A Novel Multi-Agent Spatiotemporal Fusion Framework for Intelligent Skin Cancer Diagnosis

Peiyao Zheng¹Jin Yang²Xuanru Wen¹Liao Hu^3*Boqian Hu^4*

• ¹Hubei University of Chinese Medicine, Wuhan, China
• ²Inner Mongolia University of Science and Technology, Baotou, China
• ³University of Illinois Chicago College of Medicine, Chicago, United States
• ⁴Hebei University of Chinese Medicine, Shijiazhuang, China

Skin cancer is one of the most common malignancies worldwide, and its early-stage diagnosis remains challenging due to its morphological similarity to benign lesions. Most existing computer-aided diagnostic systems rely on single static images, overlooking temporal information critical for distinguishing progressive malignancy. In this study, we propose a novel multi-agent spatiotemporal fusion framework to enhance diagnostic accuracy. The framework comprises three key components: a spatial agent based on a convolutional neural network for high-fidelity static feature extraction, a temporal agent employing gated recurrent units to model longitudinal lesion evolution, and a collaboration agent that dynamically fuses spatial and temporal representations through an attention-based weighting strategy. Experiments on large-scale public dermoscopic datasets demonstrated that our method achieved an accuracy of 94.5%, an F1-score of 93.8%, and an AUC of 0.97, outperforming traditional machine learning, CNN classifiers, and 3D-CNN baselines. Ablation studies confirmed the critical contribution of temporal modeling and adaptive fusion, particularly in differentiating early melanoma from atypical nevi. This work high-lights the potential of spatiotemporal modeling to improve early skin cancer detection and provides a promising direction for AI-assisted diagnosis of other chronic diseases requiring longitudinal monitoring.