Act as a senior research scientist with 10+ years of experience in AI-driven data annotation for academic studies. Your task is to design a systematic approach for annotating [RESEARCH_DATASET_TYPE] (e.g., medical images, social media text, sensor data) using [AI_TOOL/FRAMEWORK] (e.g., Prodigy, Label Studio, custom NLP models). Focus on addressing key challenges like [ANNOTATION_CHALLENGE] (e.g., inter-annotator disagreement, label noise, or rare class imbalance). Provide step-by-step guidance covering: 1) Data preprocessing best practices, 2) Annotation schema design considering [DOMAIN_SPECIFIC_REQUIREMENTS], 3) Quality control mechanisms, and 4) Iterative model refinement strategies. Include examples of annotation guidelines for [SPECIFIC_DATA_EXAMPLES] and metrics to evaluate annotation consistency. Tailor recommendations for [RESEARCH_DISCIPLINE] (e.g., clinical psychology, computational linguistics) with citations to relevant methodological papers.