Empirical evaluation of general-purpose image features for pathology-oriented image retrieval of Alzheimer Disease cases

Ricardo Antonio Mendoza Leon (1, 2, 3) , John Puentes (1, 2) , Fabio A. González (4) , Marcela Hernández Hoyos (3)

1 Lab-STICC_TB_CID_DECIDE
2 ITI - Département Image et Traitement Information
3 UNIANDES - Facultad de Ingeniería [Univ de los Andes Colombia]
4 MindLab Research Group (Universidad Nacional de Colombia)

Purpose: To evaluate the applicability of general-purpose image feature algo- rithms as effective and efficient descriptors for pathology-based image retrieval of Magnetic Resonance Imaging (MRI) brain volumes under the hypothesis that localized quantitative visual similarity associates consistently with the pathological state in Alzheimer Dis- ease (AD).

Mots clés: Image retrieval; Alzheimer disease; Magnetic resonance imaging; Image features

Domaines: Imagerie médicale; Traitement des images [eess.IV]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement du signal et de l'image [eess.SP]

Ricardo Antonio Mendoza Leon, John Puentes, Fabio A. González, Marcela Hernández Hoyos. Empirical evaluation of general-purpose image features for pathology-oriented image retrieval of Alzheimer Disease cases. CARS 2016 : 30th International Congress on Computer Assisted Radiology and Surgery In : International Journal of Computer Assisted Radiology and Surgery, Jun 2016, Heidelberg, Germany. pp.S39 - S40, ⟨10.1007/s11548-016-1412-5⟩. ⟨hal-01354654⟩