4 papers
Revisiting the Role of Foundation Models in Cell-Level Histopathological Image Analysis under Small-Patch Constraints -- Effects of Training Data Scale and Blur Perturbations on CNNs and Vision Transformers
Hiroki Kagiyama, Toru Nagasaka, Yukari Adachi +5
Background and objective: Cell-level pathological image analysis requires working with extremely small image patches (40x40 pixels), far below standard ImageNet resolutions. It rem…
Spatially-extended Flow Phixer (SpeF-Phixer): A Spatially Extended -Mixing Framework for Gene Regulatory Causal Inference in Spatial Gene Field
Toru Nagasaka, Takaaki Tachibana, Yukari Adachi +5
Background and objective: Spatial transcriptomics provides rich spatial context but lacks sufficient resolution for large-scale causal inference. We developed SpeF-Phixer, a spatia…
Reliability Assessment Framework Based on Feature Separability for Pathological Cell Image Classification under Prior Bias
Takaaki Tachibana, Toru Nagasaka, Yukari Adachi +5
Background and objective: Prior probability shift between training and deployment datasets challenges deep learning-based medical image classification. Standard correction methods…
Novel Methods for Analyzing Cellular Interactions in Deep Learning-Based Image Cytometry: Spatial Interaction Potential and Co-Localization Index
Toru Nagasaka, Kimihiro Yamashita, Mitsugu Fujita
The study presents a novel approach for quantifying cellular interactions in digital pathology using deep learning-based image cytometry. Traditional methods struggle with the dive…