Financial fraud detection and healthcare AI share the same core challenge: rare events in noisy, multimodal data where missing a positive case is expensive. This session presents a production-ready framework built from a real credit card fraud system (XGBoost with class weighting achieving 80% recall on 56,961 transactions) combined with hands-on long-read sequencing experience (COSS 2025 workshop: PacBio HiFi for structural variant calling and Iso-Seq isoform detection). The talk shows how ensemble methods, threshold tuning, and GPU parallelism transfer directly to clinical applications: patient phenotyping, rare disease identification, adverse event detection, and payer fraud prevention. Attendees receive reusable code templates for multimodal fusion, bias mitigation strategies, and governance checklists that satisfy regulatory requirements, enabling fast adaptation of proven finance ML pipelines to healthcare production environments.

Keywords: imbalanced learning, fraud detection, long-read sequencing, multimodal AI, clinical deployment, AI governance