cardiology-visual-system

What this skill does

# Cardiology Visual System

A unified meta-skill that intelligently routes visual content requests to the optimal tool. No more switching between Napkin.ai, NotebookLM, or other subscriptions—everything happens in Claude Code.

## Quick Reference: What Tool Does What

| You Ask For | Tool Used | Output |
|-------------|-----------|--------|
| Blog header, lifestyle photo, patient scenario | **Fal.ai** | PNG image |
| Infographic, explainer graphic, medical illustration | **Gemini** | PNG/JPG image |
| Flowchart, treatment algorithm, clinical pathway | **Mermaid** | SVG/PNG diagram |
| Slide deck, presentation | **Marp** | PPTX/PDF/HTML slides |
| Data chart, trial results, trends over time | **Plotly** | Interactive HTML or PNG |
| Interactive explainer, dashboard | **React Artifact** | Interactive HTML |
| Quick visualization prototype, exploratory data viz | **LIDA** ⚠️ | PNG + Code (prototype only) |

## Automatic Routing Logic

When you ask for visuals, I determine the best tool by analyzing your request:

### → Route to Fal.ai (Stock/Human Imagery)
Keywords: `blog image`, `header`, `hero image`, `lifestyle`, `patient photo`, `stock`, `person`, `family`, `emotional`, `scenario`

**Best for:**
- Blog post headers
- Patient experience illustrations
- Lifestyle/wellness imagery
- Emotional/human-centered scenes
- Recovery and hope imagery

**NOT for:** Medical devices, ECGs, diagrams, data, or text-heavy content

### → Route to Gemini (Infographics & Medical Illustrations)
Keywords: `infographic`, `explainer`, `illustration`, `visual summary`, `concept diagram`, `icons`, `steps`, `process visual`, `simplified`, `educational graphic`

**Best for:**
- Infographics (like Napkin.ai produces)
- Medical concept illustrations
- Simplified explainer graphics
- Educational visuals with icons
- Text-in-image content
- Visual summaries of articles

### → Route to Mermaid (Diagrams & Flowcharts)
Keywords: `flowchart`, `algorithm`, `pathway`, `decision tree`, `sequence`, `timeline`, `process flow`, `treatment algorithm`, `diagnostic pathway`, `workflow`

**Best for:**
- Clinical decision trees
- Treatment algorithms
- Diagnostic pathways
- Process workflows
- Organizational charts
- Sequence diagrams
- Gantt charts for timelines

### → Route to Marp (Slide Decks)
Keywords: `slides`, `presentation`, `deck`, `powerpoint`, `lecture`, `talk`, `keynote`

**Best for:**
- Conference presentations
- Educational lectures
- Patient education slides
- Grand rounds presentations
- CME content

### → Route to Plotly (Data Visualization)
Keywords: `chart`, `graph`, `plot`, `data`, `statistics`, `trial results`, `forest plot`, `trends`, `comparison`, `survival curve`, `Kaplan-Meier`, `bar chart`, `line graph`, `scatter`

**Best for:**
- Clinical trial results
- Statistical comparisons
- Trends over time
- Forest plots
- Survival curves
- Before/after data
- Multi-study comparisons

### → Route to React Artifact (Interactive)
Keywords: `interactive`, `dashboard`, `calculator`, `tool`, `widget`, `animated`, `explorable`

**Best for:**
- Risk calculators
- Interactive explainers
- Animated diagrams
- Patient education tools
- Explorable explanations

### → Route to LIDA (Quick Prototyping) ⚠️ PROTOTYPING ONLY
Keywords: `quick`, `prototype`, `exploratory`, `rough draft`, `multiple options`, `try`, `experiment`, `brainstorm visualization`

**⚠️ CRITICAL LIMITATIONS:**
- **PROTOTYPING ONLY** - NOT for publication or patient-facing materials
- Quality varies - ALWAYS review for medical accuracy
- Works best with ≤10 data columns
- No specialized medical charts (true forest plots, Kaplan-Meier)
- Requires manual validation before ANY use

**Best for:**
- Quick exploratory data visualization ("what does this data show?")
- Generating multiple visualization candidates
- Brainstorming chart types for new data
- Internal research review only

**NOT for:**
- Publication-ready charts → Use Plotly instead
- Patient-facing materials → Use production tools
- Regulatory submissions → Never
- Final blog posts → Use Plotly or Gemini

**When in doubt, use Plotly for data visualization instead of LIDA.**

---

## Tool 1: Fal.ai (Blog Imagery)

### Setup
```bash
export FAL_KEY="your-fal-api-key"
# Get key from: https://fal.ai/dashboard/keys
```

### Usage
```bash
python scripts/fal_image.py "A 55-year-old man experiencing chest pain at work" --output hero.png
```

### Models
| Model | Cost | Best For |
|-------|------|----------|
| `fal-ai/recraft-v3` | $0.04 | **Default** - Best quality |
| `fal-ai/flux-pro/v1.1` | $0.04 | Photorealism |
| `fal-ai/flux/schnell` | $0.003 | Fast/cheap drafts |

### What to Generate vs Not Generate

✅ **GENERATE:**
- Patient symptoms/experiences (chest pain, shortness of breath)
- Lifestyle scenes (exercise, healthy cooking)
- Doctor-patient conversations
- Family support moments
- Recovery celebrations

❌ **DO NOT GENERATE:**
- Medical devices (pacemakers, stents, valves)
- Clinical imagery (ECGs, angiograms, OR scenes)
- Anatomical diagrams
- Medications

---

## Tool 2: Gemini (Infographics)

### Setup
```bash
export GEMINI_API_KEY="your-gemini-api-key"
```

### Usage
```python
python scripts/gemini_infographic.py \
  --topic "Heart Failure Stages" \
  --style "minimalist medical" \
  --output hf_stages.jpg
```

### Prompting for Medical Infographics

**Structure your prompt:**
```
Create a [STYLE] infographic showing [TOPIC].

Include:
- [Key point 1]
- [Key point 2]
- [Key point 3]

Style: [clean/minimalist/modern medical], use icons, clear hierarchy, 
professional color palette (blues, teals for medical), easy to read text
```

**Example prompts:**

1. **Disease Progression:**
   > "Create a minimalist medical infographic showing the 4 stages of heart failure (A, B, C, D). Use icons for each stage, show progression with arrows, include brief descriptions. Clean layout, medical blue color scheme."

2. **Treatment Comparison:**
   > "Create an infographic comparing medication vs intervention for AFib. Two columns, icons for each approach, bullet points for pros/cons. Modern medical style."

3. **Risk Factor Summary:**
   > "Create a visual summary of 7 modifiable risk factors for heart disease. Icon for each factor, clean grid layout, actionable tips. Professional medical illustration style."

---

## Tool 3: Mermaid (Diagrams)

You have Mermaid Chart MCP connected. Use it for structured diagrams.

### Common Cardiology Diagram Types

**1. Treatment Algorithm:**
```mermaid
flowchart TD
    A[Acute Chest Pain] --> B{STEMI?}
    B -->|Yes| C[Primary PCI < 90 min]
    B -->|No| D{High-risk NSTEMI?}
    D -->|Yes| E[Early invasive < 24h]
    D -->|No| F[Ischemia-guided strategy]
```

**2. Clinical Pathway:**
```mermaid
flowchart LR
    A[Diagnosis] --> B[Risk Stratification]
    B --> C[Treatment Selection]
    C --> D[Follow-up Protocol]
```

**3. Diagnostic Decision Tree:**
```mermaid
flowchart TD
    A[Dyspnea] --> B{BNP elevated?}
    B -->|Yes| C{Echo findings?}
    B -->|No| D[Consider other causes]
    C -->|HFrEF| E[GDMT initiation]
    C -->|HFpEF| F[Diuretics + address comorbidities]
```

**4. Timeline (Gantt):**
```mermaid
gantt
    title Post-MI Care Timeline
    dateFormat  YYYY-MM-DD
    section Acute
    Hospital stay        :a1, 2024-01-01, 5d
    section Recovery
    Cardiac rehab        :a2, after a1, 12w
    section Long-term
    Medication titration :a3, after a1, 6m
```

---

## Tool 4: Marp (Slides)

### Setup
```bash
npm install -g @marp-team/marp-cli
```

### Usage

1. I write Markdown with Marp syntax
2. Save as `presentation.md`
3. Convert:
```bash
marp presentation.md --pptx           # PowerPoint
marp presentation.md --pdf            # PDF
marp presentation.md -o slides.html   # HTML
```

### Marp Template for Medical Slides

```markdown
---
marp: true
theme: default
paginate: true
backgroundColor: #ffffff
color: #333333
---

# Heart Failure Management
## Modern Approaches in 2024

Dr. [Your Name]
Interventional Cardiology

---

# Agenda