The nursing process
Case study: wound care visualisation with AI
Year 2 Adult Nursing wound assessment — bridging tissue-layer anatomy, pressure ulcer staging, and NICE CG179 pathways.
Cross-sectional diagrams + Mermaid flowcharts + audio narration — verified by the educator before classroom use.
Check anatomical proportions & diverse skin tones — pair AI schematics with consented clinical photographs (e.g. Mind the Gap).
This case study describes work from 2025, using the tools available then. Some, including DALL-E 3, have since been retired. For current image tools see GenAI tools, and for safe use see Images.
Author: Lincoln Gombedza, RN (LD)
Institution Type: University nursing programme, NHS practice education
Year: 2025
Student Level: Year 2 Adult Nursing students
Topic: Wound assessment and management
Context
Teaching wound care assessment and management requires students to understand complex anatomical structures, healing stages, and clinical decision-making pathways. Traditional approaches rely heavily on textbook diagrams and static images which may not adequately represent the diversity of wound presentations students will encounter in practice.
Challenge
- Limited access to diverse wound care images across healing stages and skin tones
- Static textbook diagrams do not show progression or tissue depth clearly
- Dense text care pathways are hard to follow during simulation
- Copyright restrictions limit reuse of clinical photographs
- “I can’t visualise what a Category 3 pressure ulcer actually looks like from the description.”
- “The care pathway in the textbook is confusing — too much text.”
- “I need to see the layers of the skin to understand wound depth.”
Solution: using AI images and diagrams
Phase 1: Anatomical Visualization
Tool used at the time: an AI image generator (DALL-E 3, now retired)
Cross-Sectional Tissue Depth Diagram
Create a detailed cross-sectional anatomical diagram showing the layers of
human skin (epidermis, dermis, subcutaneous tissue) with a Category 3 pressure
ulcer. Show full-thickness skin loss extending into the subcutaneous fat,
with no exposed bone, tendon or muscle. Use medical illustration
style with clear labels. Suitable for nursing education.
Anatomical & Equity Verification
- Output produced: Normal skin anatomy, Categories 1–4 and unstageable pressure ulcers, and inflammatory/proliferative/remodelling phases.
- Educator check: Verify tissue depth boundaries against EPUAP/NPIAP/PPPIA staging definitions and re-label any garbled AI text overlays in Canva before sharing.
Phase 2: Care Pathway Visualization
Tool used at the time: a general AI chatbot with Mermaid diagram generation
Process:
- Input NICE pressure ulcer management guidelines
- Ask the AI tool to create a visual flowchart
- Generate Mermaid diagram code
- Render into visual care pathway
Guideline-to-Mermaid Flowchart Generator
Convert these NICE guidelines for pressure ulcer assessment into a visual
flowchart using Mermaid diagram syntax. Include decision points for:
risk assessment, wound staging, intervention selection, and reassessment
timeframes.
Phase 3: Combined learning resource
Combined AI outputs, checked by the educator, into a learning module:
- Anatomical diagrams (text-to-image AI)
- Care pathway flowcharts (AI-generated Mermaid diagrams)
- Narrated explanations (text-to-speech AI for accessibility)
- Interactive quizzes (AI-generated questions with visual elements)
Implementation
Compare & Critique
Students review AI-generated anatomical schematics alongside consented clinical photographs and audit AI accuracy against anatomy textbooks.
Simulation & Documentation
Students use the visual care pathway flowcharts in simulation, practise wound assessment, and document findings in person-centred language.
Peer Audit & Bias Check
Students generate and critique visual aids against NICE CG179, identifying anatomical errors and skin-tone biases in AI image models.
Assessment Component:
Students create a wound care resource package combining text, images and audio, including:
- AI-generated visual aids (with prompt documentation)
- Critical commentary on AI output accuracy
- Verification against NICE guidelines and evidence-based sources
- Reflection on learning through creating resources in different formats
Tools Used
- An AI image generator (DALL-E 3 at the time) — anatomical diagram generation
- A general AI chatbot — care pathway visualisation and Mermaid code
- NotebookLM — Synthesising NICE guidelines into audio summaries
- Canva — Assembling final learning resources
- H5P — Creating interactive wound staging quizzes
Outcomes
Quantitative Results:
- 95% of students reported improved understanding of wound anatomy
- Formative assessment scores increased by average of 18%
- Student engagement metrics showed 40% more time spent with learning materials
- Accessibility: Audio narrations benefited 12% of students with dyslexia
Qualitative Feedback:
"The AI-generated diagrams helped me finally understand the difference between partial and full thickness wounds. I could see the layers!" — Year 2 student
"Creating my own visual guide using AI made me think really carefully about accuracy. I had to fact-check everything against NICE." — Year 2 student
"The care pathway flowchart is so much clearer than reading paragraphs of text. I can follow the decision points now." — Practice educator
Skills Developed:
- Anatomical knowledge of skin and wound structures
- Critical evaluation of AI-generated content
- Prompt engineering for clinical education
- Evidence-based practice (verifying AI outputs)
- Creating resources in different formats
Key Learnings
What Worked Well:
- AI democratized visual creation — Students without graphic design skills could create professional diagrams
- Multiple iterations were valuable — Refining prompts taught precision and clinical accuracy
- Combined with real images — AI diagrams complemented (not replaced) genuine clinical photos
- Critical evaluation was essential — Students caught anatomical errors in AI outputs
Challenges Encountered:
- Anatomical inaccuracies — Some AI-generated diagrams had incorrect anatomical proportions
- Solution: Created checklist for verifying anatomy against textbooks
- Copyright confusion — Students unsure about AI-generated image ownership
- Solution: Teaching session on creative commons and AI-generated content
- Over-reliance risk — Some students wanted to skip traditional learning
- Solution: Positioned AI as supplement, not replacement for core texts
Tips for Other Educators:
- Start with simple diagrams — Build confidence before complex visualizations
- Teach prompt engineering — Good prompts = better outputs
- Always verify against authoritative sources — NICE, BNF, anatomy textbooks
- Use AI to supplement, not substitute — Real clinical experience is irreplaceable
- Document your prompts - Helps with reproducibility and learning
NMC Alignment
This case study supports the following NMC Standards of Proficiency for Registered Nurses:
Platform 1: Being an accountable professional
- 1.9 Demonstrate the skills and abilities required to develop, manage and maintain appropriate relationships with people, their families, carers and colleagues
- 1.18 Demonstrate the knowledge and skills required to support people at all stages of life
Platform 2: Promoting health and preventing ill health
- 2.2 Protect health through understanding and applying the principles of infection prevention and control including communicable disease surveillance
Platform 3: Assessing needs and planning care
- 3.5 Demonstrate the ability to accurately process all information gathered during the assessment process
Platform 4: Providing and evaluating care
- 4.2 Work in partnership with people to encourage shared decision making
- 4.14 Understand and apply the principles of safe and effective administration and optimization of medicines
Annexe B: Nursing procedures
- B.11 Observing and assessing the need for intervention for people with wounds
Resources
AI tools
Clinical Guidelines:
Educational Tools:
- H5P — Interactive content creation
- Mermaid Live Editor — Diagram creation
Adaptation Ideas
This approach can be adapted for other clinical topics:
- Medication pharmacokinetics — Visualise drug absorption, distribution, metabolism, excretion
- Cardiovascular anatomy — Create heart diagrams showing blood flow pathways
- Respiratory assessment — Visualise breath sounds and chest examination findings
- Care pathways — Any clinical guideline can be converted to flowcharts
Further Reading
- Lacković, N. (2020). Thinking with digital images
- NICE (2014). Pressure ulcers: prevention and management (CG179)
- NHS England. Pressure ulcer prevention and management framework
Share Your Experience
Have you used generative AI for wound care or clinical visualisation?
Share your experience, teaching ideas, or feedback using the NCD AI Nursing Toolkit Feedback and Ideas form, or scan the QR code.
Open Feedback & Ideas Form →
