collaborators

7 papers

cs.CV2025

INTERACT-CMIL: Multi-Task Shared Learning and Inter-Task Consistency for Conjunctival Melanocytic Intraepithelial Lesion Grading

Mert Ikinci, Luna Toma, Karin U. Loeffler +6

Accurate grading of Conjunctival Melanocytic Intraepithelial Lesions (CMIL) is essential for treatment and melanoma prediction but remains difficult due to subtle morphological cue…

cs.LG2025

VeLU: Variance-enhanced Learning Unit for Deep Neural Networks

Ashkan Shakarami, Yousef Yeganeh, Azade Farshad +3

Activation functions play a critical role in deep neural networks by shaping gradient flow, optimization stability, and generalization. While ReLU remains widely used due to its si…

cs.CV2025

DeepAf: One-Shot Spatiospectral Auto-Focus Model for Digital Pathology

Yousef Yeganeh, Maximilian Frantzen, Michael Lee +3

While Whole Slide Imaging (WSI) scanners remain the gold standard for digitizing pathology samples, their high cost limits accessibility in many healthcare settings. Other low-cost…

cs.LG2025

Stress-Aware Resilient Neural Training

Ashkan Shakarami, Yousef Yeganeh, Azade Farshad +3

This paper introduces Stress-Aware Learning, a resilient neural training paradigm in which deep neural networks dynamically adjust their optimization behavior - whether under stabl…

eess.IV2025

Unit-Based Histopathology Tissue Segmentation via Multi-Level Feature Representation

Ashkan Shakarami, Azade Farshad, Yousef Yeganeh +4

We propose UTS, a unit-based tissue segmentation framework for histopathology that classifies each fixed-size 32 * 32 tile, rather than each pixel, as the segmentation unit. This a…

cs.CV2025

Latent Drifting in Diffusion Models for Counterfactual Medical Image Synthesis

Yousef Yeganeh, Azade Farshad, Ioannis Charisiadis +5

Scaling by training on large datasets has been shown to enhance the quality and fidelity of image generation and manipulation with diffusion models; however, such large datasets ar…