A global description of medical image with high precision

Abstract

Medical Imaging suffers from different problems. This paper explores our solution that aims to provide efficient retrieval of medical imaging. Depending on the user, the same image can be described through different views. In essence, an image can be described on the basis of either low-level properties, such as texture or color; context, such as date of acquisition or author; or semantic content, such as real-world objects and relations. Our approach consists of providing a global description solution capable of integrating different dimensions (or views) of a medical image. The description problem of medical images during both storage and retrieval processes is studied. Few proposed solutions take into consideration the heterogeneity of user competence (physician, researcher, student, etc.) and the necessiv of a high expressive power for medical imaging description. For example, spatial content in terms of relationships in surgical or radiation therapy of brain tumors is very decisive because the location of a tumor has profound implications on a therapeutic decision. Visual solutions are recommended and are the most appropriated for non computer-scientist users. However, current visual languages suffer from several problems, especially ambiguities generated by the user andlor the system at different levels of image description, imprecision and no respect of the integrity of spatial relations. This framework exposes our solution showing ho" this problematic can be resolved. An implementation has been realized to prove our proposition.

A Four-dimensional Approach to Medical Image Retrieval

“…Users can also define other spatial relations. In [18], we explain how spatial relations can be extended to provide high precision. -Semantic analysis.…”

“…As known, the location of medical entities is very hard to define. In [18], we explain in detail how spatial ambiguities are dealt with. The existence of IB allows the resolution of certain ambiguities.…”

“…The data model must be designed such that all required aspects are integrated into the image description. To address this problem, a recent study has been started in our laboratory (LISI) [15][16][17][18] which aims to provide an adequate retrieval solution for heterogeneous users of digital medical imaging equipment (X-rays, computed tomography, Magnetic Resonance Imaging, nuclear medicine, scanners, and ultrasound). Here, we describe our solution built on a four-dimensional data model for image description.…”

Content-Based Medical Image Retrieval

“…Given that abnormalities are deWned as gross deviations from anatomical models, the spatial relationships of body structures are often critical to the diagnosis, prognosis, and mechanics of human disease. For example, spatial content in terms of relationships in surgical or radiation therapy of brain tumors is very decisive, because the location and related adjacent structures of a tumor have profound implications on a therapeutic decision [175]. Low-level features cannot always capture or describe these complex scenarios.…”

“…As such, the development of global features that can represent an entire medical image database seems to be practically infeasible. Aiming to provide eYcient retrieval of generic medical images with coherent and eVective objectivity of interpretation at diVerent facets (or views) of medical images, Chbeir and Favetta [195] proposed a global description of medical images in which a hyperspaced image data model was constructed [175,196]. The data model was structured as a multispaced form in which each space contained a set of features (contextual, physical, spatial, and semantic), and considered the medical image as a composition of contextual and content feature spaces.…”