1. 医学成像信息学及其在转化医学中的作用;

2. 在医疗治疗护理过程中,成像与预防,诊断和治疗间的交织性质;

3. 本学科过去和当前的挑战。

Key points

1. Introduction of   medical imaging informatics and its role in transforming healthcare research.   

2. Touching upon the interwoven nature of imaging with   preventative, diagnostic, and therapeutic elements of patient care.

3. Illustrating both   past and current challenges of the discipline.






Key points

1. Reviewing clinical   imaging modalities (i.e., projectional x-ray, computed tomography   (CT), magnetic resonance (MR), ultrasound) and a primer on imaging anatomy   and physiology.

2. Modality encompassing core physics principles and image   formation techniques.

3. Presenting an overview of anatomy and physiology from the perspective of projectional and crosssectional imaging. A few systems are covered in detail.






Key points

1. Tackling the question of how we store and access  imaging and other patient information as part of a distributed and heterogeneous EMR.

2. Providing major   information systems (PACS; HIS) and different data standards (HL7 DICOM) .

3. Discussing distributed architectures applid to clinical databases (peer-to-peer, grid   computing).






Key points

1. Integrating and presenting patient information to   support the physician’s cognitive tasks.

2. Presenting works   related to the visualization of medical data including graphical metaphors   (lists and tables; plots and charts; graphs and trees; and pictograms).

3. Illustrating the   methods to combine the visual components based on a definition of (task)   context and user modeling.




1.  强度标准化;去噪;滤波设计;图像分割。

2.  线性和非线性图像配准方法及导航应用。

3.  基于外观和形状区分描述子;特征提取、选择和降维方法;

4.  基于成像的解剖图谱,详细描述其构造和用法;理解基于人群的图谱和由于疾病发展引起的差异的处理手段。

Key points

Medical imaging   informatics focus on how imaging studies, alongside other clinical data, can   be standardized and their content (automatically) extracted to guide medical   decision making processes. Unless medical images are standardized, quantitative  comparisons across studies is subject to various sources of bias/artifacts   that negatively influence assessment. For creating scientific-quality imaging   databases, this chapter starts with the groundwork for understanding what   exactly an image captures, and outlines the different aspects encompassing   the standardization process:

1.  Intensity   normalization; denoising; image segmentation.

2.  Both linear and   nonlinear image registration methods and its application in image navigation.

3.  Discussion of commonly extracted imaging features,   including appearance- and shape-based descriptors; Image feature selection and dimensionality reduction methods is also provided.

4.  Description of the   imaging-based anatomical atlases, detailing their construction and usage for   understanding population-based norms and differences arising due to a disease   process.



1. 本章描述医学成像中经常遇到的关系。

2. 本章聚集单个患者的观察结果以表征患者疾病的状态和行为,包括其自然过程和(治疗性)干预的结果。

3. 本章沿空间(如物理和解剖关系),时间(如临床事件序列)和临床导向模型(即专用于表示医疗保健抽象的模型)组织这些影像信息。

4. 讨论医疗数据模型设计的动机,描述以现象为中心的数据模型。

Key points

1. Describing the  different types of relationships commonly encountered in medical imaging.

2. Aggregate the  observations for a single patient to characterize the state and behavior of   the patient’s disease, both in terms of its natural course and as the result   of (therapeutic) interventions.

3. The chapter organizes the information along spatial (physical and anatomical relations), temporal (sequences of clinical events, episodes of care), and clinically-oriented models (those models specific to representing   a healthcare abstraction).

4. Discussion of the  motivation behind the design of a medical data model. Describing  phenomenon-centric data model.



1. 讲述贝叶斯信念网络(BBN),评估过去和当前这些模型在医疗环境的应用,讨论建立这些模型的实际考虑和必须作出的假设。讲述特殊类型的信念网络,说明其用途。

2. 讨论查询BBN的算法和工具。考虑两大类查询:信念更新和遗传推理。前者在给定一些具体证据的情况下重新计算网络中的后验概率;后者计算BBN的最优配置以最大化某些指定标准。

3. 提供精确和近似推理方法描述,讨论与信念网络评估相关的问题,寻求标准的技术准确度和参数敏感性指标。给出BBN的医学示例应用,如基于案例的检索和图像处理任务。

Key points

1.  Handling Bayesian belief networks (BBNs), appraising past and current efforts to apply these  models to the medical environment. Addressing the practical considerations in   the building of these models and the assumptions that must be made.

2.  Addressing the algorithms and tools that enable us to query BBNs. Two broad classes of  queries are considered: belief updating, and abductive reasoning. The former entails the re-computation of posterior probabilities in a network given some  specific evidence; the latter involves calculating the optimal configuration of the BBN in order to maximize some specified criteria.

3.  Describing the exact and approximate inference methods. Covering special types of belief networks. Illustrating their potential usage in medicine. Looking to standard technical accuracy metrics in parametric sensitivity analysis. Concluding with some example applications of BBNs in medicine, including to support case-based retrieval and image processing tasks.



1. 介绍生物统计学,包括对基本概念及不同情况和假设下评估假设的统计检验。

2. 引入误差和性能评估讨论,包括灵敏度、特异性和接收器操作特性分析。

3. 研究设计以不同类型的测试假设形成的实验描述,并讨论变量选择和样本大小/功率计算的过程。

4. 简单描述研究偏差/误差的来源,以及决策的统计工具。

5. 评估基于内容的图像检索和评估(系统)可用性。

Key points

1. Introducing biostatistics including basic concepts and the statistical tests that are used to evaluate hypotheses under different circumstances and assumptions.

2. Discussion of error and performance assessment is introduced, including sensitivity/specificity and receiver operative characteristic analysis.

3. Study design encompasses a description of the different types of experiments that can be formed to test a hypothesis, and goes over the process of variable selection and sample size/power calculations.

4. Sources of study bias/error are briefly described, as are statistical tools for decision making.

5. Evaluation the content-based image retrieval; and assessing (system) usability.