Skip to main content

The Promise of Big Data for Lung Disease Patients

The launch of an open-source repository of thousands of lung scans invites a new era of AI-assisted research into rare diseases.

screenshot of a blue lung scan and the website for the OSIC repository

Sometimes the trouble with big data is that it’s so big.

The information researchers want is out there, it’s just spread across the globe, tucked into different medical record platforms, owned by various universities and private companies and located in different countries with their own regulations for how to handle patient information. When the data involve rare disease, such as interstitial lung disease, gathering it can be even more difficult because there’s less to be had.

But after several years of working through these obstacles, the Open Source Imaging Consortium (OSIC) made a happy announcement this month with the launch of the OSIC Data Repository, now filling up with thousands of high-resolution computed tomography (HRCT) lung scans.

Alongside numerous partners, including CSL Behring, the consortium celebrated the September 7 launch and the opportunity the large database creates for research aided by artificial intelligence (AI). The data will enable AI-based applications, precision medicine and improved patient care, said Dr. Simon Walsh of the National Heart Institute at the Imperial College London, who is the OSIC radiology lead. The organization wants to “make radical progress for patients and caregivers,” OSIC’s Executive Director Elizabeth Estes told patients in a virtual meeting last week.

The first-of-its-kind database for interstitial lung diseases (ILD) is expected to house 15,000 scans by early 2022. Previous research relied on cohorts of several hundred at most, said Dr. Kevin Brown of National Jewish Health in Denver, Colorado, who is the OSIC pulmonology lead. The open-source data base includes accompanying clinical data that has been anonymized to protect patient privacy.

“This ability to collect and organize anonymized imaging and clinical data from across the world is the future of clinical science,” Brown said.

Having lung scans paired with clinical info about how the patient’s disease progressed will help doctors predict the best care for patients whose illnesses have similar characteristics.

Being able to reliably predict how pulmonary fibrosis will progress in an individual patient would allow doctors to initiate appropriate treatment at the earliest opportunity and slow disease progression,” Walsh said. “It remains one of the most urgent challenges for effective management for patients with fibrotic lung disease.”     

CSL Behring, which develops medicines for rare and serious diseases, partnered with OSIC to support cutting-edge research into interstitial lung diseases. ILD includes a variety of conditions that cause fibrosis – scarring – and then stiffness in the lungs.

“It’s fantastic progress. The dynamic has changed,” said CSL Behring Lars Groenke, Vice President of R&D/Co-Lead of the Respiratory Therapy Area who serves on OSIC’s Board of Directors.