From 4c8c0ad956cea4427cc04248e0ec02561b70b143 Mon Sep 17 00:00:00 2001 From: Insiyah Hajoori Date: Mon, 24 Jun 2019 23:21:31 +0530 Subject: [PATCH 1/2] Update README.md --- lymphoma-cancer-classification/README.md | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/lymphoma-cancer-classification/README.md b/lymphoma-cancer-classification/README.md index 5887b3c..52c374c 100644 --- a/lymphoma-cancer-classification/README.md +++ b/lymphoma-cancer-classification/README.md @@ -12,16 +12,21 @@ Non-overlapping patches of size 256 x 256 are extracted from each image. 15% fro ### Inception Recurrent Convolutional Neural Network The IRCNN architecture consists of general convolution layers, IRCNN blocks, transaction blocks, and a softmax logistic regression layer. + The IRCNN block, performs recurrent convolution operations on different sized kernels. + As the input and output dimensions do not change, this is simply an accumulation of feature maps with respect to the time step considered. This helps to strengthen the extraction of the target features. + In the transaction block, three operations (convolution, pooling, and drop-out) are performed depending upon the placement of the block in the network. According to Figure, all of the operations are applied in the very first transaction block and second transaction block. The third transaction block consists of convolution, global-average pooling, and drop-out layer. The GlobalAveragePooling layer is used as an alternative to a fully connected layer. + The Stochastics Gradient Descent (SGD) optimization method is used with initial learning rate 0.01 ### Usage -The models folder contains Keras model and it's corresponding tensorflowjs model. Download the .zip file of repository and load the model in caMicrscope. +The *model* folder contains Keras model and it's corresponding tensorflowjs model. Download the .zip file of repository and load the model in caMicrscope. ### References -- Advanced Deep Convolutional Neural Network Approaches for Digital Pathology Image Analysis: a comprehensive evaluation with different use cases + +- *Advanced Deep Convolutional Neural Network Approaches for Digital Pathology Image Analysis: a comprehensive evaluation with different use cases* Md Zahangir Alom, Theus Aspiras, Tarek M. Taha, Vijayan K. Asari, TJ Bowen, Dave Billiter, Simon Arkell -- M. Liang, X. Hu, "Recurrent convolutional neural network for object recognition", CVPR, pp. 3367-3375, 2015. -- Inception Recurrent Convolutional Neural Network for Object Recognition +- M. Liang, X. Hu, *"Recurrent convolutional neural network for object recognition"*, CVPR, pp. 3367-3375, 2015. +- *Inception Recurrent Convolutional Neural Network for Object Recognition* Md Zahangir Alom, Mahmudul Hasan, Chris Yakopcic, Tarek M. Taha From 1b0d7bdabc78e46ed9a97989845f3343d1bd22d3 Mon Sep 17 00:00:00 2001 From: Insiyah Hajoori Date: Wed, 24 Jul 2019 22:18:46 +0530 Subject: [PATCH 2/2] Update README.md --- .../README.md | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/classification-of-HnE-stained-histological-breast-cancer-images/README.md b/classification-of-HnE-stained-histological-breast-cancer-images/README.md index fd5fe9e..6f1e56e 100644 --- a/classification-of-HnE-stained-histological-breast-cancer-images/README.md +++ b/classification-of-HnE-stained-histological-breast-cancer-images/README.md @@ -1,3 +1,8 @@ +## Details +The image dataset is composed of high-resolution (2040 × 1536 pixels), uncompressed, and annotated H&E stain images from the Bioimaging 2015 breast histology classification challenge [3]. All the images are digitized with the same acquisition conditions, with magnification of 200× and pixel size of 0.42μm × 0.42μm. Each image is labeled with one of four classes: i) normal tissue, ii) benign lesion, iii) in situ carcinoma and iv) invasive carcinoma + +Input size = 512 x 512 x 3 + ## Citation In this notebook, there's a Keras implementation of the patch-wise network of the paper Two-Stage Convolutional Neural Network for Breast Cancer Histology Image Classification: @@ -10,4 +15,4 @@ In this notebook, there's a Keras implementation of the patch-wise network of th year={2018}, organization={Springer} } -``` \ No newline at end of file +```