Recent Findings in Cancer Research Described by Researchers from University of Electronic Science and Technology of China (Crosslinknet: an Explainable and Trustworthy Ai Framework for Whole-slide Images Segmentation).

Předmět:	IMAGE segmentation ARTIFICIAL intelligence FEATURE extraction ELECTRONIC records CANCER research
Zdroj:	Cancer Weekly; 8/13/2024, p904-904, 1p
Abstrakt:	Researchers from the University of Electronic Science and Technology of China have developed an AI framework called CrossLinkNet for segmenting whole-slide images in cancer diagnosis. The framework addresses the limitations of traditional neural networks in processing high-resolution digital pathological images by enhancing interpretability and adaptability. The researchers designed an innovative Multi-Scale Multi-Branch Feature Encoder (MSBE) and employed cross-layer encoder-decoder connections for multi-level feature fusion. The framework not only achieves accurate segmentation results but also emphasizes interpretability, enhancing trust and reliability in AI-assisted diagnostics. This research has been peer-reviewed and published in Computers Materials & Continua. [Extracted from the article]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu