OSIC Announces Winners of Pulmonary Fibrosis Progression Challenge

OSIC Announces Winners of Pulmonary Fibrosis Progression Challenge
3.8
(12)

The Open Source Imaging Consortium (OSIC), a nonprofit collaborative group focused on combatting lung diseases, announced the winners of a competition that challenged scientists to design artificial intelligence (AI) programs that predict lung function decline in people with pulmonary fibrosis.

Broadly, AI refers to computer systems that are able to accomplish tasks that typically would require human intelligence.

The OSIC Pulmonary Fibrosis Progression Challenge, which was announced earlier this year, challenged scientists to develop AI programs that could look at a patient’s data and predict how lung function would change over time, particularly in people with idiopathic pulmonary fibrosis (IPF).

The challenge was administered by the data science community platform Kaggle.

The top three researchers to create the highest-performing algorithm were: first place winner Artyom Kulakov (Russia); second place winner Yusuke Ikemoto (Japan); and third place winner Khang Pham (Japan). First place receives $30,000, second place $15,000, and the third place $10,000.

“We saw three innovative approaches from these winners, and we know we can build on this to make meaningful progress in the fight against IPF and other interstitial lung diseases,” Elizabeth Estes, OSIC’s executive director, said in a press release.

Nearly 40,000 entries in the competition were submitted by more than 2,500 global competitors.

To build and train their programs, researchers were given access to a set of data from IPF patients. The data included CT scans (a type of imaging that can be used to visualize scarring in the lungs) at diagnosis, as well as clinical information, including forced vital capacity (FVC, a standard measure of lung function) collected over about two years of follow-up.

The scientists then tested their programs using data from a second group of patients. For the test, the researchers had access only to an initial lung CT scan, clinical data, and FVC measurement. Their goal was to develop AIs that could accurately predict, based on the initial data, what each patient’s last three FVC measurements would be.

OSIC intends to conduct a second Pulmonary Fibrosis Progression Challenge next year.

“We are happy to announce that we will be hosting a second challenge in 2021 to help us continue this important work on behalf of patients around the world,” Estes said.

Marisa holds an MS in Cellular and Molecular Pathology from the University of Pittsburgh, where she studied novel genetic drivers of ovarian cancer. She specializes in cancer biology, immunology, and genetics. Marisa began working with BioNews in 2018, and has written about science and health for SelfHacked and the Genetics Society of America. She also writes/composes musicals and coaches the University of Pittsburgh fencing club.
Total Posts: 110

Patrícia holds her PhD in Medical Microbiology and Infectious Diseases from the Leiden University Medical Center in Leiden, The Netherlands. She has studied Applied Biology at Universidade do Minho and was a postdoctoral research fellow at Instituto de Medicina Molecular in Lisbon, Portugal. Her work has been focused on molecular genetic traits of infectious agents such as viruses and parasites.

×
Marisa holds an MS in Cellular and Molecular Pathology from the University of Pittsburgh, where she studied novel genetic drivers of ovarian cancer. She specializes in cancer biology, immunology, and genetics. Marisa began working with BioNews in 2018, and has written about science and health for SelfHacked and the Genetics Society of America. She also writes/composes musicals and coaches the University of Pittsburgh fencing club.
Latest Posts
  • PFF Registry
  • 3D cell model project
  • Pulmonary Fibrosis Progression Challenge
  • diabetes and lung fibrosis

How useful was this post?

Click on a star to rate it!

Average rating 3.8 / 5. Vote count: 12

No votes so far! Be the first to rate this post.

As you found this post useful...

Follow us on social media!

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?