ARCHES Funded Grant
The Jump Applied Research for Community Health through Engineering and Simulation (ARCHES) program is a partnership between OSF HealthCare, the University of Illinois Urbana-Champaign (UIUC) and the University of Illinois College of Medicine in Peoria (UICOMP).
The funding supports research involving clinicians, engineers and social scientists to rapidly develop technologies and devices that could revolutionize medical training and health care delivery.
In alignment with the strategic focus on improving patient and family care across key health care areas, the Jump ARCHES program has placed an enhanced emphasis on four critical strategic areas for development:
- OSF HealthCare Children’s Hospital of Illinois
- Neurologic and Neuro Surgical Services of the OSF HealthCare Illinois Neurologic Institute
- OSF HealthCare Cancer Institute
- OSF HealthCare Cardiovascular Institute
Enhanced focus areas should emphasize:
- Pediatric innovations: Development of tools and technologies for early diagnosis and intervention in pediatric developmental and chronic conditions.
- Chronic care management: Novel solutions for managing long-term pediatric illnesses, ensuring continuity of care beyond the hospital setting.
- Family-centered education: Interactive platforms and educational tools to engage families in the care process, improving adherence and outcomes.
- Neurologic and Neuro Surgical Services Projects should target:
- Advanced neuro-diagnostics: Leveraging artificial intelligence and machine learning for early detection and monitoring of neurological disorders.
- Innovative therapies: Development of minimally invasive or non-invasive treatments for neurological conditions.
- Surgical support technologies: Tools to improve pre-operative planning, intraoperative precision and post-operative recovery in neurosurgical patients.
- OSF Cancer Institute projects should address:
- Cancer disparities: Solutions aimed at reducing disparities in cancer care for underserved populations.
- Diagnostics and screening: Innovations in early detection, including biomarkers and imaging technologies.
- Survivorship care: Personalized approaches to post-treatment survivorship, including mental health support, rehabilitation, and lifestyle adjustments. OSF Cardiovascular Institute Projects should prioritize:
- Telehealth Solutions: Scalable telemedicine models for managing chronic cardiovascular conditions, especially in rural and underserved areas.
- Remote monitoring technologies: Wearable devices and mobile platforms for continuous monitoring and management of cardiovascular health.
- Preventive cardiovascular care: Strategies to enhance early intervention and prevention, reducing the burden of heart disease.
Expanded Goals
This supplementary notice supports the original goals of the Jump ARCHES program while emphasizing the unique challenges and opportunities within the four focus areas. Applicants are encouraged to:
- Propose solutions that integrate seamlessly into existing workflows within these institutes.
- Collaborate with domain experts from the respective institutes to ensure relevance and impact.
- Demonstrate potential for significant clinical improvements, including scalability and sustainability of proposed solutions.
Enhanced Proposal Criteria
In addition to the standard evaluation criteria outlined in the original RFP, proposals tailored to these focus areas will be evaluated on:
- Relevance to strategic areas: Clear alignment with the identified challenges and priorities of the four focus areas.
- Interdisciplinary expertise: Inclusion of collaborators with clinical, engineering and social science expertise specific to the targeted institute.
- Potential impact: Demonstrated capacity to improve patient outcomes, enhance operational efficiency and contribute to health equity.
Research grants have been awarded to the following projects:
Utilizing machine learning to predict asthma exacerbations in children
- Jimeng Sun, University of Illinois Urbana-Champaign
- Adam Cross, OSF HealthCare
This proposal aims to develop a personalized predictive model for pediatric asthma using patient-specific data from electronic health records (EHR), including medication adherence estimates. The model will integrate data analysis techniques such as regression, gradient boosting and knowledge graph-based deep learning to predict asthma exacerbations, focusing on high-risk pediatric patients. Given the disproportionate impact on ethnic minorities and children from lower socioeconomic backgrounds, the model aims to improve clinical outcomes, reduce emergency visits and lower health care costs by preventing exacerbations through better outpatient management and home care.
Virtual Patient with Mixed Reality Haptic Task Trainers for Neurologic Exam
- Elizabeth Hsiao-Wecksler, University of Illinois Urbana-Champaign
- Christopher Zallek, OSF HealthCare
This project aims to enhance clinical training in neurologic exams by combining Mixed Reality (MR) visualization with haptic feedback from robotic task trainers. The goal is to create a virtual patient that can be assessed during neurologic exams to improve health care trainees' ability to evaluate muscle tone and reflexes in patients with neurological conditions. The project will develop MR haptic trainers, integrating virtual patients with task trainer movements and evaluate their effectiveness in replicating real patient conditions. This approach seeks to improve clinician education and patient outcomes in neurological care.
Canine Pre-Clinical Imaging Models for Neurological Interventions in Glioma
- Kari Foss, University of Illinois Urbana-Champaign
- Andrew Tsung, OSF HealthCare
This project aims to develop advanced non-invasive imaging tools for the detection and characterization of gliomas in both humans and canines. Using magnetic resonance spectroscopy (MRS) and magnetic resonance elastography (MRE), the project seeks to evaluate gliomas at the functional level, moving beyond the limitations of traditional MRI. By applying these techniques to dogs with suspected gliomas, the team aims to establish the canine model as a pre-clinical tool for predicting the effectiveness of anti-tumor therapies in humans. The findings from MRS and MRE will be correlated with histopathological data to improve diagnostics and therapeutic strategies.
I-DocAssist: AI-based Assistant for Claims Optimization & Denial Reduction
- Dilek Hakkani-Tür, University of Illinois Urbana-Champaign
- Sohail Syed, OSF HealthCare
This project aims to develop an AI assistant to reduce the burden of insurance paperwork for health care providers. By utilizing large language models, the AI assistant will assist clinicians in real-time during patient assessments, highlighting insurance-specific requirements and suggesting necessary documentation. This system will track insurance guidelines, ensure compliance with requirements, and streamline the claims process, ultimately reducing denials. Long-term, the AI will automate claims preparation, generate appeal letters and monitor insurance trends, allowing providers to spend more time with patients and less on administrative tasks.
Leveraging AI to Reduce Provider Burden in Lifestyle Medicine Shared Medical Appointments
- James Rehg, University of Illinois Urbana-Champaign
- John Vozenilek, OSF HealthCare
- Sarah Zallek, OSF HealthCare
This project aims to use AI to address two challenges in Lifestyle Medicine Shared Medical Appointments (LM-SMA): (1) identifying patients and forming cohorts for lifestyle medicine and (2) providing actionable insights on sleep irregularities. By applying self-supervised learning to patient data, the project will identify high-risk individuals and group them into similar cohorts for peer support. Additionally, AI will analyze wearable sleep data to detect patterns and generate insights that will help tailor interventions and foster social support. The solution aims to automate patient selection and enhance the efficacy of LM-SMAs, improving patient care.
AI-Assisted Well-being Support for Family Caregivers of Dementia Patients
- Koustuv Saha, University of Illinois Urbana-Champaign
- Daniel S. Brown, OSF HealthCare
This project aims to develop an AI-powered system to support caregivers of individuals with dementia. The system will integrate into clinical workflows, providing real-time emotional support and connecting caregivers with resources. Key features include assessing caregiver well-being during clinical visits, continuous monitoring of stress and resilience and personalized interventions. A clinician-guided AI chatbot will deliver evidence-based mental health strategies like CBT, while a dashboard will allow clinicians to track caregiver needs. This system addresses caregiver burden by offering continuous support, improving mental health outcomes and enhancing both caregiver and patient care.
Leveraging Continuous Smartphone Sensor Data in Spine Surgery
- James Rehg, University of Illinois Urbana-Champaign
- Mark Chwajol OSF HealthCare
This project aims to improve post-operative spine care with a secure, HIPAA-compliant mobile platform that tracks real-time patient recovery using sensor data from iOS devices. The platform captures objective metrics such as light levels, device usage and geolocation alongside patient-reported outcomes and clinical event logs. By establishing baselines during the pre-operative phase, the system allows for early detection of complications and personalized monitoring. The platform predicts short-term issues like readmissions and opioid-related problems, as well as long-term outcomes such as implant failure. This project will help enhance patient outcomes, reduce rehospitalization rates and lower health care costs.
Automating Aortic Aneurysm Analysis from Medical Imaging
- Brad Sutton, University of Illinois Urbana-Champaign
- Matthew Bramlet, OSF HealthCare
This project aims to develop and expand a technology for automated 3D aortic segmentation and measurement to aid in the detection and monitoring of aortic aneurysms. Building on Phase I success, this Phase II proposal seeks to integrate this tool across various imaging modalities, such as CT and MRI, to provide consistent, accurate measurements of the aorta. By automating this process, the project aims to improve early detection, reduce human error and support clinicians in identifying at-risk patients, ultimately enhancing cardiovascular care and prevention of aortic dissection. This initiative aligns with strategic goals outlined by the OSF HealthCare Cardiovascular Center of Excellence.
Machine Learned 4D Heart Subsegmentation
- Brad Sutton, University of Illinois Urbana-Champaign
- Matthew Bramlet, OSF HealthCare
This project aims to advance cardiac imaging by automating the segmentation of heart chambers in 3D/4D models derived from CT scans. The current technology, developed by the AIM lab, uses machine learning to generate a 3D heart model, which is then analyzed to provide detailed and accurate measurements. The new approach will improve diagnostic accuracy, reduce clinician workload and offer enhanced analysis for complex conditions like congenital heart disease. By moving from 2D to 4D models, the project aims to streamline cardiovascular diagnostics, improve post-processing workflows and set a new standard in cardiac imaging technology.
VR Surgical Behavioral Analysis
- Jennifer Amos, University of Illinois Urbana-Champaign
- Matthew Bramlet, OSF HealthCare
This project aims to enhance pre-surgical planning in congenital heart surgery through a virtual reality (VR)-based decision-making model. The focus is on refining the analysis of complex data and decision-making processes in surgery, using VR to create detailed 3D models of patient anatomy. The project includes developing machine learning (ML) models to automate the segmentation and analysis of cardiac chambers, enabling precise, individualized planning. By comparing decision-making patterns between novice and expert surgeons, the goal is to improve mental preparedness, decision accuracy and ultimately surgical outcomes.
High-Fidelity Pediatric Heart and Fontan Assistive Device Simulators
- Sameh Tawfick, University of Illinois Urbana-Champaign
- Harma Turbendian, OSF HealthCare
- Matthew Bramlet, OSF HealthCare
This project aims to develop a technology platform combining 3D printing and polymeric artificial muscles to create biomimicking cardiovascular organs. Focused on pediatric heart simulators, mitral valve testing and an extravascular circulatory support device for Fontan patients, the platform will enhance education, simulation and clinical testing. Key aims include developing a Fontan circulatory assist setup, a high-fidelity beating pediatric heart model and a mitral valve testing simulation. Phase 2 will refine these technologies, establish external funding and enhance clinical applications, while addressing congenital heart disease challenges and supporting ongoing innovation in cardiovascular care.
AI-Driven Prediction and Prevention of DKA in Pediatric T1DM
- Jimeng Sun, University of Illinois Urbana-Champaign
- Adam Cross, OSF HealthCare
This project seeks to develop an AI-driven predictive modeling suite to improve diabetic ketoacidosis (DKA) risk stratification in pediatric type 1 diabetes (T1DM). The suite will include three models: a Screening Model, Confirmatory Model and Surveillance Model. These models aim to detect at-risk patients early, confirm true risk and continuously monitor over time, offering a more nuanced and accurate approach than relying solely on hemoglobin A1c levels. The goal is to prevent DKA through timely, targeted interventions while minimizing unnecessary clinical actions and reducing health care resource waste.
Advancing Physical Immersion Toward Low-Input, High-Return Simulation
- Inki Kim, University of Illinois Urbana-Champaign
- Paul Jeziorczak, OSF HealthCare
This project aims to expand the use of low-input, high-return simulations for effective training in medical education. Building on previous work, the proposal focuses on enhancing CPR training for patient families and laparoscopic skill development for surgical residents. Using Mixed Reality (MR) and Digital-Twin (DT) technology, the project improves physical immersion during CPR training and enables real-time skill assessment, including predicting risks like rib fractures. The initiative seeks to improve learning outcomes and patient care through resource-efficient simulation tools at OSF HealthCare and UICOMP, addressing training needs across clinical settings.
