We are very excited to announce the winners of our $55,000OSIC Pulmonary Fibrosis Progression Challenge, the first-ever computational challenge for interstitial lung diseases (ILDs). Our AI-focused competition was administered byKaggle, the world’s largest data science community platform, and we asked participants from around the globe to use machine learning to predict lung function decline in people living with pulmonary fibrosis.
The top three Kaggle investigators to devise the highest performing algorithm were first place winner Artyom Kulakov from Russia; second place winner Yusuke Ikemoto from Japan; and third place winner Khang Pham, also from Japan. They were selected from nearly 40,000 entries submitted by more than 2,500 competitors -- we were thrilled with the participation rate!
From the very beginning, our goal has been to drive collaboration between unlikely partners to help us look at fibrotic lung diseases differently. Prior to this Kaggle Challenge, our three winners had never heard of IPF or ILDs. Because of that, they were able to see this through a different lens and come back to us with three innovative approaches that used combinations of conventional neural networks and dense neural networks. The submissions from non-winning teams provided key learnings as well, with some creating linear regression models that proved very interesting.
We’ll be hosting a second challenge in 2022, and hope to double or even triple the amount of scans we share with participants. Thank you to the thousands of Kagglers who participated this time around for helping us continue this important work on behalf of patients around the world. We know we can build on everyone’s findings to make meaningful progress in our fight against these disease states!