How Much Off-The-Shelf Knowledge Is Transferable From Natural Images To Pathology Images?

04/24/2020
by   Xingyu Li, et al.
0

Deep learning has achieved a great success in natural image classification. To overcome data-scarcity in computational pathology, recent studies exploit transfer learning to reuse knowledge gained from natural images in pathology image analysis, aiming to build effective pathology image diagnosis models. Since transferability of knowledge heavily depends on the similarity of the original and target tasks, significant differences in image content and statistics between pathology images and natural images raise the questions: how much knowledge is transferable? Is the transferred information equally contributed by pre-trained layers? To answer these questions, this paper proposes a method to quantify knowledge gain by a particular layer, conducts an empirical investigation into pathology image centered transfer learning, and reports some interesting observations. Different from prior arts in transfer learning, we show that transferability of off-the-shelf representation heavily depends on specific pathology image sets, and may obtain classification performance close to random guess. The observation in this study encourages further investigation of specific metric and tools to quantify the feasibility of transfer learning in future.

READ FULL TEXT
research
01/27/2023

Feasibility and Transferability of Transfer Learning: A Mathematical Framework

Transfer learning is an emerging and popular paradigm for utilizing exis...
research
12/01/2022

Rethinking Two Consensuses of the Transferability in Deep Learning

Deep transfer learning (DTL) has formed a long-term quest toward enablin...
research
08/26/2020

What is being transferred in transfer learning?

One desired capability for machines is the ability to transfer their kno...
research
04/11/2022

Commonality in Natural Images Rescues GANs: Pretraining GANs with Generic and Privacy-free Synthetic Data

Transfer learning for GANs successfully improves generation performance ...
research
01/21/2023

Pre-text Representation Transfer for Deep Learning with Limited Imbalanced Data : Application to CT-based COVID-19 Detection

Annotating medical images for disease detection is often tedious and exp...
research
11/30/2021

Beyond Flatland: Pre-training with a Strong 3D Inductive Bias

Pre-training on large-scale databases consisting of natural images and t...
research
03/04/2021

Contrastive Learning Meets Transfer Learning: A Case Study In Medical Image Analysis

Annotated medical images are typically rarer than labeled natural images...

Please sign up or login with your details

Forgot password? Click here to reset