Inference by Turing Post

The Future of Cancer Diagnosis: Digital Pathology and AI

Dec 4, 2025
In this conversation, Akash Parvatikar, an AI scientist leading the PathologyMap platform at HistoWiz, dives into the transformative world of digital pathology. He explains how scanning glass slides revolutionizes cancer diagnosis and the potential of AI to enhance diagnostic processes. Akash outlines the exciting future of telepathology, addresses challenges like data bottlenecks, and highlights the importance of explainability in AI tools. He emphasizes that while technology evolves, the role of pathologists remains crucial. A must-listen for anyone interested in the future of medical diagnostics!
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
ADVICE

Introduce Agents In Small Chunks

  • Implement agentic workflows in modular chunks like quality control rather than full diagnosis.
  • Use agents first to standardize image quality and prepare data for downstream AI.
INSIGHT

Annotations Are The Core Bottleneck

  • The biggest bottleneck is scarce, high‑quality annotated data with inter-pathologist disagreement.
  • Models often learn site-specific biases and may not generalize across hospitals or regions.
ANECDOTE

PathologyMap Lets You Compare AIs

  • HistoWiz built PathologyMap to manage and analyze high‑resolution slides and run multiple AI apps side-by-side.
  • Users can test different AI vendors on the same slide and upload custom annotations for comparison.
Get the Snipd Podcast app to discover more snips from this episode
Get the app