The Superiority of Tsallis Entropy over Traditional Cost Functions for Brain MRI and SPECT Registration

Author(s)

Amaral-Silva, Henrique ^{[1, 2]} ; Wichert-Ana, Lauro ^{[3, 1, 2, 4, 5]} ; Murta, Luiz Otavio ^[6] ; Romualdo-Suzuki, Larissa ^[7] ; Itikawa, Emerson ^{[4, 5]} ; Bussato, Geraldo Filho ^[3] ; Azevedo-Marques, Paulo ^{[1, 2, 5]}

Affiliation

^[1] Univ Sao Paulo, Ctr Imaging Sci & Med Phys, Dept Internal Med, BR-14049900 Ribeirao Preto - Brazil

^[2] Univ Sao Paulo, Grad Program, Ribeirao Preto Med Sch, BR-14049900 Ribeirao Preto - Brazil

^[3] Univ Sao Paulo, Ctr Interdisciplinary Res Appl Neurosci NAPNA, BR-05403010 Sao Paulo - Brazil

^[4] Univ Sao Paulo, Nucl Med Sect, Ribeirao Preto Med Sch, BR-14049900 Ribeirao Preto - Brazil

^[5] Univ Sao Paulo, Sao Carlos Sch Engn, Bioengn Interunits Grad Program, BR-13566590 Sao Carlos, SP - Brazil

^[6] Univ Sao Paulo, Dept Comp & Math, BR-14040901 Ribeirao Preto - Brazil

^[7] UCL, Dept Comp Sci, London WC1E 6BT - England

• Document type: Journal article
• Source: Entropy; v. 16, n. 3, p. 1632-1651, MAR 2014.
• Web of Science Citations: 5

Abstract

Neuroimage registration has an important role in clinical (for both diagnostic and therapeutic purposes) and research applications. In this article we describe the applicability of Tsallis Entropy as a new cost function for neuroimage registration through a comparative analysis based on the performance of the traditional approaches (correlation based: Entropy Correlation Coefficient (ECC) and Normalized Cross Correlation (NCC); and Mutual Information (MI) based: Mutual Information using Shannon Entropy (MIS) and Normalized Mutual Information (NMI)) and the proposed one based on MI using Tsallis entropy (MIT). We created phantoms with known geometric transformations using Single Photon Emission Computed Tomography (SPECT) and Magnetic Resonance Imaging from 3 morphologically normal subjects. The simulated volumes were registered to the original ones using both the proposed and traditional approaches. The comparative analysis of the Relative Error (RE) showed that MIT was more accurate in the intramodality registration, whereas for inter-modality registration, MIT presented the lowest RE for rotational transformations, and the ECC the lowest RE for translational transformations. In conclusion, we have shown that, with certain limitations, Tsallis Entropy has application as a better cost function for reliable neuroimage registration. (AU)

• FAPESP's process: 10/10979-6 - CLINICAL VALIDATION OF USE OF TSALLIS ENTROPY IN NEUROIMAGE CO-REGISTRATION FOR THE LOCATION OF THE EPILEPTOGENIC ZONE USING THE STATISTICAL PARAMETRIC MAPPING (SPM)
• Grantee: Henrique Tomaz Do Amaral Silva
• Support Opportunities: Scholarships in Brazil - Doctorate
• FAPESP's process: 13/08240-0 - Validation of Tsallis entropy in neuroimage registration
• Grantee: Henrique Tomaz Do Amaral Silva
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• FAPESP's process: 08/03369-7 - MULTIMODALITY IMAGE REGISTRATION USING TSALLIS ENTROPY IN STATISTICAL PARAMETRIC MAPPING.
• Grantee: Henrique Tomaz Do Amaral Silva
• Support Opportunities: Scholarships in Brazil - Master