An official website of the United States government
Digital Health and Artificial Intelligence Tools for Biomedical and Behavioral Research: Validity and Utility
When beginning your next investigator-initiated application, consider the following NIH highlighted topic. The area of science described below is of interest to the listed NIH Institutes, Centers, and Offices (ICOs). This is not a notice of funding opportunity (NOFO).
Apply through an appropriate NIH Parent Funding Announcement or another broad NIH opportunity available on Grants.gov. Learn how to interpret and use Highlighted Topics.
Topic Description
Post Date: June 26, 2026
Expiration Date: June 26, 2028
This topic encourages evaluation of the validity and clinical utility of digital health and artificial intelligence (AI) tools and technologies in biomedical and behavioral research. In this context, digital health technologies include computing platforms, software, sensors, mobile and wearable device-based technology, health information technology, connectivity, telehealth, telemedicine, and internet of things when used for healthcare or health-related research. AI tools include integrated and stand-alone computational technologies, risk assessment and prognosis algorithms, and software used for health promotion, disease prevention, and treatment of health conditions.
Rapidly evolving use of digital and AI technologies in research and health care has revealed both promises and perils. These technologies have accelerated scientific breakthroughs in multiple domains, expanded clinical reach, and surpassed human performance on several task-specific benchmarks (e.g., information extraction and image generation). However, their analytical validity, clinical validity, reliability, and/or utility in research and clinical care settings have often not been thoroughly examined. Optimizing dissemination and implementation and ensuring widespread use of these technologies requires more rigorous evaluation across research, community, and clinical settings, as well as in different populations and health contexts.
NIH encourages research projects that evaluate reliability and sensitivity, as well as projects that validate digital health and/or AI tools for use in research and clinical settings. NIH sees these technologies as important for assessing chronic and mental health conditions and identifying factors that can impact or predict morbidity and mortality. It is critical to determine the utility of using emerging technologies to improve health outcomes across all populations, particularly among those with limited access to traditional healthcare. Technical factors that might affect the use of these tools at different stages of the lifespan should also be assessed. Applicants are strongly encouraged to justify the impact of validating the specific tool or technology in the context or population under study.
Projects should include rigorous risk assessments, data security, and AI governance. Studies proposing secondary analysis should address the sufficiency of existing datasets for rigorous validation. Investigators are strongly encouraged to design projects that will deepen the evidence base for digital health and AI applications and to employ techniques that minimize potential imbalances or biases in data sources. Applications should include plans for risk mitigation that ensure safety, privacy, and effectiveness for all individuals. Before submission, applicants are strongly encouraged to review the specific research interests of participating NIH Institutes and Centers and to direct inquiries to the listed Scientific Contacts.
Participating ICOs
Janine Simmons, M.D.
[email protected]
NCCIH supports the development and validation of digital health and AI/ML tools in the context of complementary and integrative health (CIH) approaches and their effects on whole person health, pain, resilience, and well-being. Priority areas include:
- Developing and validating digital health technologies (e.g., sensors, mobile/wearable devices, telehealth) and AI tools to assess pain, promote whole person health, and optimize CIH approaches.
- Developing and validating tools to advance understanding of the behavioral and biological mechanisms of CIH approaches.
- Evaluating the analytical and clinical validity, and/or reliability of digital health and AI tools for CIH approaches in both research and real-world settings.
- Assessing risk, data security, privacy protection, AI governance, and bias mitigation to ensure safe, effective, and equitable use in CIH studies.
Emrin Horgusluoglu, PhD
[email protected]
NCI is interested in analytical and clinical validation of digital health tools and AI technologies across the cancer control continuum. Priority areas include but are not limited to:
- Validating digital health and AI tools for measuring cancer-related risk factors (e.g., physical activity, sleep, diet/nutrition, sun exposure, cannabis use, and alcohol use) across the lifespan
- Assessing validity and utility digital health and AI tools not yet validated for use in cancer-related contexts
- Validating digital and AI platforms that collect, harmonize and aggregate cancer incidence prevalence, and survival
- Validating digital and AI platforms used in cohort identification for clinical studies
- Validating digital and AI biomarkers for use in cancer-related contexts such as monitoring cancer risk factors, prediction and diagnosis of cancer, or monitoring symptoms and toxicities
- Validating digital health and AI tools that improve patient access, clinical decision-making, and cancer outcomes
Dana Wolff-Hughes
[email protected]
NHLBI encourages validation of digital health and AI/ML tools for heart, lung, blood, and sleep (HLBS) conditions, aligned with its Strategic Vision and Research Priorities. Emphasis is on validating existing devices and algorithms for HLBS-relevant, condition-specific diagnosis, monitoring, or treatment guidance. NHLBI prioritizes rigorous assessment of system performance, including reliability, robustness, and risks from non-generalizable models and non-representative data to reduce health disparities. Areas of interest include, but not limited to:
- AI for risk prediction, early detection and prevention of HLBS diseases
- AI for cardiovascular and pulmonary imaging, phenotyping, and diagnosis AI-assisted clinical decision support and workflow optimization
- Remote monitoring for detection and follow up AI-enabled point-of-care testing
- Sleep assessment for screening and diagnosis Hematologic diagnostics
- Wearable-based studies capturing HLBS physiology and outcomes
NHLBI Highlighted Topics
[email protected]
The National Institute on Aging (NIA) supports evaluation and validation of digital/AI tools relevant to aging and Alzheimer’s disease and related dementias (AD/ADRD). NIA seeks research that strengthens methodological rigor, reproducibility, and real-world utility across data sources, populations, and care contexts. Areas of interest include:
- AI/ML analysis of multimodal data to study aging biology, function, behavior, longevity, and biomarkers or therapeutic targets
- Data linkage methods, natural language processing, and neural networks to integrate heterogeneous datasets
- Analytical and clinical validation of digital and AI tools as biomarkers, clinical outcome assessments, prognostic indicators, or decision-support
- Evaluation of reliability and performance across diverse datasets and care settings
- Real-time monitoring or prediction of health, function, and social or environmental determinants
- Clinical effectiveness and use of tools in decentralized or traditional clinical trials
Joe Chiarenzelli, MPH
[email protected]
NIAAA encourages the development and validation of novel tools, as well as the rigorous evaluation of existing approaches using objective benchmarks, diverse data sources, and robust designs. Emphasis is on improving measurement accuracy, acceptability, and utility across populations and settings.
Examples of interest include:
- Smartphone, wearable, and AI-based sensing tools to detect alcohol use, craving, relapse risk, and related harms
- AI prediction tools for early identification of high-risk drinking and related harms using multimodal data
- Natural language processing (NLP)–based AI tools to quantify alcohol consumption and identify alcohol-related risk or treatment needs
- AI-enabled just-in-time adaptive interventions to reduce heavy drinking
- Scalable digital SBIRT (Screening, Brief Intervention, and Referral to Treatment) and recovery support tools
Wenxing Zha, Ph.D.
[email protected]
NIAMS encourages research that advances generalizable methodological frameworks and rigorous methods to validate, optimize, or improve digital health and/or AI technologies in these areas.
Example areas of interest include:
- Repurposing validated digital health, or AI/machine learning (ML) tools for disease detection, stratification, or longitudinal monitoring
- Rigorous validation of AI-enabled measures of physical activity, sleep, nutrition, and other health behaviors
- AI/ML applications using real-world data collection (e.g., electronic health records, wearables) to support quality-of-life assessment or symptom management
- Validation of decision support systems to enhance shared decision-making and patient engagement
- Evidence-based technologies for education and capacity building (e.g., tools for community-led co-design)
Raj Srinath
[email protected]
Mei Qin, M.D., Ph.D.
[email protected]
NIDA encourages research to validate digital-health tools and AI/ML technologies to prevent, diagnose, and treat substance use disorders (SUDs), including associated health and social consequences. SBIR/STTR applications are encouraged when market need is defined and the product has strong commercialization potential.
Some examples of areas of interest include:
- Using technology to advance understanding of both behavioral and neurobiological components of SUD, including the influence of environmental and social factors
- Developing and validating technologies that help individuals collect, manage, and use SUD-related and personal health data
- Developing clinical-grade mobile, web, or other software to deliver safe, effective SUD treatments
- Diagnostic and monitoring tools to optimize SUD interventions.
- Validating commercial digital health tools in research and real-world settings, including those with functional SUD prototypes or those repurposed from other indications for SUD populations
Ming Zhan
[email protected]
Division of Extramural Research
[email protected]
NIMH prioritizes research validating digital health/AI tools for mental health and/or HIV prevention, diagnosis, treatment, and service delivery, including:
- Validation of AI/ML tools for risk assessment, diagnosis, relapse prediction, treatment selection, and symptom management in real-world settings
- Evaluation of AI clinical decision support tools, including accuracy, reliability, safety, subgroup performance, and impacts on workflow and outcomes
- Validation of multimodal AI systems integrating clinical, behavioral, sensor, imaging, or genomic data, with attention to bias, robustness, and human oversight
- Safety evaluation and lifecycle management of digital/AI tools, including monitoring for model drift, performance degradation, automation bias, and post-deployment harms
- Implementation validation of digital health and AI/ML tools used across the HIV prevention/care continuum – including clinical decision support - and the intersection of mental health and HIV, in real-world settings
Lori Scott-Sheldon (HIV-related MH applications)
[email protected]
Michele Ferrante (all other MH applications)
[email protected]
NLM strongly encourages research focused on developing generalizable, disease-agnostic methods, tools, and validation standards to assess and optimize digital health AI. Rigorous evaluation of rapidly evolving AI technologies and health data use is critical to ensure reliability, wide adoption, and safe use. A thorough understanding of key development factors for digital health tools and platforms, along with user perspectives, is essential to inform design, optimize long-term health outcomes and promote sustained use of evidence-based prevention. NLM will support digital health research across settings, lifespan, and populations, leveraging robust AI tools to enhance prevention, care engagement, and treatment adherence for chronic diseases and other conditions, including HIV and autism. Such cross-cutting benchmarking frameworks will be increasingly important for promoting efficient, trustworthy, and user-friendly AI-enabled digital health solutions.
ICO Scientific Contact:Yanli Wang, PhD
[email protected]
Office of Research on Women’s Health (ORWH) supports research on digital health and AI tools in women’s health in alignment with Congressional language:
- AI/ML tools and computational approaches to assess the role of sex as a biological variable in digital health
- AI tools to characterize hormonal changes across the life course, including midlife/menopausal transition
- Use of AI tools for diagnosis/treatment of conditions unique to or affecting women
Office of Autoimmune Disease Research in ORWH is interested in rigorous, reproducible research of digital health and AI tools in autoimmune disease, including:
- Development of AI/ML and computational tools for autoimmune disease research
- Digital health technologies investigating drivers of autoimmune disease signs, symptoms, and flares
- Platforms and tools to improve modeling of predictors and risk factors for autoimmunity across the lifespan
- Digital applications to support rigorous implementation science in autoimmune disease research
Raven Hardy Richard, Ph.D.
[email protected]
Victoria Shanmugam, MBBS, MRCP, FACR, CCD
[email protected]
For technical issues E-mail OER Webmaster