeting is tedious, given the busy schedule of physicians, resulting in delayed reviews. To overcome this challenge, in recent years, web-based tools are being developed to ensure faster communication between resources. A company that developed such a tumor board solution software was interested in evaluating the impact of its product. The cost and time effectiveness was analyzed using propensity score matching technique.
Analyzing Nurses’ Electronic Medical Record Documentation Patterns in an Intensive Care Unit
The primary objective of this study is to conduct a multi-level cognitive task analysis of nurse’s EMR work patterns by using the time study data and the nurses’ EMR log data. The study focuses on two specific aims: (1) Elicitation of EMR documentation patterns through EMR log data and conducts cognitive task analysis of these patterns and (2) Conduct the usability study to identify existing problems of the EMR system in ICUs.
Real-time Workflow and Cognitive Workload Analysis in an Intensive Care Unit
Measuring accurate health processes is crucial for improving the quality of healthcare service. However, most measuring methods of workflow, such as time-motion study and manual observation, are made in an obtrusive way. Hence, the primary objective of this study is to develop a new way to analyze nursing activities in an intensive care unit (ICU). A Near Field Electromagnetic Ranging (NFER) system was used to diagnose the care flows related to intensive care nursing. The study advanced our understanding of how the variable clinical status of patients influences the time and clinical processes in ICU.
Developing NGOMSL Model for Electronic Health Record Use in an Emergency Department
The primary objective of this project was to develop a new workload evaluation method for the EHR system. The findings from this study improved our understanding of how the time and workload related to managing EHR system influence clinical processes around care delivery with patients. The benefits of conducting this study were 1) reducing information overload caused by networked information technologies in the healthcare system, and 2) revealing the important form factors that contribute a heavy workload during the EHR documentation in an emergency department.
Visualizing Alzheimer’s Disease Progression in Low Dimensional Manifolds
In this project, we developed a new visualization tool, derived from manifold-based nonlinear dimension reduction of brain MRI features, to track AD progression over time. In specific, we investigated the locally linear embedding (LLE) using a dataset from Alzheimer’s Disease Neuroimaging Initiative (ADNI), which includes the longitudinal MRIs from 562 subjects. About 20% of them progressed to the next stage of dementia. Using only the baseline data of cognitively unimpaired (CU) and AD subjects, LLE reduces the feature dimension to two and a subject’s AD progression path (curve) can be plotted in this low dimensional LLE feature space. In addition, the likelihood of being categorized to AD is indicated by color. This LLE map is a new data visualization approach that can assist in tracking AD progression over time.