Novel paradigms for computational analysis for structure and function in medical images

在过去的二十年中，医学成像已成为现代医疗保健的主要支柱之一。越来越丰富的可用结构和功能成像技术提供了精美的医学数据，为解剖学和生理学的非侵入性可视化和定量分析提供了巨大的机会。应用，其中一些是临床现实，从基本的计算机辅助诊断到精心制作的图像引导干预。然而，利用这种医学图像数据中嵌入的大量信息是一项非常复杂的任务。事实上，缺乏高效、准确和鲁棒的计算分析技术带来了巨大的挑战，严重阻碍了图像信息的最佳提取和实际使用。

Over the past two decades, medical imaging has become one of the main pillars of modern healthcare. The increasingly rich spectrum of available structural and functional imaging technologies provides exquisite medical data that offer tremendous opportunities for non-invasive visualization and quantitative analysis of anatomy and physiology. Applications, some of which are clinical realities, vary from basic computer-assisted diagnostics to elaborate image-guided interventions. Exploiting the enormous amounts of information embedded in such medical image data is, however, a very complicated task. In fact, the lack of efficient, accurate, and robust computational analysis techniques poses huge challenges that seriously hinder optimal extraction and use of image information in practice.