Acute Myeloid Leukemia Classifier 2021

Acute Myeloid Leukemia Classifier 2021

Acute Myeloid Leukemia Classifier 2021 on Linkedin

The Acute Myeloid Leukemia (AML) Classifier is an implementation of a Convolutional Neural Network designed to assist with the early detection of Acute Myeloid Leukemia. The project is inspired by Human-level recognition of blast cells in acute myeloid leukemia with convolutional neural networks by Matek, C., Schwarz, S, Spiekermann, K., Marr, C. CapsNets. Project by Adam Milton-Barker

 




Introduction

Acute Myeloid Leukemia (AML) is a cancer of the blood cells in the Myeloid blood cell lineage. AML is caused by abnormal Myeloid Blasts, or Myeloblasts, produced by the Myeloid progenitors in the bone marrow. Myeloblasts normally develop into healthy red and white blood cells, and platelets which help stop bleeding by forming clots.

Early detection of AML remains an unsolved problem. Our first research project began in 2018 due to Acute Myeloid Leukemia being missed in a routine blood test one month before Peter Moss was diagnosed as terminal. Due to the lack of available datasets we have not been able to work on classifiers for AML.

This project will be our first Acute Myeloid Leukemia project with the goals of developing a Convolutional Neural Networks based on the research proposed in Human-level recognition of blast cells in acute myeloid leukemia with convolutional neural networks by Matek, C., Schwarz, S, Spiekermann, K., Marr, C. CapsNets. The project uses the Single-cell Morphological Dataset of Leukocytes from AML Patients and Non-malignant Controls (AML-Cytomorphology_LMU), a dataset comprised of 18,365 single cell images from peripheral blood smears from 100 AML positive patients and 100 AML negative patients.





Project Contributors

Adam Milton-Barker

Adam Milton-Barker

President/Founder

 




Project Presentation

 

There is no presentation for this project.






Project Videos

No videos available.