Abstract:Due to the malignancy of gliomas varies with their grade, classifying gliomas of different grades can assist doctors in developing personalized surgical plans during surgery, thereby improving prognosis. Raman spectroscopy is an optical method for real-time glioma diagnosis. However, high-grade glioma (HGG, WHO Grades III and IV), low-grade glioma (LGG, WHO Grades I and II) and normal tissue have similar biochemical components, leading to similar spectral characteristics. This similarity reduces classification accuracy when using traditional machine learning methods. In contrast, deep learning offers enhanced feature extraction capabilities without the need for extensive feature engineering. Nevertheless, the diversity in scale of the spectral features presents challenges in designing a neural network that effectively adapts to these characteristics. To address these issues, this paper proposes a Multi-Scale Convolutional Attention Residual Network (M-SCA ResNet), which incorporates Multi-Scale channel and spatial attention mechanisms along with residual structures to improve the model's feature extraction capabilities. The algorithm presented in this study, was employed to classify HGG, LGG, and healthy tissue in comparison with conventional machine learning and neural networks. The results indicate that M-SCA ResNet achieved an identification accuracy exceeding 85% for all three tissue types, along with the highest weighted F1-score. Furthermore, to enhance the interpretability of deep learning models, Gradient-weighted Class Activation Mapping (Grad-CAM) was utilized to extract and visualize key Raman shifts that significantly contribute to classification. Most of the extracted Raman shifts correspond to characteristic peaks of brain tissue which have been demonstrated to be effective in distinguishing between glioma of different grades and normal tissue in previous studies. This finding proves the strong correlation between the feature extraction capabilities of M-SCA ResNet and the biomolecular characteristics of various tissue. The experiments conducted in this study validate the feasibility of using M-SCA ResNet for glioma grading and provide valuable support for formulating subsequent surgical and treatment plans, indicating promising application in in-vivo and in-situ spectral diagnosis of glioma grading.

1D Gradient-Weighted Class Activation Mapping, Visualizing Decision Process of Convolutional Neural Network-Based Models in Spectroscopy Analysis

Exclusive Feature Constrained Class Activation Mapping for Better Visual Explanation.

Statistic-CAM: A Gradient-Free Visual Explanations for Deep Convolutional Network

Channel and band attention embedded 3D CNN for model development of hyperspectral image in object-scale analysis

Understanding the Learning Mechanism of Convolutional Neural Networks in Spectral Analysis

Cluster-CAM: Cluster-weighted visual interpretation of CNNs' decision in image classification

A Survey of Class Activation Mapping for the Interpretability of Convolution Neural Networks

Integrated Grad-CAM: Sensitivity-Aware Visual Explanation of Deep Convolutional Networks via Integrated Gradient-Based Scoring

MSG-CAM:Multi-scale inputs make a better visual interpretation of CNN networks

Grad-CAM: Why did you say that?

Grad-cam: Visual explanations from deep networks via gradient-based localization

Golf Guided Grad-CAM: Attention Visualization Within Golf Swings Via Guided Gradient-Based Class Activation Mapping

Respond-CAM: Analyzing Deep Models for 3D Imaging Data by Visualizations

Exploring how deep learning decodes anomalous diffusion via Grad-CAM

Feature Activation Map: Visual Explanation of Deep Learning Models for Image Classification

Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs

VS-CAM: Vertex Semantic Class Activation Mapping to Interpret Vision Graph Neural Network

Generating Self-Attention Activation Maps for Visual Interpretations of Convolutional Neural Networks

An Interpretable Multi-Scale Convolutional Attention Residual Neural Network for Glioma Grading with Raman Spectroscopy

Better Visual Interpretation for Remote Sensing Scene Classification