Democratizing Artificial Intelligence in Healthcare: A Study of Model Development Across Two Institutions Incorporating Transfer Learning

• Main Authors: Gupta1, Vikash, Roth, Holger, Buch3, Varun, Rockenbach, Marcio A. B. C, White, Richard D, Yang, Dong, Laur, Olga, Ghoshhajra, Brian, Dayan, Ittai, Xu, Daguang, Flores, Mona G, Erdal, Barbaros Selnur
• Format: Journal Article
• Language: English
• Published: 25-09-2020
• Subjects: Computer Science - Computer Vision and Pattern Recognition
• DOI: 10.48550/arxiv.2009.12437

The training of deep learning models typically requires extensive data, which
are not readily available as large well-curated medical-image datasets for
development of artificial intelligence (AI) models applied in Radiology.
Recognizing the potential for transfer learning (TL) to allow a fully trained
model from one institution to be fine-tuned by another institution using a much
small local dataset, this report describes the challenges, methodology, and
benefits of TL within the context of developing an AI model for a basic
use-case, segmentation of Left Ventricular Myocardium (LVM) on images from
4-dimensional coronary computed tomography angiography. Ultimately, our results
from comparisons of LVM segmentation predicted by a model locally trained using
random initialization, versus one training-enhanced by TL, showed that a
use-case model initiated by TL can be developed with sparse labels with
acceptable performance. This process reduces the time required to build a new
model in the clinical environment at a different institution.