MCMollie R. Cummins

Mollie R. Cummins

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mollie R. Cummins has a total of 214 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Accelerating environmental health sciences through data and metadata standards to support FAIR principles

    by David Hines, Jiang Bian, Patricia Ceger, Nicole C. Deziel, Nomi Harris, Xing He, Hui Hu, Mollie R. Cummins, Sierra Moxon, Christopher J. Mungall, et al.

    Environmental health sciences (EHS) generate varied and highly specialized datasets that span fields including exposure science, toxicology, epidemiology, and molecular biology. Despite their shared relevance to EHS, these data streams can be siloed due to inconsistent metadata practices, limited use of common data elements, and uneven adoption of community standards. This fragmentation limits data integration, reduces reproducibility, and hampers the development and application of emerging computational approaches, including artificial intelligence (AI) and machine learning (ML). This article provides an overview of existing data and metadata standards most relevant to EHS, including public health initiatives, controlled vocabularies and ontologies, and platforms that support findability, accessibility, interoperability, and reusability (FAIR data practices). We outline practical steps for researchers who wish to incorporate standards into their workflows, emphasizing early planning, use of established templates, and availability of institutional resources. We then provide examples of ongoing efforts across EHS domains to develop interoperable infrastructures that enable data integration and highlight how standards enable leveraging of emerging technologies such as AI/ML to enhance data use. This work makes a key contribution to the EHS field by demonstrating the importance of community engagement and active collaboration among developers and users, demystifying the processes of adoption of these tools across EHS domains. As AI/ML technologies stand poised to revolutionize cross-disciplinary science, we provide a vision for the future of an AI-ready, interoperable EHS data ecosystem to support interdisciplinary research, enhance transparency, and accelerate scientific discovery.

  • Open Access

    An Agentic AI Pipeline for an Evidence-Grounded Environmental Health Public Information System: Development and Evaluation

    by Stein H. Ingebretsen, Ram Gouripeddi, Mollie R. Cummins, Julio C. Facelli, Fatemeh Shah-Mohammadi

    Background: Existing environmental and health information systems typically provide isolated data streams, requiring users to interpret heterogeneous information sources without personalized guidance. Residents and health providers of urban regions with complex environmental conditions, such as the Salt Lake Valley, frequently seek timely, location-specific information about weather, air quality, and their associated health impacts, but may find it difficult to integrate this information in their decision-making, precluding the effective translation of valuable environmental health information into action. Objective: This study aimed to develop and evaluate an agentic artificial intelligence (AI) pipeline that delivers evidence-grounded, neighborhood-scale responses to health-related environmental questions raised lay persons by dynamically orchestrating authoritative weather, neighborhood-scale environmental observations, air quality, and health-evidence resources. Methods: We developed an agentic AI pipeline using LangGraph and OpenAI large language models to interpret natural language queries, select and invoke relevant external tools, retrieve real-time environmental data and evidence-based health guidance, and synthesize responses. The pipeline integrates multiple authoritative resources, including weather forecasts, air-quality observations, geospatial services, and a retrieval-augmented generation (RAG) module built from asthma-related clinical guidance. The system incorporates fallback mechanisms to improve robustness when external services are unavailable. We evaluated the pipeline using a curated benchmark of representative user questions, assessing tool-needed recognition, correct tool selection, avoidance of irrelevant tools, parameter mapping, analytic pathway selection, and incorporation of tool outputs into the final response. Results: The developed system successfully integrated heterogeneous environmental and health information sources into a unified conversational interface that produces evidence-grounded, neighborhood-scale responses for the Salt Lake Valley. Evaluation demonstrated generally strong tool recognition, parameter mapping, and incorporation of retrieved information across diverse question types, supporting the pipeline’s ability to retrieve and integrate authoritative environmental and health information consumable by lay persons and providing an important step for the T3 and T4 phases of the translational pipeline of exposure health informatics. Conclusions: By combining existing environmental health knowledge with local sensor observations and neighborhood-scale environmental measurements, and authoritative state and national weather and air quality resources, the system delivers more localized and context-aware guidance than approaches that rely on authoritative regional or state-level information. Rather than depending on the internal knowledge of a large language model, the pipeline dynamically retrieves, synthesizes, and attributes evidence from authoritative sources. The modular architecture enables seamless integration of additional environmental data streams, health knowledge sources, and geographic regions, providing a scalable foundation for future translational pipelines, completing the T3 to T4 path for environmental health.

  • Open Access

    Toward Citation-Guided Sensor Metadata Enrichment: A Large Language Model-based Approach (Preprint)

    by Fatemeh Shah-Mohammadi, Sunho Im, Urvi Varma, Julio Facelli, Mollie Cummins, Ramkiran Gouripeddi

    UNSTRUCTURED Background: Sensor metadata is critical for exposure health research because it supports accurate sensor identification, deployments, data integration, interoperability, and reproducibility. Yet it is often fragmented across multiple heterogeneous sources, such as scientific literature and manufacturer guides, where key specifications are frequently reported indirectly through citation chains, making reference tracing essential for metadata enrichment and completeness. Objective: To address this bottleneck, we developed and evaluated an LLM-based automated, citation-aware pipeline that enriches sensor metadata extracted from a primary article by identifying sensor-related citation markers and extracting additional metadata from the referenced sources. Methods: We extend our prior LLM-based metadata extraction approach by (i) detecting sensor mentions in full-text articles, (ii) capturing nearby citation markers, (iii) resolving markers to full bibliographic entries in the reference list, and (iv) retrieving cited papers to extract additional sensor metadata that may be absent from the primary document and using it to enrich and complete the base metadata. Results: Across 20 primary papers, the citation extraction component achieved 74.2% precision, 92.0% recall, 82.1% F1-score, and 69.7% accuracy, and all extracted bibliographic entries were correctly matched to their source references. This component increased sensor extraction by about 261%, yielding 94 additional sensors overall. Conclusions: The developed citation-guided pipeline improved sensor discovery and metadata completeness, thereby supporting the development of richer, more complete sensor metadata repositories.

  • Open Access

    Leveraging Digital Health for Sustainable Care: Advancing Telemedicine, Climate Change Mitigation, and Global Health Equity

    by Manijeh Berenji, Mollie R. Cummins, Ram Gouripeddi, Titus K. Schleyer, Jonathan Suh, Betty Villantay

    Objectives: This narrative review describes the role of digital health in addressing climate change adaptation, mitigation, insight, and equity. We survey existing evidence and efforts and identify high-value opportunities.Methods: A comprehensive review of recent literature was conducted, focusing on the intersection of digital health technologies, climate adaptation, and mitigation strategies.Results: Digital health technologies have emerged as powerful tools in addressing climate change. Healthcare organizations are investing in digital health to reduce greenhouse gas emissions, develop adaptive strategies for climate-related health impacts, and address emerging challenges like climate-driven migration. This approach can enhance sustainability, improve patient care, and contribute to global health security.Conclusions: While digital health offers significant potential in combating climate change, challenges remain in reducing the healthcare sector's greenhouse gas emissions and bridging the digital divide. Future research should prioritize developing more energy-efficient AI models, improving the resilience of digital infrastructure to climate events, and integrating climate considerations into health informatics. The field must also address the ethical implications of AI in climate decision-making and ensure equitable access to digital climate solutions globally.

  • Open Access

    Development of Prediction Models for Mortality and Debility Among Critically Ill Adults Receiving High-flow Nasal Cannula Therapy

    by Jace Johnny, Katherine A. Sward, Bob Wong, Mollie R. Cummins, Ramkiran Gouripeddi, Sean J. Callahan

    ObjectivesThe objective of this study was to examine whether factors known at high-flow nasal cannula (HFNC) initiation could predict inpatient mortality and debility.MethodsMultivariable logistic regression models were developed and evaluated using a secondary data analysis of the Medical Information Mart for Intensive Care database (MIMIC-IV). The cohort consisted of adult patients admitted to Intensive Care Units at Beth Israel Deaconess Medical Center between 2008 and 2019. Patients with a tracheostomy before HFNC use and intubated within 1 hour after HFNC initiation were excluded.ResultsOf the 2,031 patients, 463 (22.8%) died, and 1059 (52.1%) met the composite debility outcome. The final mortality model had an AUC of 0.77 (95% CI: 0.74-0.79). Age, heart rate, blood urea nitrogen (BUN), and need for hemodialysis at HFNC initiation were associated with increased odds of inpatient mortality. The debility model had an AUC of 0.65 (95% CI: 0.63-0.68). Extubation to HFNC therapy increased the odds of debility more than two-fold.ConclusionsA prediction model for inpatient mortality was developed, had good overall performance, and identified age, heart rate, and renal dysfunction as key risk factors. A debility model was also developed, highlighting how patients’ risk profile evolved, but requires further refinement.Implications for Clinical PracticeRisk prediction in patients receiving HFNC therapy should extend past predicting intubation, as other factors lead to mortality, such as concomitant renal dysfunction. Additionally, debility is a common outcome in those receiving HFNC therapy and warrants further study. The ratio of oxygenation (ROX) index performed poorly, in isolation, at predicting mortality and debility.

  • Open Access

    Target Trial Emulation of Vaccine Effectiveness in 5- to 17-years-olds with Prior SARS-CoV-2 Infection

    by Yuqing Lei, Jiajie Chen, Qiong Wu, Ting Zhou, Bingyu Zhang, Michael J. Becich, Yuriy Bisyuk, Saul Blecker, Elizabeth A. Chrischilles, Dimitri A. Christakis, et al.

    The effectiveness of COVID-19 vaccination in children and adolescents with prior SARS-CoV-2 infection remains unclear, particularly for Omicron subvariants. We evaluate vaccine effectiveness against reinfection with Omicron BA.1/BA.2, BA.4/BA.5, XBB, and later subvariants among 5- to 17-year-olds using data from the RECOVER initiative, a national electronic health record database covering 37 U.S. children's hospitals and health institutions. We emulate target trials by age group and variant period, comparing previously infected participants between January 2022 and August 2023. During the BA.1/BA.2 period, vaccination reduces the risk of reinfection, with effectiveness rates of 62% in children and 65% in adolescents. During the BA.4/BA.5 period, protection effectiveness in children was 57%, whereas no statistically significant protection is observed in adolescents. During the XBB and later period, no significant protection is observed in either group. In summary, COVID-19 vaccination provides protection against reinfection during the early and mid-Omicron periods in previously infected pediatric populations, but effectiveness declines for later variants.

  • Open Access

    Inflection Point: Artificial Intelligence, Digital Health, and the Evolution of Pre-Licensure Nursing Informatics Competencies

    by Mollie R. Cummins, Seneca Perri Moore, Katherine A Sward

    Recent advancements in artificial intelligence, telehealth, and sensor technologies have fundamentally transformed health care, creating an urgent need to update and revise nursing informatics competencies for pre-licensure education. The purpose of this narrative, non-systematic review is to (1) describe and characterize the evolution of pre-licensure nursing informatics competencies; (2) discuss the central phenomena driving current digital transformation in health care; and (3) discuss future directions to evolve pre-licensure nursing informatics competency frameworks in an era of digital health. We conducted multiple literature searches in support of the narrative review. Current pre-licensure nursing informatics competency frameworks require evolution, given the new knowledge and skills needed for nurses to engage with rapidly advancing digital health technologies. The review reveals the need for a fifth wave of competency development that prepares nurses as informed users, contributors, evaluators, and stewards of digital health tools. We must advance nursing informatics education to meet the workforce needs of the digital health era by building on a robust foundation of past competency development efforts.

  • Open Access

    Scaling sensor metadata extraction for exposure health using LLMs

    by Fatemeh Shah-Mohammadi, Sunho Im, Julio C Facelli, Mollie R Cummins, Ramkiran Gouripeddi

    Background: The rapid evolution and diversity of sensor technologies, coupled with inconsistencies in how sensor metadata is reported across formats and sources, present significant challenges for generating exposomes and exposure health research. Objective: Despite the development of standardized metadata schemas, the process of extracting sensor metadata from unstructured sources remains largely manual and unscalable. To address this bottleneck, we developed and evaluated a large language model (LLM)-based pipeline for automating sensor metadata extraction and harmonization from publicly available exposure health literature. Methods: Using GPT-4 in a zero-shot setting, we constructed a pipeline that parses full-text PDFs to extract metadata and harmonizes output into structured formats. Results: Our automated pipeline achieved substantial efficiency gains in completing extractions much faster than manual review and demonstrated strong performance with 88.0% accuracy, 88.0% precision, 93.0% recall, and an F1-score of 90.0%. Conclusions: This study demonstrates the feasibility and scalability of leveraging LLMs to automate sensor metadata extraction for exposure health, reducing manual burden while enhancing metadata completeness and consistency. Our findings support the integration of LLM-driven pipelines into exposure health informatics platforms.


  • Open Access

    Needs, Preferences, and Challenges of Environmental Health Researchers in Using Sensors: A Content Analysis (Preprint)

    by Rachel Nelson, Peter Taber, Pavan Motagi, Sukrut Shishupal, Sunho Im, Julio Facelli, Katherine Sward, The Smarter Expert Panel, Ram Gouripeddi, Mollie R. Cummins

    BACKGROUND Environmental exposures can influence human health in complex ways. It remains difficult for researchers to integrate exposure data, partly due to an unmet need for informatics and metadata tools. OBJECTIVE The purpose of this study was to understand the needs, preferences, and pain points of environmental health researchers regarding the selection, deployment, and integration of sensors for their research studies, to inform user requirements for a sensor metadata repository. METHODS We purposively recruited six exposure health researchers with expertise entailing sensors, corresponding to one of eight role types, and conducted semi-structured interviews between February 7-26, 2025. Interviews centered on understanding the needs, preferences, and pain points of environmental health researchers seeking to use sensors for their research projects. We conducted deductive content analysis of interview transcripts, guided by the HITREF framework. RESULTS The participants held primary roles of primary investigator, study coordinator, sensor developer, biomedical informaticist, and study sponsor. Content analysis revealed that participants consider multiple characteristics of sensors when selecting sensors for studies, including cost, physical parameters and limitations of the sensors, reliability, suitable environments for deployment, and software and processes required for data acquisition, transfer, integration, and analysis. User training and interaction are important considerations, often conceptualized as burdens on study participants that research teams seek to minimize. Participants described a desire for adequate support from sensor developers and flexibility in data transfer and analysis. CONCLUSIONS Participants in varied roles described many similar themes regarding considerations for sensor selection, deployment, and integration as well as desired features for a sensor metadata repository.

  • Open Access

    Scaling Sensor Metadata Extraction for Exposure Health Using LLMs

    by Fatemeh Shah-Mohammadi, Sunho Im, Julio C. Facelli, Mollie R. Cummins, Ramkiran Gouripeddi

    Background: The rapid evolution and diversity of sensor technologies, coupled with inconsistencies in how sensor metadata is reported across formats and sources, present significant challenges for generating exposomes and exposure health research. Objective: Despite the development of standardized metadata schemas, the process of extracting sensor metadata from unstructured sources remains largely manual and unscalable. To address this bottleneck, we developed and evaluated a large language model (LLM)-based pipeline for automating sensor metadata extraction and harmonization from exposure health literature publicly available. Methods: Using GPT-4 in a zero-shot setting, we constructed a pipeline that parses full-text PDFs to extract metadata and harmonizes output into structured formats. Results: Our automated pipeline achieved substantial efficiency gains in completing extractions much faster than manual review and demonstrated strong performance with average accuracy and precision of 94.74%, recall of 100%, and F1-score of 97.28%. Conclusions: This study demonstrates the feasibility and scalability of leveraging LLMs to automate sensor metadata extraction for exposure health, reducing manual burden while enhancing metadata completeness and consistency. Our findings support the integration of LLM-driven pipelines into exposure health informatics platforms.