Recently, there has been growing interest in medical image analysis methods based on multi-layer perceptron (MLP), which have demonstrated remarkable performance in this field. Typically, the MLP has been employed to flatten feature maps along the channel and spatial dimensions and encode them to capture the overall feature information. However, for MLP-based architecture, the whole learning process requires high computational complexity, a large number of parameters, and large-scale training datasets. Meanwhile, quantum tensor network (QTN), derived from quantum many-body systems, have gained popularity in machine learning due to their ability to reduce the number of parameters and explore the correlation of remote elements in an image through a tensor contraction operation. In this paper, we propose a novel framework named QTN-MLP based on QTN and MLP to tackle the weaknesses of MLP. This paper has been extensively validated on the COVID-CT dataset and ISIC 2018 dataset, and the experiments show that our method competes with state-of-the-art methods with lower computational complexity and fewer parameters.

In recent years, deep convolutional neural networks (CNNs) have been spectacularly successful in the classification and diagnosis of breast cancer and its histopathological images. However, for CNNs, the whole learning process requires high computational complexity, a large number of parameters, and loss of certain global feature information. Meanwhile, the flexibility of tensor networks (TNs) algorithms to machine learning leads to creativity in devising new approaches. In this paper, we propose a novel framework named ConTenNet based on the pre-trained CNNs and quantum TNs (QTNs) to address the weaknesses in CNNs. We propose ConTenNet on the BreakHis dataset, and the experiments show that our model competes with the state-of-the-art methods on both original and normalized images with lower computational complexity, a less number of parameters, and global feature information. Moreover, we adopt the color normalization method to avoid the interference of color in model learning, using the gradient-weighted class activation mapping (Grad-CAM) to prove the necessity of color normalization and the reliability of model learning.