Graphical Tools and Retrieval Strategies for Medical Image Databases

Kofakis, Petros G.; Orphanoudakis, Stelios C.

doi:10.1007/978-3-662-00807-2_84

Cited by 4 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system is based on the integration of tools and methodologies which support the interactive processing, classification and browsing of medical images, as well as methodologies for the efficient automated indexing, storage and retrieval of such images by content. Earlier implementations of this system have been previously described in [114,115,16,116].…”

Section: Idb System Characteristicsmentioning

confidence: 99%

“…The user interface is equipped with interactive graphical tools and mechanisms for both displaying and selecting image classes or class properties, and for exploring their interrelationships. Such display tools are referred to as "graph browsers" in [115] .Furthermore, the user interface is equipped with an "image browser" whose purpose is to display reduced resolution images (miniatures) or sketches corresponding to images stored in the database.…”

Section: Image Browsingmentioning

confidence: 99%

See 1 more Smart Citation

Image Representation Indexing and Retrieval Based on Spatial Relationships and Properties of Objects

Πετρακησ¹

View full text Add to dashboard Cite

J.N THIS thesis, a new methodology is presented which supports the efficient representation, indexing and retrieval of images by content. Images may be indexed and accessed based on spatial relationships between objects, properties of individual objects, and properties of object classes. In particular, images are first decomposed into groups of objects, called "image subsets", and are indexed by computing addresses to all such groups. All groups up to a predefined maximum size are considered. This methodology supports the efficient processing of queries by image example and avoids exhaustive searching through the entire image database.The image database consists of a "physical database" which stores the original image files and a "logical database" which stores all image subsets together with their representations.The logical database consists of a set of data files each storing subsets of equal size.Queries are resolved through the logical database. Searching is performed in two steps, namely "hypothesis generation" and "hypothesis verification". Queries are distinguished into "direct access", corresponding to queries specifying a number of objects which is less than or equal to the maximum size of image subsets stored, and into "indirect access", corresponding to queries specifying more objects than the maximum size of image subsets stored.The performance of the proposed methodology has been evaluated based on simulated images, as well as images obtained with computed tomography and magnetic resonance imaging. Measurements of the size of the answer sets and of the retrieval response times have been obtained. The results of this evaluation are presented and discussed. This work completes and extends the work of others. In particular, the image representations used in this work may be considered as extensions of the representation of "2-D strings".The classical framework of representation of 2-D strings is specialized to the cases of scaled and unsealed images. Based on 2-D strings, an indexing scheme and a retrieval strategy are proposed, which in contrast to 2-D strings, avoid the exhaustive search through the entire image database. The performance of the proposed methodology has been compared with the performance of existing techniques of 2-D string indexing and retrieval. The results demonstrate significant retrieval performance improvements. iv 6 Epilog 6.1 Conclusions 6.2 Directions for Further Research 89 6.3 IDB System Characteristics A Image Indexing by 2-D Strings A.l Introduction A.2 Image Representation by 2-D Strings A.3 Image Matching Using 2-D Strings A.4 Extensions to 2-D Strings 101 A.5 Correctness of 2-D String Matching 103 Β User Interface 110 Vil A.2 Function agree determines whether the pair of objects (μ, ν) is a type-i (i = 0,1,2) match of the pair (j,k) 99 A.3 Function matchlD determines whether the 2-D string (x\, r\, s\) is a type-z (z = 0,1,2) subsequence of the 2-D string (a;2,r 2 ,52) A.4 Function sequences produces all type-z (i = 0,1,2) matched subsequences when called before matchl...

show abstract

Section: Idb System Characteristicsmentioning

confidence: 99%

Section: Image Browsingmentioning

confidence: 99%