Skip to main content
Premium Trial:

Request an Annual Quote

Philips, PathAI Collaborate to Develop Deep Learning for Breast Cancer Diagnosis

NEW YORK (360Dx) – Royal Philips said today that it is collaborating with PathAI, a company that develops artificial intelligence for pathology, to develop solutions that make diagnosis of cancer and other diseases more precise and accurate. 

The companies aim to build deep-learning applications in computational pathology that may be applied to massive pathology data sets to better inform diagnostic and treatment decisions. 

The initial focus of the collaboration is on developing applications to automatically detect and quantify cancerous lesions in breast cancer tissue, Philips said. Clinicians diagnose more than 250,000 new breast cancer cases each year in the US alone, it noted. 

Philips noted that historically pathologists have manually reviewed and analyzed tumor tissue slides using a microscope. But a rising shortage of pathologists and an increase in cancer caseloads require digital-pathology solutions and smart-image analysis software that reduce pathologists' routine workload, improve diagnostic accuracy and precision, and reduce error rates, it said. 

The objective of the collaboration is to help patients receive fast and accurate diagnosis, and empower pathologists with decision support tools enabled by artificial intelligence, PathAI CEO Andy Beck said in a statement. 

Identifying the presence or absence of cancer in lymph nodes is a routine and critically important task for a pathologist, but it can be extremely laborious using conventional methods, he said, adding that research indicates that pathologists work faster and more accurately using computational tools. 

Philips has already implemented deep learning in its clinical informatics solutions for radiology, such as Illumeo and IntelliSpace Portal 9.0. With the proliferation of digital pathology and whole-slide imaging, computers may soon learn and unlock the potential of thousands of digital tumor tissue images and related patient data, Philips said.