Project Awardees
The Advanced Research Projects Agency for Health (ARPA-H) funds individual research projects that align with ARPA-H’s research focus areas but fall outside the scope of an ARPA-H program or initiative.
Projects are most often awarded through Mission Office Innovative Solution Openings (ISO) and historically through the Open BAA announcement. As of March 2024, ARPA-H is no longer accepting submissions for the Open BAA solicitation.
ARPA-H is pleased to announce the following project awardees.
MIGHTY: Microbe / phage Investigation for Generalized Health TherapY
The MIGHTY project aims to improve healthcare outcomes for all Americans in an era of rising antibiotic resistance, a significant threat to public health. MIGHTY will address the limitations of current therapeutic treatments by focusing on the human microbiome's role in health and disease. While antibiotics are often vital to kill harmful bacteria, they can also disrupt beneficial microbes. Phages are a class of viruses that only infect bacteria and offer an opportunity for more precise targeting of specific bacteria of interest. MIGHTY will harness the power of phages to alter the microbiome composition in favor of healthy outcomes. These phages will be provided in chewing gum or lozenges for initial targeting of the oral microbiome, particularly focusing on communities with high rates of oral disease. The MIGHTY project has the potential to revolutionize healthcare by using phages to maintain a balance between healthy and unhealthy bacteria, influencing every organ in the human body throughout our lifespan.
CLINAC-BP: Unobtrusive Near-field Continuous Monitoring of Clinically Accurate Blood Pressure
This project will develop technology for continuous, real-world monitoring of blood pressure, a key vital sign for assessing cardiovascular health. The proposed technology will improve accuracy, efficiency, and accessibility to enable individuals and their clinicians to monitor blood pressure (BP) “beat-to-beat,” facilitating diagnosis and management of cardiovascular diseases, which are the leading cause of mortality in the United States and globally. Moreover, digital access to unprecedented continuous data will generate novel insights for disease prediction and improved care for cardiovascular and circulatory conditions. To date, an accurate and continuous blood pressure monitor in a wearable form factor for consumer use is not available. The CLINAC-BP solution leverages technology to measure changes in the arteries’ electrical properties as the heart beats that can then be converted into blood pressure values. This interdisciplinary team has designed an approach that will integrate advances in radio frequency monitoring technology, hardware, and AI/software to develop a wearable and connected continuous blood pressure monitor for clinical accuracy as the user goes about daily life, at home, at work, and in the community.
TrustShadow: Towards Scalable and Strengthened Protection of Mobile Health Data with a Novel Trusted Execution Environment
In this project, Penn State University will expand the capabilities of a novel cybersecurity solution, TrustShadow, for Body Area Networks (BANs) consisting of multiple wearable devices and unmodified apps running on smartphones/tablets. Presently, only custom Operating System (OS) kernels and custom apps are compatible with TrustZone, limiting its applicability. Developed by AMD in 2004, TrustZone is a secure execution environment as an optional extension to hardware processers. Currently, extensive modification is needed to allow a usermode app to work with TrustZone. Enabling unmodified apps to operate within TrustZone through proposed technological innovations promises greater transparency, scalability, and practicality in addressing cybersecurity risks and safeguarding personal health data compared to conventional TrustZone implementations. This project aims to utilize the proposed TrustShadow technology to provide comprehensive protection for health data outside of clinical settings.
Closing the Doors to All Pathogens Through Integrated Digital Experiments
Understanding the human health importance of pathogen surveillance data is not currently possible from sequence or other molecular information alone. The Closing the Doors to All Pathogens Through Integrated Digital Experiments (DOORs) project aims to 1) Create a comprehensive computational map of pathogen-host interactions, 2) Uncover the best drug targets for an entire family of pathogens, instead of each individual pathogen as it’s done today, and 3) Test approved therapies and discover new ones via computer models. The vision for DOORs is to protect U.S. national security from emerging biological threats with an integrated digital experimental workflow that closes the loop for detecting, understanding, and responding to infectious diseases, while making these activities commercially viable through host-based therapeutic discovery.
AFC Health: The American Family Cohort: Enhancing and Validating Primary Care Data for Improving Urban and Rural Health
The AFC-HEALTH project will leverage the PRIME registry, the largest primary care data registry in the U.S., which contains Electronic Health Record (EHR) data on 8 million patients. The project will establish data partnerships to expand the registry by 50 percent, building an enriched dataset to help elucidate the social determinants of health and causal drivers of disease. By addressing critical barriers in data collection and utilization, the project will enable large-scale evaluation of the health effects of environmental, social, and educational policies. This will help enable the creation, implementation, and monitoring of evidence-based, equitable, and pragmatic primary care interventions. The use of responsible AI in healthcare will also be enhanced.
D3PH: Decentralized Diabetes Data Platform for Senior Health Management
The D3PH project will develop edge computing technology to create advanced machine learning models that will predict and improve diabetes foot ulcer (DFU) risk prediction and wound healing prediction for senior citizens with diabetes in multiple long-term care clinics. While the goal is to pilot the technology for diabetes complications, success would enable availability of important technology for federated data sharing, and model development and deployment, that could be used for countless conditions, enabling efficient and affordable preventative health interventions. This will ensure advanced disease and health prediction models and improved health outcomes, decreased costs, and a decreased need for acute interventions throughout the medical system.
Tissue Preservation under Stress
Traumatic tissue injuries can happen unexpectedly and lead to life-threatening conditions. Despite advancements in wound care, millions of Americans lack immediate access to specialized medical facilities, increasing the risk of chronic wounds or death. Aging populations and individuals with underlying health conditions like diabetes and obesity are at greater risk of developing chronic wounds that fail to properly heal. The Tissue Preservation under Stress project seeks to develop cutting-edge technologies aimed at preserving the health and function of biological tissues under urgent stress or injury conditions. These therapies have the potential to significantly improve health care access for all Americans, especially in trauma care in under-resourced and rural communities. The TPS effort is a two-year focused effort on advanced development to translate these technologies toward clinical trials.
Cyber Health Intelligence (CHI): Sensing and Self-Healing Cyber Vulnerabilities in Healthcare Ecosystems
The Cyber Health Intelligence: Sensing and Self-Healing Cyber Vulnerabilities in Healthcare Ecosystems project will extend a previously DoE-funded framework - Common Weakness Enumerations (CWE) - into the healthcare sector. This technology, applied to the full ecosystem of connected things in healthcare and combined with novel orchestration and machine learning capabilities will dramatically improve cyber defenses. MD Anderson will achieve this by combining a) the digital ecosystem of healthcare at MD Anderson, b) a proven approach using machine learning technologies to enumerate and characterize cyber vulnerabilities and corresponding weaknesses, and c) an approach to facilitate weakness prioritization and mitigation through interactive prompting.
Eliminating Out-of-Hospital Adverse Events for Everyone with Congenital Heart Disease (CHD)
This project seeks to create a noninvasive, integrative hardware and software monitoring solution that prevents low cardiac output syndrome (LCOS), the leading cause of postoperative mortality in children and adults with congenital heart disease (CHD). This wearable technology would allow for continuous, high quality data acquisition of cardiac output information from children with congenital heart disease, leveraging a novel multi-modal approach that integrates four different sensors on a non-invasive patch. These advances in hardware integration will be combined with machine learning methods to match the combined sensor signals and cardiac output.
MEDAGuard: Medical Device Functionality Shielding Against Adverse Effects of Authorized Updates by Leveraging EM Side-Channels
The MEDAGuard: Medical Device Functionality Shielding Against Adverse Effects of Authorized Updates by Leveraging EM Side-Channels project will observe electromagnetic (EM) emissions from medical devices to detect when a device behaves abnormally. By analyzing embedded information with EM signals, MEDAGurad can enhance systems assurance for medical devices such as remote patient monitors. This "cyber smoke detector” compares captured signals with a baseline model derived from a known, properly functioning system version. Any deviations from expected behavior trigger an alert, indicating abnormal system activity or unexpected side-effects of updates, patches, etc. MEDAGuard stands as a notable leap forward in monitoring and attestation domain, especially for critical medical devices where system stability is vital. This sets it apart from state-of-the-art-systems, which often demand shared hardware or network connections with the device under scrutiny. In contrast, MEDAGuard relies solely on physical proximity.