Diagnosing Cancer Using Artificial Intelligence

To define the behavior of a patient’s cancer, pathologists often look at the interface between cancerous and healthy cells. This interface is called the “Pattern of Invasion.” By defining a patient’s Pattern of Invasion, physicians can design patient-specific treatment plans to improve patient outcomes. However, the identification of cancer cells can vary among different pathologists. To make the identification of cancerous cells more standardized, computers can be “trained” to recognize different types of cells. We hope to use this technology to improve the speed and accuracy of cancer diagnoses, and provide specialized diagnostic care in remote places.

Read the Abstract