Mode
Text Size
Log in / Sign up

Scoping review maps knowledge graph technology research landscape in nursing healthcare fieldsNursing’s New Brain: How AI Maps Are Quietly Reshaping Patient Care

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway
Note that knowledge graph research in nursing is dominated by education scenarios with scarce clinical validation.

This scoping review examines the research landscape of knowledge graph applications in nursing across 30 studies in healthcare field settings. The authors categorize the evidence maturity framework distribution, identifying that 12 studies (40%) are in the Construction Phase, 6 studies (20%) in the System Performance Evaluation Phase, 8 studies (26.7%) in the Usability Evaluation Phase, 1 study (3.3%) in the Preliminary Application Phase, and 3 studies (10%) in the Application Phase.

Regarding research focus distribution, the review indicates that studies on nursing education scenarios dominated the literature. Conversely, studies on clinical nursing scenarios were relatively scarce, highlighting a distinct domain bias within the existing body of work.

The authors acknowledge several limitations, including the lack of literature that systematically sorts out the overall research landscape from a macro perspective. Furthermore, there is no objective validation of the clinical application effects of these technologies. The review concludes that these insights should inform subsequent in-depth research and practical applications.

Imagine a nurse on a busy ward, needing to find the best care plan for a patient with a complex condition. In the past, this meant flipping through thick manuals or searching endless databases. Now, a new tool is emerging that could act like a smart map, connecting all the right information in seconds.

This tool is called a knowledge graph. Think of it as a visual web that links symptoms, treatments, and research in one place. A new review in Frontiers in Medicine looks at how ready this technology is for nursing.

Nursing is a knowledge-heavy profession. Nurses must remember drug interactions, care protocols, and patient histories—all while under pressure. Mistakes can happen when information is scattered or hard to find.

Knowledge graphs aim to solve this. They organize data like a family tree, showing how different pieces of information connect. For example, a graph could link a patient’s diabetes diagnosis to specific diet advice, warning signs, and relevant research studies.

But is this technology ready for the real world? That’s what this review set out to find.

The Old Way vs. The New Way

Traditionally, nurses rely on textbooks, guidelines, and digital libraries. These are static. If a new study comes out, the book doesn’t update itself.

Knowledge graphs are dynamic. They can grow and change as new evidence appears. They can also personalize information. Instead of a generic guide, a graph could tailor advice based on a patient’s specific age, health history, and even genetics.

But here’s the twist: most current research isn’t testing this in real hospitals yet. The review found that most studies are still building the basic structure, not seeing if it actually helps patients.

Think of a knowledge graph like a city map. Old guidelines are like a paper map—useful but fixed. A knowledge graph is like Google Maps with live traffic. It shows the best route based on current conditions.

In nursing, the “traffic” is patient data. The graph connects dots: a symptom leads to a possible diagnosis, which links to a treatment, which connects to the latest research. This helps nurses see the whole picture quickly.

The review found studies using these graphs for education, disease management, and even predicting risks. For example, one graph might help a nurse student learn how heart failure progresses by showing all related symptoms and treatments in one visual web.

Researchers from China and other institutions reviewed 30 studies on knowledge graphs in nursing. They used a standard method (the Arksey framework) to find and analyze all relevant papers. They then sorted these studies into five stages, from building the graph to actually using it in practice.

The big finding? We’re still in the early days. Of the 30 studies, 12 (40%) were just building the basic graph structure. Only 3 studies (10%) had reached the final stage of real-world application.

Most research focused on nursing education. This makes sense—teaching is a safe place to test new tech. But there’s a gap: very few studies tested these graphs in actual clinical care, like in a hospital ward.

Translation barriers exist. The review notes that moving from a working graph to a tool nurses use daily is a big leap. Data standards are also lacking; different hospitals use different systems, making it hard to share these graphs.

But there’s a catch.

The authors suggest that knowledge graphs in nursing are like a promising prototype. They work in theory and early tests, but they need more real-world trials. The focus should shift from just building graphs to testing if they improve patient outcomes and nurse efficiency.

If you’re a nurse or a patient, this technology isn’t available in your hospital tomorrow. It’s still in research and development. However, it points to a future where nursing care is more data-driven and personalized.

If you’re a nurse, you might see this tech in continuing education or training modules soon. For patients, it could mean more tailored care plans down the road. But for now, it’s a tool to watch, not one to ask for at your next appointment.

This review only included 30 studies, mostly from China. The evidence is early-stage, and many studies were small or focused on theory. We don’t yet have large trials showing these graphs improve patient health or reduce errors in busy hospitals.

Next steps are clear: researchers need to design larger studies that test knowledge graphs in real clinical settings. This means working with hospitals to integrate these tools into daily workflows. Data standardization is also key—so graphs can be shared across different health systems.

While we wait, the idea of a smart, connected map for nursing knowledge is gaining ground. It’s not a miracle cure, but it could quietly make nursing care safer and smarter.

This doesn’t mean this treatment is available yet.

Study Details

Study typeSystematic review
EvidenceLevel 1
PublishedApr 2026
View Original Abstract ↓
The application of knowledge graph technology in the healthcare field is increasingly in-depth, yet there is a lack of literature that systematically sorts out the overall research landscape of its application in nursing from a macro perspective. This study aimed to systematically depict the research panorama of knowledge graph applications in nursing, stratify existing studies by stages through the construction of an exploratory evidence maturity framework, reveal structural gaps and translation barriers, and provide insights for subsequent in-depth research and practical applications. This study adopted the Arksey scoping review reporting framework and followed the PRISMA-ScR checklist for reporting. We systematically searched databases including Wanfang, CNKI, VIP, SinoMed, PubMed, Embase, Web of Science, CINAHL, and the Cochrane Library, and summarized and analyzed the included articles. The research results were comprehensively collated and divided into five phases: the Construction Phase, System Performance Evaluation Phase, Usability Evaluation Phase, Preliminary Application Phase, and Application Phase. A total of 30 studies were included, with methodological studies as the main research design (n = 15), covering themes such as nursing education, disease management, health education, clinical nursing decision support, risk prediction, and psychological support. Analysis based on the evidence maturity framework showed that there were 12 studies (40%) in the knowledge graph Construction Phase, six (20%) in the System Performance Evaluation Phase, eight (26.7%) in the Usability Evaluation Phase, one (3.3%) in the Preliminary Application Phase, and three (10%) in the Application Phase. The research focus exhibited obvious domain bias: studies on nursing education scenarios dominated, while those on clinical nursing scenarios were relatively scarce. Knowledge graph research in nursing is still in the exploratory stage, dominated by evidence on technical construction, with no objective validation of its clinical application effects. Future research should adopt an evidence-driven approach, focusing on clinical application, optimizing study design and advancing data standardization, thereby enabling knowledge graphs to deliver evidence-based support in nursing practice.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.