Graph-Based Fusion of Imaging and Non-Imaging Data for Disease Trajectory Prediction

This study proposes a graph convolutional neural networks (GCN) architecture for fusion of radiological imaging and non-imaging tabular electronic health records (EHR) for the purpose of clinical event prediction. We focused on a cohort of hospitalized patients with positive RT-PCR test for COVID-19 and developed GCN based models to predict three dependent clinical events (discharge from hospital, admission into ICU, and mortality) using demographics, billing codes for procedures and diagnoses and chest X-rays. We hypothesized that the two-fold learning opportunity provided by the GCN is ideal for fusion of imaging information and tabular data as node and edge features, respectively. Our experiments indicate the validity of our hypothesis where GCN based predictive models outperform single modality and traditional fusion models. We compared the proposed models against two variations of imaging-based models, including DenseNet-121 architecture with learnable classification layers and Random Forest classifiers using disease severity score estimated by pre-trained convolutional neural network. GCN based model outperforms both imaging-only methods. We also validated our models on an external dataset where GCN showed valuable generalization capabilities. We noticed that edge-formation function can be adapted even after training the GCN model without limiting application scope of the model. Our models take advantage of this fact for generalization to external data. © 2023 IEEE.

GeoHealth Perspectives on Integrated, Coordinated, Open, Networked (ICON) Science

GeoHealth research both characterizes and predicts problems at the nexus of earth and human systems like climate change, pollution, and natural hazards. While GeoHealth excels in the area of integrated science, there is a need to improve coordinated and networked efforts to produce open science to enable environmental justice. There is a need to resource and empower frontline populations that are disproportionately marginalized by environmental injustice (i.e., the unequal protection from environmental harms and lack of access and meaningful engagement in decision making for a healthy environment;EPA, 2022, https://www.epa.gov/environmentaljustice). GeoHealth practice has the opportunity to advance environmental justice or the “fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income” with respect to how research and collaboration of GeoHealth professionals supports the “development, implementation, and enforcement of environmental laws, regulations, and policies” that produce equal protection from environmental and health hazards and access to the decision making for a health environment (EPA, 2022, https://www.epa.gov/environmentaljustice). Here we highlight barriers and opportunities to apply an equity‐centered ICON framework to the field of GeoHealth to advance environmental justice and health equity.

Community conversations on cancer: Creating and implementing a community engagement strategy for the 2022-2027 Illinois Comprehensive Cancer Control Plan through an academic-state public health department partnership

Background: Comprehensive cancer control (CCC) plans are region-specific blueprints that identify cancer priorities and health equity informed strategies to address cancer burden and are supported by the National Comprehensive Cancer Control Program through the Centers for Disease Control and Prevention (CDC). Although CCC plans are created by stakeholder coalitions, few have focused on community engaged approaches, which may diminish their applicability for community members. Thus, in preparation for its forthcoming 2022-2027 CCC plan, the Illinois Comprehensive Cancer Control Program collaborated with the University of Illinois Cancer Center's Community Engagement and Health Equity office to implement a community engagement strategy to address cancer burden. Objective: To describe the development and implementation of a community engagement strategy for the 2022-2027 Illinois CCC plan. Method: The goal of the community engagement strategy was to identify barriers, facilitating factors and recommendations related to cancer burden and equity in Illinois by engaging diverse community stakeholders. A statewide town hall and focus groups (FGs) were implemented in early 2021. Thedevelopment and analysis of the community engagement strategy were guided by the Model for Analysis of Population Health and Health Disparities, CDC's CHANGE Action Guide, and the Community ToolBox. Semistructured guides included questions about fundamental causes of health, social and physical contexts, individual demographics and risk factors, and biologic responses and pathways. The town hall was open to Illinoisians over 18 years of age. FG participants were selected using purposive sampling to maximize group heterogeneity. Eight FGs were held, one each for: rural residents, survivors, young survivors, caregivers, and Spanish speakers, and three that were a mix of community members. Town hall notes and FGs were analyzed using content analysis. Results were synthesized and a final report was included in the forthcoming plan. Results: Town hall and FG (n=8) participants (n=115) included cancer survivors (36%), caregivers (27%), Latinos (17%), African Americans (23%), and rural residents (14%). Throughout the development of the plan, data were continuously reviewed with the coalition developing the CCC Plan. The final report described multi-level factors that contribute to cancer disparities among Illinoisians, proposed recommendations to improve health across the cancer continuum across multiple levels, funding priorities, and the impact of COVID-19 on cancer care. Participant quotes supported strategies throughout the plan. Conclusion: A robust community engagement strategy for the forthcoming 2022-2027 Illinois CCC Plan was implemented through a successful academic-state public health department partnership. This strategy ensures that the plan reflects the expertise and voices of Illinoisians impacted by cancer. This engagement strategy, framed around health determinants that impact cancer risk and outcomes, may be replicated by other coalitions creating CCC plans.

A roadmap for patient- and family-centered care during the pandemic

The COVID-19 pandemic has affected profoundly all parties and health-care delivery. Most hospitalized patients find that separation from loved ones takes a toll on them and the safety measures instituted come at a huge cost to their physical and emotional well-being. Family members are distressed and traumatized because they are unable to support or advocate for their loved ones and feel disconnected from the health-care team. Health-care professionals (HCPs) experience psychological and moral distress because the absence of families negatively affects their ability to provide quality, compassionate, and timely care. As a result, there have been calls for revisions of policies and procedures to reprioritize patient- and family-centered care (PFCC). Building upon their experience as frontline HCPs, health-care team leaders, and patients and family members who are also frontline HCPs, the interprofessional team (psychologists, psychiatrists, hospitalists, nurses, and specialists) of co-authors reviewed the literature and developed collegial consensus on a roadmap for PFCC during the pandemic that includes guiding principles along with associated best practices and implementation strategies for their operationalization. The guiding principles focus on providing safe yet compassionate and ethical PFCC;balancing community health and the mitigation of viral transmission with appreciating family members as essential partners in care;fostering communication between patients and their families;and promoting interactions and decision-making among HCPs, patients, and families. We hope this roadmap facilitates the transformation of the health-care culture during the pandemic to one that is truly patient and family centered. (PsycInfo Database Record (c) 2021 APA, all rights reserved) © 2021 American Psychological Association