Computer Vision News - May 2023

8 Computer Vision Code Hi everyone! Today, we're diving into another fascinating application of deep learning in medicine: blood cell classification using computer vision. This has far-reaching implications, from diagnosing blood disorders to analyzing infection responses. It’s a medical application indeed, but I’m actually using a super-nice version of the net we developed in the lab and it has great use in the broad field of computer vision applications! So, let's get started! Introduction Blood cell classification is crucial in diagnosing various medical conditions, such as anemia, leukemia, and infections. Traditional classification methods often involve manual examination of blood samples under a microscope, which can be time-consuming and prone to human error. Deep learning models, particularly convolutional neural networks (CNNs), have shown great potential for automating this process with high accuracy. One popular CNN architecture for image classification tasks is the ResNet, which employs residual connections to facilitate the flow of information between layers. In this article, we will explore how ResNet can be adapted for blood cell classification. Blood cell classification can be performed on different types of cells, such as red blood cells, white blood cells, and platelets. Here, we will focus on classifying white blood cells, which play a vital role in our immune system. You can find real blood cell images online, such as this dataset on Kaggle: White Blood Cell Classification Dataset . Here is a very useful video for understanding blood cells, their types and why we attempt the classification approach: