Leukemia detection based on microscopic blood smear images using deep learning

TL;DR

This paper presents a deep learning-based system for early leukemia detection from microscopic blood smear images, achieving high accuracy and potentially improving diagnosis speed and cost-effectiveness.

Contribution

The study introduces a novel deep neural network approach for leukemia detection with 97.3% accuracy, streamlining traditional diagnostic processes.

Findings

Achieved 97.3% classification accuracy

System enables rapid, cost-effective leukemia screening

Supports early diagnosis to improve treatment outcomes

Abstract

In this paper we discuss a new method for detecting leukemia in microscopic blood smear images using deep neural networks to diagnose leukemia early in blood. leukemia is considered one of the most dangerous mortality causes for a human being, the traditional process of diagnosis of leukemia in blood is complex, costly, and time-consuming, so patients could not receive medical treatment on time; Computer vision classification technique using deep learning can overcome the problems of traditional analysis of blood smears, our system for leukemia detection provides 97.3 % accuracy in classifying samples as cancerous or normal samples by taking a shot of blood smear and passing it as an input to the system that will check whether it contains cancer or not. In case of containing cancer cells, then the hematological expert passes the sample to a more complex device such as flow cytometry to generate complete information about the progress of cancer in the blood.