Iconic indexing by 2-D strings
IEEE Transactions on Pattern Analysis and Machine Intelligence
An image algebra for pictorial data manipulation
CVGIP: Image Understanding
Building future medical education environments over ATM networks
Communications of the ACM
Image Information Systems: Where Do We Go From Here?
IEEE Transactions on Knowledge and Data Engineering
An Intelligent Image Database System
IEEE Transactions on Software Engineering
ICIAP '95 Proceedings of the 8th International Conference on Image Analysis and Processing
Virtual Images for Similarity Retrieval in Image Databases
IEEE Transactions on Knowledge and Data Engineering
Anatomical sketch understanding: Recognizing explicit and implicit structure
Artificial Intelligence in Medicine
Hi-index | 0.00 |
The paper describes a significant part of an experimental system for producing digital medical images, processing them to extract suitable spatial indexes, and to store and retrieve by content such images in order to provide users with an assisted visual browser to navigate a distributed archive. A prerequisite for the system described in this paper is that a physician should be able to manipulate the diagnostic images by simple visual commands that allow content-based access. In particular, the physician have to identify abnormalities (hot spots) in each image by deteriming their spatical locations, opacities, shapes and geometrical measures.Since our system needs the capability of retrieving images based on the presence of given patterns, it is necessary to define a similarity matching between the query and an image to retrieve. To efficiently perform such a matching, each image is stored together with a collection of metadata that are a very compact representation of the spatical contents of the image. These metadata forms the index of the image.We illustrate an experimental image browser for medical imaging diagnosis implementing the query-by-pictorial-example philosophy for user interface.