Skip to main content Skip to secondary navigation

Browse Stanford's available technologies by keyword or collection today by exploring our Techfinder Catalog.

Non-invasive Diagnosis of Skin Cancer

Main content start

Photo credit: Pixel-Shot via Shutterstock.com

A machine learning approach for transforming OCT images, enabling fast, non-invasive skin cancer diagnosis.

About the Technology

Stanford researchers have developed an innovative alignment methodology using Optical Coherence Tomography (OCT) in conjunction with histopathology to diagnose cancer or determine tumor margins. High resolution alignment of OCT volumes and histology sections provides histological information alongside OCT structural and temporal information. This invention provides for the first time, a machine learning approach to directly predict histological images from a given OCT image, producing non-invasive histology-like images with high accuracy and interpretability.

The inventors are currently developing a robust AI dataset to predict histological images from an OCT image. Applying deep learning to transform difficult-to-read OCT images into Hematoxylin and Eosin (H&E) images, will enable fast, non-invasive diagnostics (see figure). 

Figure depicting the vision of the project:  OCT images converted into Hematoxylin and Eosin (H&E) images
Figure: conversion of OCT image to H&E image using the machine learning approach developed by the inventors. (Image credit: the inventors)

 Augmenting capabilities of OCT-based diagnosis through machine learning has the potential to replace the traditional biopsy and can fundamentally improve the cosmetics, speed, and accuracy of diagnosis in multiple types of cancers.

Team Members

Kavita Sarin

Kavita Sarin

Associate Professor of Dermatology

View Stanford Profile

Stanford University Seal with Red Background

Yonatan Winetraub

Instructor, Structural Biology

View Stanford Profile

Pengpeng Wang

Pengpeng Wang

HIT Fund MBA Intern

Related Web Links