|
Activity Number:
|
219
- Artificial Intelligence and Machine Learning in Medical Imaging for Drug Development
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Statistics in Imaging
|
|
Abstract #316739
|
|
|
Title:
|
Using Deep Learning on HRCT Scan to Predict Pulmonary Fibrosis Progression and Support Early Clinical Development Decision
|
|
Author(s):
|
Bushi Wang* and Yi Liu and Hao Li
|
|
Companies:
|
Boehringer Ingelheim Pharmaceuticals Inc. and Boehringer Ingelheim and Boeringer-Ingelheim
|
|
Keywords:
|
deep learning;
CT image;
early clinical development;
biomarker endpoints;
pulmonary fibrosis
|
|
Abstract:
|
In early clinical development, small sample size and shorter study duration are usually limiting factors to selecting endpoints for early signal of efficacy. Exploratory biomarker endpoints that can be measured earlier and easier sometimes does not correlate well with long term clinical benefits. In this talk, we demonstrate the effort of Biostatistics and Data Sciences at Boehringer Ingelheim to leverage deep learning models and consortium database to gain additional insight in pulmonary fibrosis. A proof of concept Kaggle challenge demonstrated the connection of baseline lung CT scan and rate of lung function decline. We further explores the utility of early signals in CT scan to predict long term treatment effect.
|
Authors who are presenting talks have a * after their name.
