You are a professional scientific communication designer specializing in IEEE, Springer, biomedical engineering, medical imaging, and healthcare AI conference posters. Design a publication-quality academic research poster that looks professionally created by a university research laboratory or medical AI research group. The poster must not look AI-generated, promotional, or template-based. It must resemble a human-designed conference poster suitable for submission through a company to an international research conference. Poster specifications Create a single A0 portrait poster (841 × 1189 mm) at print-ready quality (300 DPI equivalent). Use a traditional three-column academic conference layout with a clear left-to-right and top-to-bottom reading flow. Use a white background, dark navy blue section headers, black body text, and subtle gray dividers. Use only two fonts: Helvetica or Arial for headings and body text. Maintain a clean academic aesthetic with strong alignment, consistent spacing, and generous white space. Target composition: - 40–45% graphics and diagrams - 20–25% text - 30–35% white space The poster must be readable from 1–2 meters away. Typography hierarchy - Title: 88–96 pt - Authors: 34–38 pt - Affiliations: 24–28 pt - Section headings: 42–48 pt - Body text: 24–28 pt - Figure captions: 20–22 pt - References: 18–20 pt Do not use decorative fonts, gradients, shadows, glowing effects, 3D icons, or marketing-style graphics. Title block Title: Stroke Detection Using AI and IoT-Based Monitoring: A Multimodal CT Imaging and Wearable Sensor Framework Authors: Nikhil Savita, Jonathan Raposo, Sakshi Kamble, Yashraj Dilip Naik, Louella M. Colaco, Meghana Pai Kane Affiliation: Department of Computer Engineering, Padre Conceição College of Engineering, Verna, Goa, India Prepared for conference poster presentation. Include a clean institutional header with a company logo placeholder on the top right. Column 1: Background, problem statement, objective Background Use concise bullet points. - Stroke is a leading cause of mortality and long-term disability. - Early detection is critical because brain tissue damage progresses rapidly. - Hospital CT diagnosis is accurate but episodic. - Wearable monitoring is continuous but prone to false positives. - Existing systems do not integrate imaging diagnosis with continuous physiological monitoring. Research gap State prominently: No existing platform combines hospital-grade CT image intelligence with continuous wearable physiological monitoring for real-time stroke risk assessment. Objective Develop and evaluate a dual-layer AI-IoT framework that combines CT image classification with continuous wearable monitoring to improve stroke detection reliability and reduce false-positive alerts. Add a small visual callout box titled Clinical Motivation with an icon of brain imaging and wearable monitoring. Column