Visual Abstractを1枚で作る

Create a Visual Abstract in BMJ journal style for a clinical trial. 
Layout: left sidebar (20% width, dark blue #005EB8) with study name 
and journal logo area. Main content (80%) with white background.

Content flow top to bottom:
- Study name: "CEREBRO-PROTECT Trial" (bold, large)
- Design icon: "Randomized Controlled Trial, 24 countries"
- Population icon: "1,850 patients, age 55-80, prior TIA"
- Intervention: "XR-2847 200mg/day + standard care vs standard care alone"
- Primary outcome with large number: "42% reduction in stroke"
  (HR 0.58, 95% CI 0.44-0.76)
- Bar chart: Intervention 8.2% vs Control 14.1%

Use medical icons (stethoscope, brain, pill). 
Color scheme: BMJ blue, teal #00A9CE, magenta #DC004E for emphasis.
Clean, professional, infographic style. High resolution.

Create a Visual Abstract in JAMA journal style. Clean, minimal design.

Top section (red banner #D22630):
"QUESTION: Does AI-assisted CT screening improve early lung cancer detection?"

Middle sections (gray boxes on white):
- Population: "3,240 adults, age 50-75, smoking ≥30 pack-years"
- Setting: "48 centers in US, Canada, UK, Germany"
- Intervention: "AI-assisted vs standard CT reading"

Bottom section (large, bold):
"FINDING: AI screening improved Stage I detection by 38% 
(RR 1.38, 95% CI 1.15-1.67) and reduced false positives by 24%"

JAMA color scheme: red, gray, beige, gold accents.
Question-and-answer format. Professional medical journal style.

Create a Visual Abstract in New England Journal of Medicine style. 
Emphasis on data visualization.

Header: "EMPA-KIDNEY Trial" in NEJM red (#C80000)

Center: Kaplan-Meier survival curve showing two lines diverging:
- Empagliflozin group (blue line, lower) 
- Placebo group (red line, higher)
- X-axis: "Months" (0-36)
- Y-axis: "Kidney Disease Progression (%)" (0-30%)
- HR 0.72 (95% CI 0.64-0.82) in bold

Bottom: Key secondary outcomes in 3 columns:
- eGFR decline: -2.1 vs -3.5 ml/min/year
- Hospitalization: 18% reduction
- Cardiovascular death: NS

White background, minimal decoration, data-centric design.