About this document

Semi-Supervised Transfer Learning in Medical Imaging by Ferdi Jr is a document available to read on EtoBox.

This document proposes a new lightweight neural network architecture called MAKNet for medical image classification using semi-supervised transfer learning. MAKNet uses mixed asymmetric kernels and attention modules to reduce parameters significantly compared to popular architectures. It is trained on a large unlabeled medical dataset along with a smaller labeled dataset to generate pseudo-labels, which helps improve domain-specific transfer learning for medical imaging tasks. Experimental results show MAKN

Author

Ferdi Jr

Language

EN