Visualization and extraction of Diagnostic Information from functional dynamic images

这个多学科的研究项目建议开发一个决策支持系统，以协助医务人员。该系统将重建，显示和分析动态核医学图像。它将包括动态数据的迭代重建、四维图像的实时可视化、数据参数分割及其定量分析工具以及新颖的用户界面。该方法将提供定量准确的诊断信息，但快速，简单，直观和实用。创新的多学科研究开发环境将使我们的合作伙伴能够直接参与重建，特征提取和可视化的新算法的开发。这是我们系统的一个关键特征。将开发的算法紧密集成，将使来自分析阶段的信息能够反馈到重建阶段，从而改善整体结果。该方法将为医疗用户提供对使用标准临床设备获取的功能性患者图像的简单且直观的访问，并增强其诊断能力。虽然我们的方法将是一般的，适用于不同的动态临床情况下，该项目的特别重点是心脏和肿瘤应用的动态单光子发射计算机断层扫描（SPECT）的研究。我们假设，除了3D空间信息的使用时间将提高诊断，允许更准确地识别心肌灌注缺损和更好地分离重叠的心脏SPECT图像中的肝脏组织。这项技术有能力将目前在昂贵得多和有限的正电子断层扫描中心使用的分子成像的许多好处带到社区医院的水平，将癌症，神经病学和心脏病的先进成像和生物技术的革命性突破扩展到成千上万的加拿大人。我们的合作伙伴是Philips Medical Systems和Dynapix Intelligence Imaging。

This multidisciplinary research project proposes to develop a decision support system to assist medical personnel. The system will  reconstruct, display and analyze dynamic nuclear medicine images. It will include iterative reconstruction of the dynamic data, real-time visualization of four-dimensional images, tools for parametric segmentation of the data and their quantitative analysis as well as novel user interfaces. The method will provide quantitatively accurate diagnostic information, yet be fast, simple, intuitive and practical. An innovative multidisciplinary research development environment will enable our partners to participate directly in the development of novel algorithms for reconstruction, feature extraction and visualization. This is a key feature of our system. This tight integration of the algorithms to be developed will enable information from the analysis stage to be fed back into the reconstruction stage, improving the overall result. This approach will provide the medical user with a simple and intuitive access to functional patient images acquired using standard clinical equipment and enhance their diagnostic power. Although our methods will be general and applicable to different dynamic clinical situations, the particular focus of this project is on cardiac and oncological applications of dynamic single photon emission computed tomography (SPECT) studies. We postulate that the use of temporal in addition to 3D spatial information will enhance diagnosis by allowing more accurate identification of myocardial perfusion defects and better separation of liver tissue overlapping that of the heart in SPECT images. This technique has the ability to bring many of the benefits of molecular imaging currently being utilized in the much more expensive and limited Positron Tomography centers to the level of the community hospital, extending the revolutionary breakthroughs in advanced imaging and biotechnology in cancer, neurology, and cardiac disease to thousands of additional Canadians. Our partners are Philips Medical Systems and Dynapix Intelligence Imaging.