Hybrid CNN-Transformer Framework Enhances Acute Lymphoblastic Leukaemia Classification

This research introduces a novel hybrid deep learning framework combining EfficientNet and Vision Transformer (ViT) models for multi-class classification of Acute Lymphoblastic Leukaemia (ALL). The approach aims to improve diagnosis accuracy by addressing biases, generalizability, and interpretability issues found in existing AI-based detection methods. This development could support more precise and prompt diagnosis of ALL, potentially improving patient survival rates.

Originally published by gnews