Tensor networks are emerging architectures for implementing quantum classification models. Branching Multi-Scale Entanglement Renormalization Ansatz (BMERA) is a tensor network known for its enhanced entanglement properties. This paperintroduces a hybrid quantum-classical classification model based on BMERA and explores the correlation between circuit layout,expressiveness, and classification accuracy. Additionally, we introduce an autodifferentiation method for computing the costfunction gradient, which presents a viable option for other hybrid quantum-classical models. Through numerical experiments,we demonstrate the superior accuracy and robustness of our classification model in tasks such as image recognition andcluster excitation discrimination, outperforming quantum classification models based on tree tensor networks and multi-scaleentanglement regularization ansatz.