Enhanced blood cell classification using neural networks
Ventrapragada Teju, Naga Bhavani Devarasetti, Swathi Boga, Kireeti Dumpala, Lakshmi Priya Challa, Krishna Prakash, Kambhampati Venkata Sowmya, Lingamallu Naga Srinivasu · IET conference proceedings. · 2025
The automatic classification of blood cells is important for early-stage diagnosis and treatment of blood diseases, it minimizes the need for human sorting and classification of blood cells. Deep learning and Convolutional Neural Networks (CNNs) in particular, have improved classification accuracy, due to their strength in identifying unique characteristics within microscopic images. This paper is about developing a CNN-based model that helps predicting the type of blood cell from an input blood cell image. The paper also analyzes model's effectiveness through a comparison with conventional machine learning approaches. It will additionally explore the effects of preprocessing, optimization and data augmentation techniques to address issues like class imbalance and data variability. A graphical user interface (GUI) will also be implemented for user-friendly interaction with the model, enabling users to submit image for instant blood cell type detection. In summary, this work contributes to the development of an effective and accurate deep learning-based classification model to the field of automated haematology.