Multi-modal relevance feedback for medical image retrieval
| Art der Publikation: | Artikel in einem Konferenzbericht |
| Zitat: | |
| Buchtitel: | SIGIR workshop on Medical Information Retrieval |
| Jahr: | 2014 |
| Ort: | Gold Coast, Australia |
| Abriss: | Medical image retrieval can assist physicians in finding in- formation supporting their diagnosis. Systems that allow searching for medical images need to provide tools for quick and easy navigation and query refinement as the time for information search is often short. Relevance feedback is a powerful tool in information re- trieval. This study evaluates relevance feedback techniques with regard to the content they use. A novel relevance feed- back technique that uses both text and visual information of the results is proposed. Results show the potential of relevance feedback tech- niques in medical image retrieval and the superiority of the proposed algorithm over commonly used approaches. Future steps include integrating semantics into relevance feedback techniques to benefit of the structure knowledge of ontologies and experimenting on the fusion of text and visual information. |
| Nutzerfelder: | A SIGIR 2014 workshop - Gold Coast, Australia - July, 11th 2014 |
| Schlagworte: | Content-based image retrieval, Medical image retrieval, relevance feedback |
| Autoren | |
| Hinzugefügt von: | [] |
| Gesamtbewertung: | 0 |
|
Anhänge
|
|
|
|
|
Notizen
|
|
|
|
|
|
Themen
|
|
|
|
|
