Peter Moss Acute Lymphoblastic Leukemia Detection System 2019

Adam Milton-Barker

Bangor, Wales

21 2

The Peter Moss Acute Lymphoblastic Leukemia (ALL) Detection System is our open source demo project utilizing Intel technologies for deep learning neural networks on the edge, including classifiers for Acute Lymphoblastic Leukemia & natural language understanding. ...learn more

Project status: Under Development

Mobile, Internet of Things, Artificial Intelligence

Groups
Internet of Things, DeepLearning, Artificial Intelligence Europe, Movidius™ Neural Compute Group

Intel Technologies
Movidius NCS, AI DevCloud / Xeon

Code Samples [1]Links [7]

Overview / Usage

The Acute Myeloid / Lymphoblastic Leukemia (AML / ALL) AI Research Project is an open source and free project in the memory of Peter Edward Moss, my grandfather who sadly passed away after a year long battle with Acute Myeloid Leukemia in 2019. The goal of the project is to showcase our the work of our team members and students, in a publicly available project.

Acute Myeloid Leukemia is a rare and aggressive form of Leukemia where mutated white blood cells destroy health red cells in masses. Early detection is very hard, if not impossible. In my grandfather's case, he was given the all clear in a standard blood test, one month before being diagnosed terminally ill.

The Acute Lymphoblastic Leukemia Detection System 2019 is part of the AML / ALL AI Research Project, and uses Intel technologies to provide a locally hosted management system for data management and inference. In addition to detection / early detection, the project incorporates facial recognition, natural linguistics, speech recognition and speech synthesis.

The Acute Lymphoblastic Leukemia Image Database for Image Processing dataset is used for this project. The dataset was created by Fabio Scotti, Associate Professor Dipartimento di Informatica, Università degli Studi di Milano. Big thanks to Fabio and his colleagues for their research and time put in to creating the dataset and documentation.

In Feb 2019, myself and Estela Cabezas demonstrated the Acute Lymphoblastic Leukemia Detection System at Embedded World in Nuremburg, Germany with Intel through the Intel Software Innovators Program. It was a privilege to be invited to demonstrate at the event and we were able to connect with medical researchers, students, doctors and AI/IoT enthusiasts.

In May 2019, Estela gave her first solo presentation representing the project to Intel executives and Intel Innovators / Student Ambassadors at an invite only event at Intel in Munich Germany, Intel Developer Affinity Day.

In August 2019, my grandfather passed away 1 day short of a year since being diagnosed. Myself and the rest of the team remain committed to the project.

In September 2019, Estela represented the project at Codemotion in Madrid with Intel. Estela gave a talk in the Code Motion Diversity In Tech Topic. You can view her presentation here.

In 2020 the team was honored to be one of 4 projects (and 1st from Europe) awarded the Intel® DevMesh AI Spotlight Award which is a new designation granted by Intel recognizing inspiring and breakthrough Artificial Intelligence projects in development from the Intel software community. A massive thank you to the team for all of their efforts last year, and to Intel for this amazing award!

We formed Peter Moss Leukemia AI Research in January 2020 in anticipation of registering as a non-profit. In June 2020 we became Asociacion De Investigacion En Inteligencia Artificial Para La Leucemia Peter Moss (Peter Moss Leukemia AI Research Association).

In January 2020 we started a crowdfunding campaign on GoFundMe, the donations from this crowdfunding campaign have helped pay for services/software and hardware required for our projects, and to help raise funds to take on staff. I have been volunteering full time since December, and as we are now an association, this campaign will become a personal campaign to help me raise funds to allow me to continue as a full time volunteer until we raise enough money for salaries or become funded. Please follow/share our mission/campaign using the following link: https://bit.ly/36YSsnv

Methodology / Approach

Server V1: A local PHP/MySQL server hosting a web based UI for managing and classifying data.
Facial-Auth V1: Hosts a REST API with access to the Siamese Neural Networks classifier used for facial authentication.
Augmentation V1: Applies filters to the original dataset and increases the amount of training / test data.
NCS1 Tensorflow Classifier V1: Hosts a REST API with access to the ALL NCS1 Classifier using NCS & NCSDK.
Chatbot V1: Hosts a REST API with access to the Natural Language Understanding Engine trained with basic knowledge of AML/ALL.