Zero AI Grants in Nursing Departments: A Systemic Blind Spot in Endoscopy Innovation (FY2020–2025)

Abstract

Background: Artificial intelligence (AI) technologies are rapidly transforming gastroenterology and endoscopy practice, with FDA-approved computer-aided detection systems now deployed in colonoscopy suites nationwide. However, the extent to which nursing scientists lead or participate in AI research development remains unclear, despite nurses being the primary operators of these technologies in clinical settings.

Purpose: This analysis examined NIH funding patterns for AI research relevant to endoscopy and gastroenterology across fiscal years 2020–2025, with particular attention to departmental distribution and nursing leadership in funded projects.

Methods: Using the NIH RePORTER database API, we identified and analyzed all active grants combining artificial intelligence with endoscopy-related terms. Projects were categorized by hosting department, principal investigator credentials, and research focus.

Results: Twenty-three AI-endoscopy projects were identified. All 23 grants were housed within medical, surgical, engineering, or computer science departments. Zero grants were identified within nursing departments. Nurse scientists were absent from principal investigator roles across all funded projects.

Conclusions: A significant gap exists between AI development for endoscopy practice and nursing science involvement in that development. Given that nurses are the primary implementers and operators of AI-assisted endoscopy technologies, this exclusion may contribute to implementation challenges, alert fatigue, and workflow disruption. Strategic investment in nurse-led AI research is needed to ensure technologies are developed with clinical nursing expertise from inception.

Keywords: artificial intelligence, nursing informatics, endoscopy, gastroenterology, research funding, implementation science, computer-aided detection

Introduction

Artificial intelligence is rapidly transforming healthcare delivery, with gastrointestinal endoscopy emerging as one of the most active domains for AI application (Hassan et al., 2024). In April 2021, the U.S. Food and Drug Administration granted de novo clearance to the GI Genius intelligent endoscopy module, marking the first FDA-approved AI system for detecting potential colorectal lesions during colonoscopy (FDA, 2021). Since then, multiple AI-assisted colonoscopy systems have demonstrated significant improvements in adenoma detection rates (ADR) and reductions in adenoma miss rates (AMR), positioning AI as an increasingly standard component of quality colonoscopy practice (Glissen Brown et al., 2022; Wang et al., 2020).

Despite this rapid technological advancement, a critical question remains underexplored: Who is developing these AI tools, and are the healthcare professionals who will implement them meaningfully involved in the research enterprise? This question is particularly relevant to nursing, as nurses comprise the largest segment of the healthcare workforce and serve as the primary operators of endoscopy equipment, managers of procedural sedation, and monitors of patient safety during gastrointestinal procedures (International Council of Nurses, 2017; Ronquillo et al., 2021).

The Nursing and Artificial Intelligence Leadership (NAIL) Collaborative has identified three priorities requiring immediate attention: (a) nurses must understand the relationship between clinical data and AI technologies; (b) nurses must be meaningfully involved in all stages of AI development and implementation; and (c) substantial untapped potential exists for nursing to contribute to AI technology development (Ronquillo et al., 2021). However, the extent to which these priorities are reflected in current research funding patterns has not been systematically examined within the specific domain of gastroenterology and endoscopy.

The purpose of this analysis was to examine NIH funding patterns for AI research relevant to endoscopy and gastroenterology, with particular attention to departmental distribution and nursing leadership in funded projects.

Background and Literature Review

AI in Gastrointestinal Endoscopy

Colorectal cancer (CRC) is the third leading cause of cancer-related death in the United States (American Cancer Society, 2024). Colonoscopy with polypectomy remains the most effective method for CRC prevention, with detection of adenomatous polyps directly linked to reduced cancer incidence and mortality. Research indicates that each 1% increase in adenoma detection rate reduces the subsequent risk of colorectal cancer by 3% (Corley et al., 2014). However, adenoma detection rates vary substantially among endoscopists, ranging from 7% to 53% depending on practitioner skill and environmental factors (Spadaccini et al., 2023).

Computer-aided detection (CADe) systems have emerged as a promising solution to address this variability. The GI Genius system demonstrated a 14% absolute increase in ADR compared to standard colonoscopy, with particularly significant improvements in flat lesion detection (42% increase) and polypoid lesion detection (36% increase) (Repici et al., 2020). A United States multi-center randomized trial demonstrated that CADe reduced polyp miss rates from 31.3% to 20.1% and adenoma miss rates from 32.4% to 15.5% (Glissen Brown et al., 2022).

A recent bibliometric analysis identified significant growth in AI endoscopy research, with publications increasing dramatically from just five in 2013 to 345 in 2023, representing a 1,160% increase over the decade (Zhang et al., 2025).

The Role of Nursing in Endoscopy Practice

Within the endoscopy suite, nurses serve multiple critical functions that directly intersect with AI technology implementation. Endoscopy nurses manage pre-procedure patient preparation, administer and monitor procedural sedation, operate and position endoscopy equipment, monitor patient vital signs and safety indicators, assist with tissue acquisition and polypectomy, and manage post-procedure recovery (SGNA, 2023).

Procedural sedation during endoscopy represents a particularly critical nursing responsibility. AI systems are now being developed to optimize sedation delivery during endoscopy, including machine learning models predicting sedation requirements and closed-loop systems adjusting propofol dosing in real-time (Syed et al., 2022; Xu et al., 2022). The ENDOANGEL system, a deep learning-based quality control tool for digestive endoscopy, has demonstrated improvements in emergence time and recovery time when used alongside traditional anesthesia monitoring (Xu et al., 2022).

Lessons from Electronic Health Record Implementation

The history of electronic health record (EHR) implementation offers cautionary lessons regarding technology development without adequate clinician input. Multiple studies have documented significant associations between EHR use and clinician burnout (Alobayli et al., 2023; Wu et al., 2024). The American Medical Association has emphasized that EHRs were developed without the input of frontline clinicians and were implemented without considering the impact on workflows (Ehrenfeld, 2024). As AI systems become integrated into clinical practice, ensuring frontline practitioners—including nurses—are involved from development through implementation becomes essential for successful adoption.

Nursing Informatics and AI

Recent analysis of NIH-funded grants related to AI and nursing identified 370 unique projects between fiscal years 2013–2023, representing a significant increase over the decade. However, the majority of National Institute of Nursing Research (NINR)-funded projects were training-focused rather than primary AI research grants (Wu et al., 2025). Key themes in nursing-related AI research include cognitive decline, inpatient predictive analytics, and chronic disease self-management, with limited focus on procedural nursing environments such as endoscopy (Wu et al., 2025).

Methods

Data Source

Data were obtained from the NIH RePORTER database (api.reporter.nih.gov), a comprehensive publicly accessible resource that tracks and reports on federally funded research projects.

Search Strategy

We searched the RePORTER database using the following query structure: (“artificial intelligence” OR “machine learning” OR “deep learning” OR “computer-aided detection” OR “neural network”) AND (“endoscopy” OR “colonoscopy” OR “gastroenterology” OR “gastrointestinal”). The search was limited to active grants with project periods overlapping fiscal years 2020 through 2025.

Data Extraction and Categorization

For each identified grant, we extracted: (a) project title and abstract; (b) principal investigator name and credentials; (c) organizational affiliation and hosting department; (d) NIH institute and center; (e) funding mechanism; and (f) total project funding. Projects were categorized by hosting department into four categories: (1) medical/surgical departments; (2) engineering/computer science departments; (3) nursing departments; and (4) other departments.

Results

Departmental Distribution

The search identified 23 active grants combining artificial intelligence with endoscopy or gastroenterology-related research. Funded institutions included VA Boston Healthcare System, MD Anderson Cancer Center, Memorial Sloan Kettering Cancer Center, Mayo Clinic, Stanford University, Johns Hopkins University, and others.

Table 1. Distribution of NIH AI-Endoscopy Grants by Hosting Department (FY2020–2025). Medical/Surgical departments include gastroenterology, internal medicine, surgery, and oncology divisions.

Principal Investigator Credentials

No principal investigators held nursing credentials (RN, MSN, DNP, or PhD in Nursing). The majority held MD degrees (n=15, 65.2%) or PhD degrees in non-nursing disciplines (n=8, 34.8%), including biomedical engineering, computer science, and biostatistics. Analysis of co-investigator teams similarly revealed no nursing representation on funded project teams.

Research Focus Areas

The 23 identified grants addressed the following primary focus areas:

• Polyp detection and characterization during colonoscopy (n=12)
• Image analysis and computer vision for endoscopy (n=5)
• Natural language processing for endoscopy reports (n=3)
• Quality metrics and documentation (n=2)
• Sedation optimization (n=1)

No identified grants focused on nursing-specific outcomes such as workflow integration, nurse-AI interaction, implementation science from a nursing perspective, sedation monitoring by nurses, or nursing education for AI technologies.

Discussion

This analysis reveals a striking gap between the development of AI technologies for endoscopy and the involvement of nursing science in that development. Despite nurses serving as the primary operators of endoscopy equipment, managers of procedural sedation, and frontline implementers of AI-assisted systems, not a single identified NIH grant for AI in endoscopy was housed within a nursing department or led by a nurse scientist.

Implications for Implementation

Alert design and timing. Current CADe systems generate visual markers (typically green boxes) and auditory alerts when potential polyps are detected. Without nursing input into alert design, there is risk that alerts may occur at inopportune moments during procedures, compete with existing monitoring alarms, or generate excessive false positives that contribute to desensitization. Studies of CADe implementation have reported false positive rates as high as 60% in some settings (Ladabaum et al., 2023).

Workflow integration. Endoscopy nurses manage multiple concurrent responsibilities including equipment operation, patient monitoring, documentation, and sedation management. AI systems designed without understanding of nursing workflow may create additional cognitive burden rather than reducing workload (Alobayli et al., 2023).

Sedation interface considerations. AI systems for sedation optimization are being developed that provide real-time feedback on procedure progress and anesthesia depth. However, the design of these interfaces without nursing input may overlook critical human factors that affect clinical decision-making during sedation monitoring.

Parallels to EHR Implementation Challenges

The current pattern of AI development for endoscopy mirrors the historical trajectory of electronic health record implementation. As documented extensively in the literature, EHRs developed without adequate clinician input contributed to documentation burden, workflow disruption, and clinician burnout (Downing et al., 2018; Collins et al., 2023). This recognition should serve as a cautionary precedent for AI development in endoscopy.

Opportunities for Nursing Science

The identified funding gap represents both a problem and an opportunity. Nursing science is well-positioned to address research questions that are currently underexplored:

• How do experienced endoscopy nurses integrate AI alerts into existing clinical judgment?
• What training approaches optimize nurse trust in and verification of AI recommendations?
• How does AI-assisted documentation affect nurse-patient interaction during procedures?
• What interface designs minimize cognitive load during high-acuity procedural moments?
• How do AI sedation support systems affect nursing clinical decision-making?

The National Institute of Nursing Research has signaled growing interest in AI and informatics, hosting an annual AI Bootcamp and establishing an AI and Data Science Working Group (NINR, 2025).

Limitations

This analysis has several limitations. First, the search was limited to NIH-funded grants and may not capture AI research funded by other agencies, industry, or international sources. Second, the search terms may have missed relevant grants using alternative terminology. Third, nursing involvement may exist at the co-investigator or consultant level that is not readily apparent in RePORTER listings. Fourth, NINR may fund AI-related nursing research under broader categories not captured by endoscopy-specific search terms. Fifth, the analysis focused specifically on endoscopy and gastroenterology applications and may not generalize to other clinical domains.

Implications for Practice and Research

For Nursing Researchers: The identified funding gap suggests an underexplored research domain where nursing science could make significant contributions. Nurse scientists should consider framing AI research within established nursing science frameworks such as implementation science, human factors, and patient outcomes research.

For Healthcare Systems: Healthcare organizations evaluating AI technologies for endoscopy should ensure nursing representation on evaluation committees. Including clinical nurse specialists and nurse informaticists in AI adoption decisions may facilitate smoother implementation.

For AI Developers: Technology developers should partner with nursing programs early in the development cycle. The insights from experienced endoscopy nurses regarding actual procedural workflows, sedation monitoring patterns, and equipment operation sequences may save years of iteration and reduce failed pilot implementations.

For Professional Organizations: Professional nursing organizations including SGNA, ANA, and specialty nursing societies should advocate for dedicated AI-in-nursing funding streams. The American Association of Colleges of Nursing has released core competencies identifying informatics and emerging technologies as critical to professional practice (AACN, 2021).

Conclusion

This analysis identified a significant gap between the development of AI technologies for gastroenterology and endoscopy and the involvement of nursing science in that development. Of 23 NIH grants combining AI and endoscopy research between fiscal years 2020–2025, zero were housed within nursing departments and none were led by nurse scientists.

This finding is particularly concerning given that nurses serve as the primary operators of endoscopy equipment, managers of procedural sedation, and frontline implementers of AI-assisted technologies. The historical experience of electronic health record implementation demonstrates the risks of developing clinical technologies without adequate input from end users.

Strategic investment in nurse-led AI research is needed to ensure that technologies for endoscopy are developed with clinical nursing expertise from inception. The nursing profession has an opportunity to shape the future of AI in healthcare rather than remaining passive recipients of technologies designed without nursing input.