Articles

Computer Vision: Buy or DIY?

Computer Vision: Buy or DIY?

Four reasons why you might need to build your own computer vision solution.

Continue Reading
Vision Foundation Models: When Does Size Matter?

Vision Foundation Models: When Does Size Matter?

Large vision models may seem attractive, but domain-specific models can get you farther.

Continue Reading
Getting Started with Computer Vision for Histopathology

Getting Started with Computer Vision for Histopathology

A reading list for anyone looking to adapt their machine learning expertise to computational pathology.

Continue Reading
Unlock the Potential of Your Proprietary Images with a Domain-Specific Foundation Model

Unlock the Potential of Your Proprietary Images with a Domain-Specific Foundation Model

The benefits of a self-supervised vision model for generalizability, adaptability, and more.

Continue Reading
How to Reduce the Trial-and-Error of Machine Learning Development

How to Reduce the Trial-and-Error of Machine Learning Development

Machine learning projects are complex and iterative with uncertain outcomes. These are some ways to improve the efficacy of ML development.

Continue Reading
Impact AI: 10 Most Downloaded Episodes

Impact AI: 10 Most Downloaded Episodes

The episodes that I highlight below are the 10 most downloaded. The common themes are food and disease: farming, climate risks, food waste, cancer, and medical imaging.

Continue Reading
Pathology AI Pitfalls: Top 10 Challenges in Applying Computer Vision to Histopathology

Pathology AI Pitfalls: Top 10 Challenges in Applying Computer Vision to Histopathology

Common challenges in machine learning for pathology.

Continue Reading
How to Design a Roadmap for a Machine Learning Project

How to Design a Roadmap for a Machine Learning Project

The core components to successful machine learning project.

Continue Reading
Building Robust AI Solutions for Whole Slide Images

Building Robust AI Solutions for Whole Slide Images

Some of the key components for robust models: quality control, generalizable, and properly validated.

Continue Reading
Predicting Molecular Tumor Biomarkers from H&E: A Review of the State-of-the-Art

Predicting Molecular Tumor Biomarkers from H&E: A Review of the State-of-the-Art

New deep learning technology using H&E images has created an alternative path for molecular diagnostics

Continue Reading
5 Ways to Make Histopathology Image Models More Robust to Domain Shifts

5 Ways to Make Histopathology Image Models More Robust to Domain Shifts

Exploring a variety of approaches: stain normalization, color augmentation, adversarial domain adaptation, model adaptation, and finetuning.

Continue Reading
Finding prognostic patterns in gigapixel images

Finding prognostic patterns in gigapixel images

Advances in AI applying deep learning to digital pathology images can stratify patients by risk.

Continue Reading