Webinar | AI in Diagnostics: From Validation to Lifecycle Management

What does it mean to validate an AI-enabled diagnostic tool—and how do we ensure that validation remains meaningful as technologies evolve?

“Validation” is a cornerstone of reliability and trust in diagnostics. Yet, despite its widespread use, the term itself carries different meanings across disciplines, regulatory frameworks, and clinical settings. What constitutes validation for a developer, regulator, laboratory director, or clinician may differ substantially. These often unspoken distinctions can lead to miscommunication, misaligned expectations, and avoidable delays.

As artificial intelligence and machine learning (AI/ML) become increasingly integrated into computational pathology and diagnostics, establishing a shared understanding of validation—and the evidence needed to support it across the technology lifecycle—is increasingly important.

Join MDIC and the Pathology Innovation Collaborative Community (PIcc) for a discussion examining how context-specific validation approaches, regulatory expectations, lifecycle management, and reference data can support the responsible development and clinical implementation of AI-enabled diagnostics.

• Validation: What do we actually mean by validation? Examining discipline-specific definitions, contextual qualifiers, evidence expectations, and practical implementation.
• Regulatory Position: How regulatory science defines and evaluates validation evidence, and how expectations can accommodate evolving AI-enabled technologies.
• Lifecycle Management: How validation extends beyond initial deployment to encompass ongoing monitoring, model changes, performance assessment, and evolving clinical use.
• Reference Data: Why high-quality, representative reference datasets are essential for credible, reproducible, and comparable validation across developers, laboratories, and institutions.

Professionals involved in the development, regulation, evaluation, and clinical implementation of digital health technologies, including AI/ML developers, clinical laboratory directors, regulatory scientists, and business strategists.