Machine Learning to Advance Research

Ambra Health’s www.ambriscloud.com archive provides instant access to secondary versions of medical images and priors of primary data when available to the cloud-based medical image management suite.

In order for researchers to use MRI, CT, or PET scans for machine learning experiments, researchers typically must remove Protected Health Information (PHI) from medical imaging data, including the patient’s name and date of birth in Digital Imaging and Communications in Medicine (DICOM) tags.

Google Cloud’s Healthcare API and the Ambra Health Cloud PACS solution for Google Cloud https://cloud.google.com allows researchers to turn data into insights by easily de-identifying patient medical imaging data for use in research studies.

A major academic medical center is among the first institutions to leverage the shared capability. The center wanted to use a more secure method to view medical imaging data and wanted to collaborate with fellow researchers during the development of machine learning tools to improve patient care. The use of the new technology together has enabled the medical center to use an open source data set when collaborating globally.

“The future of healthcare depends on the ability to more securely store, curate, view, and share research data,” said Morris Panner, CEO, for Ambra Health. “The collaboration with Google Cloud offers opportunities for providers to work together more effectively and furthers deep learning capabilities for medical imaging.”