A detailed breakdown of 23 active NIH-funded projects at the intersection of artificial intelligence and endoscopy — examining institutional affiliations, departmental homes, PI credentials, funding levels, and the absence of nursing-led principal investigators.
This data brief provides a detailed companion analysis to the MilenAi research paper “Zero AI Grants in Nursing Departments”. Using the NIH RePORTER database API, we identified and categorized all 23 active grants combining artificial intelligence with endoscopy or gastroenterology-related research across fiscal years 2020–2025.
The findings reveal a systematic exclusion pattern: while AI-endoscopy research receives substantial federal funding across 17 institutions, every single project is housed outside of nursing departments, and no principal investigator holds nursing credentials.
The 23 grants were categorized by the department hosting the principal investigator. The distribution reveals a clear concentration in two sectors—medical/surgical departments and engineering/computer science—with a complete absence of nursing representation.
| Department Category | Grants (n) | Percentage | Unique Institutions |
|---|---|---|---|
| Medical / Surgical Departments | 14 | 60.9% | 10 |
| Engineering / Computer Science | 9 | 39.1% | 7 |
| Nursing Departments | 0 | 0.0% | 0 |
| Total | 23 | 100.0% | 17 |
Table 1. Distribution of NIH AI-Endoscopy Grants by Department (FY2020–2025). Medical/Surgical includes gastroenterology, internal medicine, surgery, and oncology divisions. Engineering/CS includes biomedical engineering, electrical engineering, and computer science.
Zero out of 23 NIH-funded AI-endoscopy grants are housed in nursing departments. The complete absence is not a funding gap—it is a structural exclusion of the professionals who will implement these technologies at the bedside.
Analysis of principal investigator terminal degrees confirms the absence of nursing scientists from AI-endoscopy research leadership.
| PI Credential | n | % |
|---|---|---|
| MD (Gastroenterology, Surgery, Oncology) | 15 | 65.2% |
| PhD (Biomedical Engineering, CS, Biostatistics) | 8 | 34.8% |
| Nursing Credentials (RN, MSN, DNP, PhD-Nursing) | 0 | 0.0% |
| Total Principal Investigators | 23 | 100.0% |
Table 2. Principal Investigator Credentials for NIH AI-Endoscopy Grants. Co-investigator analysis also revealed no nursing representation on funded project teams.
The 23 grants span 17 unique institutions, including some of the nation’s most prominent academic medical centers and VA healthcare systems:
Every funded institution hosts the grant within a medical, surgical, or engineering department. No institution houses an AI-endoscopy grant within its school of nursing or nursing research center.
The 23 grants address five primary focus areas. Notably, the dominant focus is on polyp detection—a technology that directly impacts endoscopy nursing workflows—yet no projects examine the nursing dimension of implementation.
| Research Focus | Grants (n) | % |
|---|---|---|
| Polyp detection and characterization during colonoscopy | 12 | 52.2% |
| Image analysis and computer vision for endoscopy | 5 | 21.7% |
| Natural language processing for endoscopy reports | 3 | 13.0% |
| Quality metrics and documentation | 2 | 8.7% |
| Sedation optimization | 1 | 4.3% |
| Total | 23 | 100.0% |
Table 3. Primary Research Focus Areas of NIH AI-Endoscopy Grants. No grants focused on nursing-specific outcomes such as workflow integration, nurse-AI interaction, or nursing education for AI technologies.
Zero funded projects address: nurse-AI workflow integration, alert fatigue in endoscopy nursing, sedation monitoring by nursing staff, nursing education for AI systems, or implementation science from a nursing perspective. These are the questions that will determine whether AI in endoscopy succeeds or fails in practice.
Nurses are the primary operators. In the endoscopy suite, nurses manage pre-procedure preparation, administer and monitor procedural sedation, operate endoscopy equipment, monitor patient vital signs, assist with tissue acquisition and polypectomy, and manage post-procedure recovery. AI systems designed without understanding their workflows risk repeating the documented failures of electronic health record implementation.
Implementation requires nursing insight. Current CADe systems generate visual markers and auditory alerts during procedures. Without nursing input, alerts may occur at inopportune moments, compete with sedation monitoring alarms, or generate false positives that lead to desensitization. Studies have reported false positive rates as high as 60% in pragmatic implementation settings (Ladabaum et al., 2023).
The EHR lesson is clear. Technologies developed without frontline clinician input contributed to documentation burden, workflow disruption, and clinician burnout. The AI development pipeline for endoscopy is following the same trajectory.
Data were obtained from the NIH RePORTER database API (api.reporter.nih.gov). Search query: (“artificial intelligence” OR “machine learning” OR “deep learning” OR “computer-aided detection” OR “neural network”) AND (“endoscopy” OR “colonoscopy” OR “gastroenterology” OR “gastrointestinal”). Limited to active grants, FY2020–2025. Full methodology available in the companion research paper.
This data brief is a companion to the comprehensive analysis: “Zero AI Grants in Nursing Departments: A Systemic Blind Spot in Endoscopy Innovation (FY2020–2025)”, which provides the complete background, literature review, discussion, and implications for practice and research.