Skip to main content Skip to footer
by
Ana Maria Sanchez
Business Development Leader – Healthcare & Life Sciences

Pallavi Nigam
Head - Trust and Safety

Healthcare

Navigating trust & safety in healthcare AI: medical annotation

Navigating Trust & Safety in Healthcare AI: Medical Annotation features an interview with Ana Maria Sanchez (Business Development Leader, Healthcare & Life Sciences) and Pallavi Nigam (Head of Trust & Safety, Healthcare & Life Sciences). They discuss the critical role of medical data annotation in enabling safe, accurate, and responsible healthcare AI systems.

Recent Podcasts

Transforming healthcare appeals & grievances with automation and agentic AI
Beyond the prototype: scaling AI for real impact
Simon Kuper on the football world cup and the momentum of sport

Podcast Audio Transcript

Mimi: Welcome to BPM Healthcare and Lifesciences, the podcast where we explore the intersection of healthcare, technology, ethics, and more. I’m your host, and today we’re diving into a topic that’s shaping the future of medicine—trust and safety in healthcare AI, with a focus on medical data annotation.

Joining us are two incredible guests:

  • Ana, a tenured nurse with deep clinical experience.
  • Pallavi, an expert in trust and safety, especially in healthcare data governance.
Let’s get started!

Mimi: Ana, as an experienced nurse, why is trust and safety in healthcare AI so important?

Ana: Trust and safety are the foundation of patient care. Patients share sensitive information expecting confidentiality. If they fear misuse, they may withhold critical details, which can compromise care. From the ground level, I see how essential it is to ensure data is handled ethically and securely—because trust impacts outcomes.

Mimi: Pallavi, from an operations and governance perspective, how do we ensure safety standards are met?

Pallavi: It starts with strict compliance frameworks like HIPAA, but governance goes beyond checklists. We need robust processes—access controls, encryption, audit trails—and continuous monitoring. AI can make procedures more efficient, but only if implemented responsibly. Responsible AI ensures fairness, transparency, and clinical soundness.

Mimi: Technological developments are growing rapidly, especially with AI at the forefront. How does the healthcare industry keep up, and what bridges operations, technology, and governance?

Ana: The bridge is collaboration. Clinicians provide context, technologists bring innovation, and governance ensures safety. For example, in medical annotation, nurses validate AI-suggested labels to maintain accuracy and ethical standards.

Pallavi: Exactly. We use secure cloud platforms, encryption, and human-in-the-loop systems to combine efficiency with oversight. Ethics is the glue—ensuring AI decisions are explainable and bias-free. This approach builds trust while leveraging technology for better outcomes.

Mimi: Ana, what types of data are annotated?

Ana: Three main types:
  • Image annotation – marking tumors or fractures in X-rays.
  • Text annotation – labeling symptoms or medications in clinical notes.
  • Video annotation – tagging steps in surgical procedures.
Mimi: Pallavi, how does this tie back to trust and safety?

Pallavi: Annotation powers AI models for diagnosis and treatment recommendations. But accuracy and compliance are critical—mislabeling or mishandling PHI can lead to harm. That’s why we embed ethical checks and security measures at every step.

Mimi: What are the biggest challenges in HIPAA-compliant annotation?

Pallavi: Security and expertise. PHI must remain protected, and annotators need medical knowledge. We address this through clinician involvement, secure platforms, and continuous audits.

Ana: And accuracy matters—mislabeling can lead to incorrect care. Human oversight is key.

Mimi: What’s next for trust and safety in healthcare AI?

Pallavi: More human-in-the-loop systems, advanced encryption, and AI governance frameworks. We’re also seeing proactive risk management to prevent bias and misinformation.

Ana: And more clinician engagement. AI should complement—not replace—human judgment.

Mimi: Thank you, Ana and Pallavi, for this insightful discussion. Trust and safety aren’t just compliance—they’re about protecting patients and building confidence in technology. Stay tuned for more episodes on how ethics and AI are transforming healthcare.

Explore Infosys BPM’s Healthcare Offering >>