Biological molecules live in an environment in which most of the forces that determine their activity are (at least apparently) different from those that guide the world visible to humans. These forces can be roughly classified as Brownian motion, lipophilic/hydrophilic interactions, and electrostatic potential. In the study and representation of proteins and other biological molecules, and especially their activity and interaction, it is therefore extremely important to be able to observe these forces in a meaningful way. This will lead to better understanding of dynamic interactions, and for a greater capacity for description and manipulation. While the calculation of these forces typically provides numerical data, it is not always easy and intuitive to have them represented in a way that makes sense to human beings, whose visual intelligence is one of the most highly developed.

This book contains the papers presented at the Eurographics 2010 Workshop on 3D Object Retrieval, organized by the Eurographics Association in cooperation with ACM SIGGRAPH, which took place on May 2, 2010, in Norrköping, Sweden.

One of the challenging directions of the research in modern medicine is towards modeling a \ndigital patient, i.e. a digital counterpart that should represent and abstract the real patient in \nall of his/her medically relevant aspects. The digital patient is nowadays a realistic vision: we \nwitness a huge flow of data every day, born-digital or easy to digitalize. These range from \nnumerical measurements (lab data, bedside measurements, home instrumentation) to coded \ndata (family history, patient medical history, current complaint, symptoms), from recorded \nsignals (ECG, EEG, EMG) to images (X-ray, MRI, CT, Ultrasound), from physical \nexaminations to narrative text (e.g. doctor’s notes, discharge summaries, ontologies, data \nbases) . \nGiven this variety of data sources, digital medicine needs a multimodal and combined access \nto interrelated heterogeneous data. \nKnowledge formalization techniques provide a good support to multimodal searches as they \nmay be used to build a formal framework to organize data at the various biological scales \nwhich contribute information related to the digital patient (shared conceptualization). Based \non this integrated view, various modalities to access and visualize data, concepts and \ninformation relevant to characterize deseases can be devised. Knowledge specification can \nbe also used to bridge properly the morphological, structural and functional aspects of \nmedically-relevant parts of the human body and used to navigate interactively the data and \nits properties (e.g. anatomical information and image orientation). \nFor the aforementioned purposes, ontologies are efficient tools for structuring and storing \ndifferent types of information and their dependencies. Ontologies are not new in the medical \ndomain: they have been introduced and used since years in order to systematize medical \nknowledge. Also, the importance of semantics-driven annotation of medical data has been \nsubject of many scientific and technological efforts and supported the development of \nsystems to support clinicians in their activities. Existing biological and medical ontologies offer complementary information (e.g., \ngeometric/topological information, disease evolution) on the human body and its functioning \nor deaseses. Some of them also make available navigation, browsing, and search \nmodalities, which suggest new ways to exploit the multi-scale and multi-modal nature of the \nknowledge captured. \nThis report surveys ongoing work related to ontology design and development in the medical \ndomain. This includes an overview of existing biomedical ontologies and some relative \ninitiatives, an overview of projects where multiple scales are integrated in advanced \nvisualization systems, and finally an overview of semantics-driven annotation systems for \nmedical data.