The Peter Moss Leukemia AI Research Blog is the place to keep up to date with our latest news / info & tutorials.
Our blog not only provides easy access to articles published via our website, but also provides links to our publications off site.
Team member Dr Amita Kapoor published her findings on the COVID-19 pandemic. This research was referenced in the peer-reviewed paper: Covid-19 spread: Reproduction of data and prediction using a SIR model on Euclidean network by Kathakali Biswas, Abdul Khaleque, and Parongama Sen.
The final classifier achieves 98 (97.979)% using Tensorflow 2 & Ubuntu/GTX 1050 ti . You can run the classifier independently and classify local images, serve an API endpoint for HTTP requests, or you can use it as part of the VR experience which will be uploaded soon.
Here we will train the network we created in part 1, using the augmented dataset proposed in the Leukemia Blood Cell Image Classification Using Convolutional Neural Network paper by T. T. P. Thanh, Caleb Vununu, Sukhrob Atoev, Suk-Hwan Lee, and Ki-Ryong Kwon.
Here we will replicate the network architecture and data split proposed in the Acute Leukemia Classification Using Convolution Neural Network In Clinical Decision Support System paper and compare our results.
In the first part of this series: Introduction to convolutional neural networks in Caffe*, I covered the steps to recreate the basics of the convolutional neural network proposed in the paper: Acute Myeloid Leukemia Classification Using Convolution Neural Network In Clinical Decision Support System. In this article I will cover the steps required to create the dataset required to train the model using the network we defined in the previous tutorial. The article will cover the paper exactly, and will use the original 108 image dataset (ALL_IDB1).
As part of my R&D for the Acute Myeloid/Lymphoblastic Leukemia (AML/ALL) AI Research Project, I am reviewing a selection of papers related to using Convolutional Neural Networks (CNN) for detecting AML/ALL. These papers share various ways of creating CNNs, and include useful information about the structure of the layers and the methods used which will help to reproduce the work outlined in the papers. This article will take you through some information about Inception V3, transfer learning, and how we use these tools in the Acute Myeloid/Lymphoblastic Leukemia AI Research Project.
Inception V3 by Google is the 3rd version in a series of Deep Learning Convolutional Architectures. Inception V3 was trained using a dataset of 1,000 classes (See the list of classes here) from the original ImageNet dataset which was trained with over 1 million training images, the Tensorflow version has 1,001 classes which is due to an additional "background' class not used in the original ImageNet. Inception V3 was trained for the ImageNet Large Visual Recognition Challenge where it was a first runner up. This article will take you through some information about Inception V3, transfer learning, and how we use these tools in the Acute Myeloid/Lymphoblastic Leukemia AI Research Project.
The AML/ALL Classifier Data Augmentation program applies filters to datasets and increases the amount of training / test data available to use. The program is part of the computer vision research and development for the Peter Moss Acute Myeloid/Lymphoblastic (AML/ALL) Leukemia AI Research Project.