Reduction in camera-specific variability in [123I]FP-CIT SPECT outcome measures by image reconstruction optimized for multisite settings: impact on age-dependence of the specific binding ratio in the ENC-DAT database of healthy controls

Buchert, Ralph; Kluge, Andreas; Tossici-Bolt, Livia; Dickson, John; Bronzel, Marcus; Lange, Catharina; Asenbaum, S.; Booij, Jan; Atay, Lütfiye Özlem; Svarer, Claus; Koulibaly, Pierre Malick; Nobili, Flavio; Pagani, Marco; Sabri, Osama; Séra, Teréz; Tatsch, Klaus; Laere, Koen Van; Varrone, Andrea; Iida, Hidehiro

doi:10.1007/s00259-016-3309-5

Cited by 35 publications

(19 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While BRASS, based on direct measure of counts concentration from tight striatal ROIs, is susceptible to partial volume losses, the Southampton method, based on the Specific Uptake Size Index (SUSI) approach, is able to overcome them [ 5 ]. Furthermore, the respective choices for the reference region—the occipital cortex for BRASS and the whole brain without the striatal VOI for the Southampton method—will also contribute to their different outcome [ 13 ]. Their difference can be fully appreciated by comparing their respective pre-calibration graphs in Figs.…”

Section: Discussionmentioning

confidence: 99%

[123I]FP-CIT ENC-DAT normal database: the impact of the reconstruction and quantification methods

et al. 2017

Self Cite

View full text Add to dashboard Cite

Background[123I]FP-CIT is a well-established radiotracer for the diagnosis of dopaminergic degenerative disorders. The European Normal Control Database of DaTSCAN (ENC-DAT) of healthy controls has provided age and gender-specific reference values for the [123I]FP-CIT specific binding ratio (SBR) under optimised protocols for image acquisition and processing. Simpler reconstruction methods, however, are in use in many hospitals, often without implementation of attenuation and scatter corrections. This study investigates the impact on the reference values of simpler approaches using two quantifications methods, BRASS and Southampton, and explores the performance of the striatal phantom calibration in their harmonisation.ResultsBRASS and Southampton databases comprising 123 ENC-DAT subjects, from gamma cameras with parallel collimators, were reconstructed using filtered back projection (FBP) and iterative reconstruction OSEM without corrections (IRNC) and compared against the recommended OSEM with corrections for attenuation and scatter and septal penetration (ACSC), before and after applying phantom calibration. Differences between databases were quantified using the percentage difference of their SBR in the dopamine transporter-rich striatum, with their significance determined by the paired t test with Bonferroni correction.Attenuation and scatter losses, measured from the percentage difference between IRNC and ACSC databases, were of the order of 47% for both BRASS and Southampton quantifications. Phantom corrections were able to recover most of these losses, but the SBRs remained significantly lower than the “true” values (p < 0.001). Calibration provided, in fact, “first order” camera-dependent corrections, but could not include “second order” subject-dependent effects, such as septal penetration from extra-cranial activity. As for the ACSC databases, phantom calibration was instrumental in compensating for partial volume losses in BRASS (~67%, p < 0.001), while for the Southampton method, inherently free from them, it brought no significant changes and solely corrected for residual inter-camera variability (−0.2%, p = 0.44).ConclusionsThe ENC-DAT reference values are significantly dependent on the reconstruction and quantification methods and phantom calibration, while reducing the major part of their differences, is unable to fully harmonize them. Clinical use of any normal database, therefore, requires consistency with the processing methodology. Caution must be exercised when comparing data from different centres, recognising that the SBR may represent an “index” rather than a “true” value.

show abstract

Section: Discussionmentioning

confidence: 99%

[123I]FP-CIT ENC-DAT normal database: the impact of the reconstruction and quantification methods

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…The total volume in the cartridge was ~16 ml. Pixel values were set to give a striatum to brain count density ratio of 8 : 1, reflective of a normal patient 16 , 17 .…”

Section: Methodsmentioning

confidence: 99%

The subresolution DaTSCAN phantom

Taylor

Vennart

Negus

et al. 2018

Nuclear Medicine Communications

View full text Add to dashboard Cite

show abstract

“…This is due to the fact that the intensity profile of this type of images heavily depends on a variety of factors including acquisition hardware and settings, reconstruction algorithm and physiological-related factors, such as age, gender, and metabolism [5,7]. This is illustrated in Fig.…”

Section: Dat Spect Intensity Profilementioning

confidence: 99%

“…This bias has negative consequences in the analysis of these images, including (i) the limitation to pool and compare images from multiple sites, (ii) the generation of non-comparable estimates in machine learning algorithms and a resulting poor generalization, and (iii) lack of reproducibility of single-site findings. There have been recent efforts to reduce this variability in ENC-DAT dataset by revisiting the reconstruction algorithm [7]. However, this approach is far from being a straightforward step since it requires dealing with detailed camera-specific factors such as correction of scatter and septal penetration for image reconstruction, which may turn the normalization into a cumbersome process.…”

Section: Introductionmentioning

confidence: 99%

Quantitative Intensity Harmonization of Dopamine Transporter SPECT Images Using Gamma Mixture Models

et al. 2018

View full text Add to dashboard Cite

Purpose: Differences in site, device, and/or settings may cause large variations in the intensity profile of dopamine transporter (DAT) single-photon emission computed tomography (SPECT) images. However, the current standard to evaluate these images, the striatal binding ratio (SBR), does not efficiently account for this heterogeneity and the assessment can be unequivalent across distinct acquisition pipelines. In this work, we present a voxel-based automated approach to intensity normalize such type of data that improves on cross-session interpretation. Procedures: The normalization method consists of a reparametrization of the voxel values based on the cumulative density function (CDF) of a Gamma distribution modeling the specific region intensity. The harmonization ability was tested in 1342 SPECT images from the PPMI repository, acquired with 7 distinct gamma camera models and at 24 different sites. We compared the striatal quantification across distinct cameras for raw intensities, SBR values, and after applying the Gamma CDF (GDCF) harmonization. As a proof-of-concept, we evaluated the impact of GCDF normalization in a classification task between controls and Parkinson disease patients. Results: Raw striatal intensities and SBR values presented significant differences across distinct camera models. We demonstrate that GCDF normalization efficiently alleviated these differences in striatal quantification and with values constrained to a fixed interval [0,1]. Also, our method allowed a fully automated image assessment that provided maximal classification ability, given by an area under the curve (AUC) of AUC = 0.94 when used mean regional variables and AUC = 0.98 when used voxel-based variables. Conclusion: The GCDF normalization method is useful to standardize the intensity of DAT SPECT images in an automated fashion and enables the development of unbiased algorithms using multicenter datasets. This method may constitute a key pre-processing step in the analysis of this type of images.Alberto Llera and Ismael Huertas shared first authorship. Electronic supplementary material The online version of this article (https://

show abstract

Reduction in camera-specific variability in [123I]FP-CIT SPECT outcome measures by image reconstruction optimized for multisite settings: impact on age-dependence of the specific binding ratio in the ENC-DAT database of healthy controls

Cited by 35 publications

References 29 publications

[123I]FP-CIT ENC-DAT normal database: the impact of the reconstruction and quantification methods

[123I]FP-CIT ENC-DAT normal database: the impact of the reconstruction and quantification methods

The subresolution DaTSCAN phantom

Quantitative Intensity Harmonization of Dopamine Transporter SPECT Images Using Gamma Mixture Models

Contact Info

Product

Resources

About