HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pérault, C. Articles by Liehn, J.-C. Search for Related Content PubMed PubMed Citation Articles by Pérault, C. Articles by Liehn, J.-C.

Computer-Aided Intrapatient Comparison of Brain SPECT Images: The Gray-Level Normalization Issue Applied to Children with Epilepsy

¹ Nuclear Medicine and Biophysics Unit, Jean Godinot Institute, Reims, France
² Department of Nuclear Medicine, Rouen University Hospital and Henri Becquerel Center, Rouen, France
³ Department of Neuropediatrics, Saint Vincent de Paul Hospital, Paris, France
⁴ Unit 29, Institut National de la Santé et de la Recherche Médicale, Marseille, France
⁵ Department of Neurology and Neurosurgery, Reims University Hospital, Reims, France

View larger version (83K): [in a new window]   FIGURE 1. Representative DSC criterion maps obtained for 1 patient with different noise amplitudes: 2% (A), 5% (B), 8% (C), and 10% (D). Criterion value (c-axis) is drawn as function of (a-axis) and ß (b-axis), the 2 parameters of gray-level normalization model.

View larger version (18K): [in a new window]   FIGURE 2. DSC criterion value as function of noise amplitude obtained for 1 patient. Horizontal line at top (below 20,000) shows theoretic maximum criterion value, toward which criterion value should converge asymptotically with increasing noise.

View larger version (33K): [in a new window]   FIGURE 3. Correlation scatterplots with overlay of different normalization lines. Plots are shown from 3 representative patients, patients 1 (A), 4 (B), and 2 (C). Each voxel within reference volume is represented by a point whose coordinates are equal to interictal and ictal counts. These counts have been scaled to maximum interictal and ictal counts, respectively. Diagonal dashed line is normalization line obtained by method Max. All other lines through origin are normalization lines obtained by 1-parameter methods; the 2 lines deviating from origin are normalization lines obtained by 2-parameter methods.

View larger version (41K): [in a new window]   FIGURE 4. Reference images (interictal), registered and normalized images (ictal), and corresponding subtraction images (ictal–interictal), which were obtained by 3 different normalization methods (Max, Mean, and DSC-2). Images in A, B, and C are from same patients as in Figures 3A, 3B, and 3C, respectively. Isocontours drawn from reference images are superimposed on all images. Subtraction images are saturated at 30% of maximum of reference images, but no background has been removed.

View larger version (51K): [in a new window]   FIGURE 5. Reference images (interictal), registered and normalized images (ictal), and corresponding subtraction images (ictal–interictal) for same patient as in Figures 3C and 4C are shown in 3 orthogonal planes. Isocontours delimiting cerebral VOI drawn from reference images, and contours delimiting parallelepiped target VOI, are superimposed on all images.