Iterative brain tumor retrieval for MR images based on user’s intention model
作者：
Highlights：
• Involve the physician in the retrieval loop (“user-in-the-loop”).
• Model the intention of the user based on the eye-tracking data of him/her collected when inspecting the relevance between different images.
• Define similarity based both on the intention model and image visual features.
• Use visual features as a bridge to transfer eye movement data.
• Continuously improve retrieval performance through iterative feedback.
摘要
•Involve the physician in the retrieval loop (“user-in-the-loop”).•Model the intention of the user based on the eye-tracking data of him/her collected when inspecting the relevance between different images.•Define similarity based both on the intention model and image visual features.•Use visual features as a bridge to transfer eye movement data.•Continuously improve retrieval performance through iterative feedback.
论文关键词：CBIR,Brain tumor images,Eye-tracking,Intention similarity,Iterative retrieval,Relevance feedback
论文评审过程：Received 11 August 2020, Revised 3 March 2022, Accepted 11 March 2022, Available online 14 March 2022, Version of Record 21 March 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108650
