H
Hamid R. Tizhoosh
Researcher at University of Waterloo
Publications - 319
Citations - 9545
Hamid R. Tizhoosh is an academic researcher from University of Waterloo. The author has contributed to research in topics: Image retrieval & Image segmentation. The author has an hindex of 41, co-authored 291 publications receiving 7786 citations. Previous affiliations of Hamid R. Tizhoosh include University of Toronto & Otto-von-Guericke University Magdeburg.
Papers
More filters
Journal Article
Automated Resolution Selection for Image Segmentation
TL;DR: A measure for defining the best resolution based on user/system criteria is proposed, offering a trade-off between accuracy and computation time, and a learning approach is then introduced for the selection of the resolution, whereby extracted image features are mapped to the previously determined best resolution.
Book ChapterDOI
A Compact Representation of Histopathology Images Using Digital Stain Separation and Frequency-Based Encoded Local Projections
TL;DR: In this article, a modified encoded local projections (ELP) algorithm is proposed to estimate local frequencies by quantifying the changes in multiple projections in local windows of greyscale images.
Journal ArticleDOI
Learning Binary and Sparse Permutation-Invariant Representations for Fast and Memory Efficient Whole Slide Image Search
TL;DR: This work introduces new loss functions for learning sparse and binary permutation-invariant WSI representations that employs instance-based training achieving better memory efficiency and outperforms Yottixel both in terms of retrieval accuracy and speed.
Journal ArticleDOI
A BERT model generates diagnostically relevant semantic embeddings from pathology synopses with active learning
Youqing Mu,Hamid R. Tizhoosh,Rohollah Moosavi Tayebi,Catherine L. Ross,Catherine L. Ross,Monalisa Sur,Monalisa Sur,Brian Leber,Brian Leber,Clinton J. V. Campbell,Clinton J. V. Campbell +10 more
TL;DR: A transformer-based natural language model can extract embeddings from pathology synopses that capture diagnostically relevant information and provide a generalizable deep learning model and approach to unlock the semantic information inherent in pathologysynopses toward improved diagnostics, biodiscovery and AI-assisted computational pathology.