Mehdi Afshari
University of Waterloo
5 Papers
1 Citations
Mehdi Afshari is an academic researcher from University of Waterloo. The author has contributed to research in topics: Digital pathology & Feature (computer vision). The author has an hindex of 1, co-authored 5 publications.
Chat about Author
Papers
Fine-Tuning and training of densenet for histopathology image representation using TCGA diagnostic slides.
Abtin Riasatian,Morteza Babaie,Danial Maleki,Shivam Kalra,Mojtaba Valipour,Sobhan Hemati,Manit Zaveri,Amir Safarpoor,Sobhan Shafiei,Mehdi Afshari,Maral Rasoolijaberi,Milad Sikaroudi,Mohd Adnan,Sultaan Shah,Charles Choi,Savvas Damaskinos,Clinton J. V. Campbell,Phedias Diamandis,Liron Pantanowitz,Hany Kashani,Ali Ghodsi,Hamid R. Tizhoosh +21 more
TL;DR: KimiaNet as discussed by the authors employs the topology of the DenseNet with four dense blocks, fine-tuned and trained with histopathology images in different configurations, using high-cellularity mosaic approach to enable the usage of weak labels of 7126 whole slide images of formalin-fixed paraffin-embedded human pathology samples publicly available through The Cancer Genome Atlas (TCGA) repository.
135
Ink Marker Segmentation in Histopathology Images Using Deep Learning
Danial Maleki,Mehdi Afshari,Morteza Babaie,Hamid R. Tizhoosh +3 more
- 05 Oct 2020
TL;DR: In this article, the authors proposed to segment the ink-marked areas of pathology patches through a deep network and showed an FPN model with the EffiecentNet-B3 as the backbone was found to be the superior configuration with an F1 score of 94.53%.
5
•Posted Content
A Similarity Measure of Histopathology Images by Deep Embeddings.
Mehdi Afshari,Hamid R. Tizhoosh +1 more
TL;DR: In this article, the authors proposed a content-based similarity measure for high-resolution gigapixel histopathology images, where each image is divided into same-size patches with a meaningful amount of information (i.e., contained enough tissue).
2
•Posted Content
Fine-Tuning and Training of DenseNet for Histopathology Image Representation Using TCGA Diagnostic Slides
Abtin Riasatian,Morteza Babaie,Danial Maleki,Shivam Kalra,Mojtaba Valipour,Sobhan Hemati,Manit Zaveri,Amir Safarpoor,Sobhan Shafiei,Mehdi Afshari,Maral Rasoolijaberi,Milad Sikaroudi,Mohd Adnan,Sultaan Shah,Charles Choi,Savvas Damaskinos,Clinton J. V. Campbell,Phedias Diamandis,Liron Pantanowitz,Hany Kashani,Ali Ghodsi,Hamid R. Tizhoosh +21 more
TL;DR: KimiaNet as discussed by the authors employs the topology of the DenseNet with four dense blocks, fine-tuned and trained with histopathology images in different configurations, and uses high-cellularity mosaic approach to enable the usage of weak labels of 7,126 whole slide images of formalin-fixed paraffin-embedded human pathology samples publicly available through the Cancer Genome Atlas (TCGA) repository.
•Posted Content
Ink Marker Segmentation in Histopathology Images Using Deep Learning
TL;DR: This study proposes to segment the ink-marked areas of pathology patches through a deep network to avoid confusing tissue pixels with ink-colored pixels for computer methods.