HER2 binary classification performance for the (a) EfficientNetV2B0 subnetworks, (b) EfficientNetV2B0 subnetworks pretrained on ImageNet, (c) ATHER2, and (d) ATHER2 pretrained on BreakHis (BCI dataset – IHC images).

HER2 binary classification performance for the (a) EfficientNetV2B0 subnetworks, (b) EfficientNetV2B0 subnetworks pretrained on ImageNet, (c) ATHER2, and (d) ATHER2 pretrained on BreakHis (BCI dataset – IHC images).

<p>HER2 binary classification performance for the (a) EfficientNetV2B0 subnetworks, (b) EfficientNetV2B0 subnetworks pretrained on ImageNet, (c) ATHER2, and (d) ATHER2 pretrained on BreakHis (BCI dataset – IHC images).</p>

Saved in:

Bibliographic Details
Main Author:	Lamiaa Abdel-Hamid (22250618) (author)
Other Authors:	Pratheepan Yogarajah (22250621) (author), Safy Hosny Ahmed Tealab (22250624) (author)
Published:	2025
Subjects:	Medicine Cell Biology Space Science Biological Sciences not elsewhere classified required laboratory equipment final testing phases complex classification tasks based her2 scoring aggressive cancer type requires special diagnosis two lightweight models customized lightweight network multiclass classifications regardless e stained images lightweight models multiclass classifications time diagnosis network pruning network parameters binary classifications treatment methods transfer learning rural regions routinely performed readily available patient samples networks resulted mobile phones generalization capability embedded systems efficientnetv2b0 pruned efficiently trained effectively evaluate e images different datasets developing countries deep networks deep learning constrained devices cbam ). attention mechanisms accuracies ranging 80 %.
Tags:	Add Tag No Tags, Be the first to tag this record!

Cannot write session to /tmp/vufind_sessions/sess_l2v2po9ubsavrs9cj31p74ubun