Vision Transformer for Intracranial Hemorrhage Classification in CT Scans Using an Entropy-Aware Fuzzy Integral Strategy for Adaptive Scan-Level Decision Fusion

2025-03-11

AI Summary

To address low accuracy, poor interpretability, and inconsistent cross-slice decisions in intracranial hemorrhage (ICH) subtype classification from CT scans, this paper proposes a multi-level attention model based on the Pyramid Vision Transformer (PVT). We introduce a SHAP-value-driven feature selection mechanism to enhance discriminative capability and clinical credibility, and design the first entropy-aware fuzzy integral operator for slice-level adaptive decision fusion. The method integrates PVT-based representation learning, gradient-boosted neural networks, and entropy-weighted fuzzy integration. Evaluated on a public ICH dataset, it significantly outperforms state-of-the-art methods: accuracy improves by 3.2%, precision by 4.1%, while robustness and inference efficiency are concurrently enhanced. The framework demonstrates strong clinical deployability.

Technology Category

• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Intracranial hemorrhage (ICH) is a critical medical emergency caused by the rupture of cerebral blood vessels, leading to internal bleeding within the skull. Accurate and timely classification of hemorrhage subtypes is essential for effective clinical decision-making. To address this challenge, we propose an advanced pyramid vision transformer (PVT)-based model, leveraging its hierarchical attention mechanisms to capture both local and global spatial dependencies in brain CT scans. Instead of processing all extracted features indiscriminately, A SHAP-based feature selection method is employed to identify the most discriminative components, which are then used as a latent feature space to train a boosting neural network, reducing computational complexity. We introduce an entropy-aware aggregation strategy along with a fuzzy integral operator to fuse information across multiple CT slices, ensuring a more comprehensive and reliable scan-level diagnosis by accounting for inter-slice dependencies. Experimental results show that our PVT-based framework significantly outperforms state-of-the-art deep learning architectures in terms of classification accuracy, precision, and robustness. By combining SHAP-driven feature selection, transformer-based modeling, and an entropy-aware fuzzy integral operator for decision fusion, our method offers a scalable and computationally efficient AI-driven solution for automated ICH subtype classification.

Problem

• Accurate classification of intracranial hemorrhage subtypes in CT scans.
• Reducing computational complexity with SHAP-based feature selection.
• Enhancing scan-level diagnosis using entropy-aware fuzzy integral fusion.

Innovation

• Pyramid Vision Transformer for spatial dependencies
• SHAP-based feature selection for discriminative components
• Entropy-aware fuzzy integral for scan-level fusion