Fully automatic diagnostic system for early- and late-onset mild Alzheimer’s disease using FDG PET and 3D-SSP

SNMMI Procedure Standard/EANM Practice Guideline for Brain [18F]FDG PET Imaging, Version 2.0

PREAMBLEThe Society of Nuclear Medicine and Molecular Imaging (SNMMI) is an international scientific and professional organization founded in 1954 to promote the science, technology, and practical application of nuclear medicine. The European Association of Nuclear Medicine (EANM) is a professional nonprofit medical association that facilitates communication worldwide between individuals pursuing clinical and research excellence in nuclear medicine. The EANM was founded in 1985. The EANM was founded in 1985. SNMMI and EANM members are physicians, technologists, and scientists specializing in the research and practice of nuclear medicine.The SNMMI and EANM will periodically define new guidelines for nuclear medicine practice to help advance the science of nuclear medicine and to improve the quality of service to patients throughout the world. Existing practice guidelines will be reviewed for revision or renewal, as appropriate, on their fifth anniversary or sooner, if indicated.Each practice guideline, representing a policy statement by the SNMMI/EANM, has undergone a thorough consensus process in which it has been subjected to extensive review. The SNMMI and EANM recognize that the safe and effective use of diagnostic nuclear medicine imaging requires specific training, skills, and techniques, as described in each document. Reproduction or modification of the published practice guideline by those entities not providing these services is not authorized.These guidelines are an educational tool designed to assist practitioners in providing appropriate care for patients. They are not inflexible rules or requirements of practice and are not intended, nor should they be used, to establish a legal standard of care. For these reasons and those set forth below, both the SNMMI and the EANM caution against the use of these guidelines in litigation in which the clinical decisions of a practitioner are called into question.The ultimate judgment regarding the propriety of any specific procedure or course of action must be made by the physician or medical physicist in light of all the circumstances presented. Thus, there is no implication that an approach differing from the guidelines, standing alone, is below the standard of care. To the contrary, a conscientious practitioner may responsibly adopt a course of action different from that set forth in the guidelines when, in the reasonable judgment of the practitioner, such course of action is indicated by the condition of the patient, limitations of available resources, or advances in knowledge or technology subsequent to publication of the guidelines.The practice of medicine includes both the art and the science of the prevention, diagnosis, alleviation, and treatment of disease. The variety and complexity of human conditions make it impossible to always reach the most appropriate diagnosis or to predict with certainty a particular response to treatment.Therefore, it should be recognized that adherence to these guidelines will not ensure an accurate diagnosis or a successful outcome. All that should be expected is that the practitioner will follow a reasonable course of action based on current knowledge, available resources, and the needs of the patient to deliver effective and safe medical care. The sole purpose of these guidelines is to assist practitioners in achieving this objective.

EANM procedure guidelines for brain PET imaging using [18F]FDG, version 3

The present procedural guidelines summarize the current views of the EANM Neuro-Imaging Committee (NIC). The purpose of these guidelines is to assist nuclear medicine practitioners in making recommendations, performing, interpreting, and reporting results of [¹⁸F]FDG-PET imaging of the brain. The aim is to help achieve a high-quality standard of [¹⁸F]FDG brain imaging and to further increase the diagnostic impact of this technique in neurological, neurosurgical, and psychiatric practice. The present document replaces a former version of the guidelines that have been published in 2009. These new guidelines include an update in the light of advances in PET technology such as the introduction of digital PET and hybrid PET/MR systems, advances in individual PET semiquantitative analysis, and current broadening clinical indications (e.g., for encephalitis and brain lymphoma). Further insight has also become available about hyperglycemia effects in patients who undergo brain [¹⁸F]FDG-PET. Accordingly, the patient preparation procedure has been updated. Finally, most typical brain patterns of metabolic changes are summarized for neurodegenerative diseases. The present guidelines are specifically intended to present information related to the European practice. The information provided should be taken in the context of local conditions and regulations.

Clinical impact of digital and conventional PET control databases for semi-quantitative analysis of brain 18F-FDG digital PET scans

PURPOSE

Digital PET cameras markedly improve sensitivity and spatial resolution of brain ¹⁸F-FDG PET images compared to conventional cameras. Our study aimed to assess whether specific control databases are required to improve the diagnostic performance of these recent advances.

METHODS

We retrospectively selected two groups of subjects, twenty-seven Alzheimer's Disease (AD) patients and twenty-two healthy control (HC) subjects. All subjects underwent a brain ¹⁸F-FDG PET on a digital camera (Vereos, Philips®). These two group (AD and HC) are compared, using a Semi-Quantitative Analysis (SQA), to two age and sex matched controls acquired with a digital PET/CT (Vereos, Philips®) or a conventional PET/CT (Biograph 6, Siemens®) camera, at group and individual levels. Moreover, individual visual interpretation of SPM T-maps was provided for the positive diagnosis of AD by 3 experienced raters.

RESULTS

At group level, SQA using digital controls detected more marked hypometabolic areas in AD (+ 116 cm³ at p < 0.001 uncorrected for the voxel, corrected for the cluster) than SQA using conventional controls. At the individual level, the accuracy of SQA for discriminating AD using digital controls was higher than SQA using conventional controls (86% vs. 80%, p < 0.01, at p < 0.005 uncorrected for the voxel, corrected for the cluster), with higher sensitivity (89% vs. 78%) and similar specificity (82% vs. 82%). These results were confirmed by visual analysis (accuracies of 84% and 82% for digital and conventional controls respectively, p = 0.01).

CONCLUSION

There is an urgent need to establish specific digital PET control databases for SQA of brain ¹⁸F-FDG PET images as such databases improve the accuracy of AD diagnosis.

Image reconstruction methods affect software-aided assessment of pathologies of [18F]flutemetamol and [18F]FDG brain-PET examinations in patients with neurodegenerative diseases

•

[¹⁸F]Flutemetamol and [¹⁸F]FDG image reconstruction.

•

Software-aided assessment of neurodegenerative disease patients.

•

New developments in brain PET image reconstruction affect quantitative measures.

•

Evaluation of SUVR and z-score measures.

•

Normalizing to pons and whole brain induced greater absolute differences between reconstructions.

Purpose

To assess how some of the new developments in brain positron emission tomography (PET) image reconstruction affect quantitative measures and software-aided assessment of pathology in patients with neurodegenerative diseases.

Methods

PET data were grouped into four cohorts: prodromal Alzheimer’s disease patients and controls receiving [¹⁸F]flutemetamol, and neurodegenerative disease patients and controls receiving [¹⁸F]FDG PET scans. Reconstructed images were obtained by ordered-subsets expectation maximization (OSEM; 3 iterations (i), 16/34 subsets (s), 3/5-mm filter, ±time-of-flight (TOF), ±point-spread function (PSF)) and block-sequential regularized expectation maximization (BSREM; TOF, PSF, β-value 75–300). Standardized uptake value ratios (SUVR) and z-scores were calculated (CortexID Suite, GE Healthcare) using cerebellar gray matter, pons, whole cerebellum and whole brain as reference regions.

Results

In controls, comparable results to the normal database were obtained with OSEM 3i/16 s 5-mm reconstruction. TOF, PSF and BSREM either increased or decreased the relative uptake difference to the normal subjects’ database within the software, depending on the tracer and chosen reference area, i.e. resulting in increased absolute z-scores. Normalizing to pons and whole brain for [¹⁸F]flutemetamol and [¹⁸F]FDG, respectively, increased absolute differences between reconstructions methods compared to normalizing to cerebellar gray matter and whole cerebellum when applying TOF, PSF and BSREM.

Conclusions

Software-aided assessment of patient pathologies should be used with caution when employing other image reconstruction methods than those used for acquisition of the normal database.

Head-to-Head Comparison among Semi-Quantification Tools of Brain FDG-PET to Aid the Diagnosis of Prodromal Alzheimer's Disease

BACKGROUND

Several automatic tools have been implemented for semi-quantitative assessment of brain [18]F-FDG-PET.

OBJECTIVE

We aimed to head-to-head compare the diagnostic performance among three statistical parametric mapping (SPM)-based approaches, another voxel-based tool (i.e., PALZ), and a volumetric region of interest (VROI-SVM)-based approach, in distinguishing patients with prodromal Alzheimer's disease (pAD) from controls.

METHODS

Sixty-two pAD patients (MMSE score = 27.0±1.6) and one hundred-nine healthy subjects (CTR) (MMSE score = 29.2±1.2) were enrolled in five centers of the European Alzheimer's Disease Consortium. The three SPM-based methods, based on different rationales, included 1) a cluster identified through the correlation analysis between [18]F-FDG-PET and a verbal memory test (VROI-1), 2) a VROI derived from the comparison between pAD and CTR (VROI-2), and 3) visual analysis of individual maps obtained by the comparison between each subject and CTR (SPM-Maps). The VROI-SVM approach was based on 6 VROI plus 6 VROI asymmetry values derived from the pAD versus CTR comparison thanks to support vector machine (SVM).

RESULTS

The areas under the ROC curves between pAD and CTR were 0.84 for VROI-1, 0.83 for VROI-2, 0.79 for SPM maps, 0.87 for PALZ, and 0.95 for VROI-SVM. Pairwise comparisons of Youden index did not show statistically significant differences in diagnostic performance between VROI-1, VROI-2, SPM-Maps, and PALZ score whereas VROI-SVM performed significantly (p < 0.005) better than any of the other methods.

CONCLUSION

The study confirms the good accuracy of [18]F-FDG-PET in discriminating healthy subjects from pAD and highlights that a non-linear, automatic VROI classifier based on SVM performs better than the voxel-based methods.

Relevance of quantification in brain PET studies with 18F-FDG

The inclusion of ¹⁸F-FDG PET as a biomarker in the diagnostic criteria of neurodegenerative diseases and its indication in the presurgical assessment for drug-resistant epilepsies allow to improve specificity of these diagnosis. The traditional interpretation of neurological PET studies has been performed qualitatively, although in the last decade, several quantitative evaluation methods have emerged. This technical development has become relevant in clinical practice, improving specificity, reproducibility and reducing the interrater reliability derived from visual analysis. In this article we update/review the main imaging processing techniques currently used. This may allow the Nuclear Medicine physician to know their advantages and disadvantages when including these procedures in daily clinical practice.

Quantifying brain metabolism from FDG-PET images into a probability of Alzheimer's dementia score

18 F-fluorodeoxyglucose positron emission tomography (FDG-PET) enables in-vivo capture of the topographic metabolism patterns in the brain. These images have shown great promise in revealing the altered metabolism patterns in Alzheimer's disease (AD). The AD pathology is progressive, and leads to structural and functional alterations that lie on a continuum. There is a need to quantify the altered metabolism patterns that exist on a continuum into a simple measure. This work proposes a 3D convolutional neural network with residual connections that generates a probability score useful for interpreting the FDG-PET images along the continuum of AD. This network is trained and tested on images of stable normal control and stable Dementia of the Alzheimer's type (sDAT) subjects, achieving an AUC of 0.976 via repeated fivefold cross-validation. An independent test set consisting of images in between the two extreme ends of the DAT spectrum is used to further test the generalization performance of the network. Classification performance of 0.811 AUC is achieved in the task of predicting conversion of mild cognitive impairment to DAT for conversion time of 0-3 years. The saliency and class activation maps, which highlight the regions of the brain that are most important to the classification task, implicate many known regions affected by DAT including the posterior cingulate cortex, precuneus, and hippocampus.

Automated assessment of FDG-PET for differential diagnosis in patients with neurodegenerative disorders

PURPOSE

To review literature until November 2015 and reach a consensus on whether automatic semi-quantification of brain FDG-PET is useful in the clinical setting for neurodegenerative disorders.

METHODS

A literature search was conducted in Medline, Embase, and Google Scholar. Papers were selected with a lower limit of 30 patients (no limits with autopsy confirmation). Consensus recommendations were developed through a Delphi procedure, based on the expertise of panelists, who were also informed about the availability and quality of evidence, assessed by an independent methodology team.

RESULTS

Critical outcomes were available in nine among the 17 papers initially selected. Only three papers performed a direct comparison between visual and automated assessment and quantified the incremental value provided by the latter. Sensitivity between visual and automatic analysis is similar but automatic assessment generally improves specificity and marginally accuracy. Also, automated assessment increases diagnostic confidence. As expected, performance of visual analysis is reported to depend on the expertise of readers.

CONCLUSIONS

Tools for semi-quantitative evaluation are recommended to assist the nuclear medicine physician in reporting brain FDG-PET pattern in neurodegenerative conditions. However, heterogeneity, complexity, and drawbacks of these tools should be known by users to avoid misinterpretation. Head-to-head comparisons and an effort to harmonize procedures are encouraged.

Clinical validity of brain fluorodeoxyglucose positron emission tomography as a biomarker for Alzheimer's disease in the context of a structured 5-phase development framework

The use of Alzheimer's disease (AD) biomarkers is supported in diagnostic criteria, but their maturity for clinical routine is still debated. Here, we evaluate brain fluorodeoxyglucose positron emission tomography (FDG PET), a measure of cerebral glucose metabolism, as a biomarker to identify clinical and prodromal AD according to the framework suggested for biomarkers in oncology, using homogenous criteria with other biomarkers addressed in parallel reviews. FDG PET has fully achieved phase 1 (rational for use) and most of phase 2 (ability to discriminate AD subjects from healthy controls or other forms of dementia) aims. Phase 3 aims (early detection ability) are partly achieved. Phase 4 studies (routine use in prodromal patients) are ongoing, and only preliminary results can be extrapolated from retrospective observations. Phase 5 studies (quantify impact and costs) have not been performed. The results of this study show that specific efforts are needed to complete phase 3 evidence, in particular comparing and combining FDG PET with other biomarkers, and to properly design phase 4 prospective studies as a basis for phase 5 evaluations.

Arterial spin labeling-based Z-maps have high specificity and positive predictive value for neurodegenerative dementia compared to FDG-PET

Objective

Cerebral perfusion analysis based on arterial spin labeling (ASL) MRI has been proposed as an alternative to FDG-PET in patients with neurodegenerative disease. Z-maps show normal distribution values relating an image to a database of controls. They are routinely used for FDG-PET to demonstrate disease-specific patterns of hypometabolism at the individual level. This study aimed to compare the performance of Z-maps based on ASL to FDG-PET.

Methods

Data were combined from two separate sites, each cohort consisting of patients with Alzheimer’s disease (n = 18 + 7), frontotemporal dementia (n = 12 + 8) and controls (n = 9 + 29). Subjects underwent pseudocontinuous ASL and FDG-PET. Z-maps were created for each subject and modality. Four experienced physicians visually assessed the 166 Z-maps in random order, blinded to modality and diagnosis.

Results

Discrimination of patients versus controls using ASL-based Z-maps yielded high specificity (84%) and positive predictive value (80%), but significantly lower sensitivity compared to FDG-PET-based Z-maps (53% vs. 96%, p < 0.001). Among true-positive cases, correct diagnoses were made in 76% (ASL) and 84% (FDG-PET) (p = 0.168).

Conclusion

ASL-based Z-maps can be used for visual assessment of neurodegenerative dementia with high specificity and positive predictive value, but with inferior sensitivity compared to FDG-PET.

Key points

• ASL-based Z-maps yielded high specificity and positive predictive value in neurodegenerative dementia.

• ASL-based Z-maps had significantly lower sensitivity compared to FDG-PET-based Z-maps.

• FDG-PET might be reserved for ASL-negative cases where clinical suspicion persists.

• Findings were similar at two study sites.

Usefulness of 3-dimensional stereotactic surface projection FDG PET images for the diagnosis of dementia

To compare diagnostic performance and confidence of a standard visual reading and combined 3-dimensional stereotactic surface projection (3D-SSP) results to discriminate between Alzheimer disease (AD)/mild cognitive impairment (MCI), dementia with Lewy bodies (DLB), and frontotemporal dementia (FTD).[F]fluorodeoxyglucose (FDG) PET brain images were obtained from 120 patients (64 AD/MCI, 38 DLB, and 18 FTD) who were clinically confirmed over 2 years follow-up. Three nuclear medicine physicians performed the diagnosis and rated diagnostic confidence twice; once by standard visual methods, and once by adding of 3D-SSP. Diagnostic performance and confidence were compared between the 2 methods.3D-SSP showed higher sensitivity, specificity, accuracy, positive, and negative predictive values to discriminate different types of dementia compared with the visual method alone, except for AD/MCI specificity and FTD sensitivity. Correction of misdiagnosis after adding 3D-SSP images was greatest for AD/MCI (56%), followed by DLB (13%) and FTD (11%). Diagnostic confidence also increased in DLB (visual: 3.2; 3D-SSP: 4.1; P < 0.001), followed by AD/MCI (visual: 3.1; 3D-SSP: 3.8; P = 0.002) and FTD (visual: 3.5; 3D-SSP: 4.2; P = 0.022). Overall, 154/360 (43%) cases had a corrected misdiagnosis or improved diagnostic confidence for the correct diagnosis.The addition of 3D-SSP images to visual analysis helped to discriminate different types of dementia in FDG PET scans, by correcting misdiagnoses and enhancing diagnostic confidence in the correct diagnosis. Improvement of diagnostic accuracy and confidence by 3D-SSP images might help to determine the cause of dementia and appropriate treatment.

Comparative Study of Voxel-Based Epileptic Foci Localization Accuracy between Statistical Parametric Mapping and Three-dimensional Stereotactic Surface Projection

Introduction

Fluorine-18-fluorodeoxyglucose positron-emission tomography (¹⁸F-FDG-PET) is widely used to help localize the hypometabolic epileptogenic focus for presurgical evaluation of drug-refractory epilepsy patients. Two voxel-based brain mapping methods to interpret ¹⁸F-FDG-PET, statistical parametric mapping (SPM) and three-dimensional stereotactic surface projection (3D-SSP), improve the detection rate of seizure foci. This study aimed to compare the consistency of epileptic focus detection between SPM and 3D-SSP for ¹⁸F-FDG-PET brain mapping analysis.

Methods

We retrospectively reviewed the clinical, electroecephalographic, and brain imaging results of 35 patients with refractory epilepsy. ¹⁸F-FDG-PET studies were revaluated by SPM, 3D-SSP, and visual assessment, and the results were compared to the magnetic resonance imaging (MRI) lesion location and to the presumed epileptogenic zone (PEZ) defined by video-electroencephalogram and other clinical data. A second consistency study compared PET analyses to histopathology and surgical outcomes in the 19 patients who underwent lesion resection surgery.

Results

Of the 35 patients, consistency with the PEZ was 29/35 for SPM, 25/35 for 3D-SSP, 14/35 for visual assessment, and 10/35 for MRI. Concordance rates with the PEZ were significantly higher for SPM and 3D-SSP than for MRI (P < 0.05) and visual assessment (P < 0.05). Differences between SPM and 3D-SSP and between visual assessment and MRI were not significant. In the 19 surgical patients, concordance with histopathology/clinical outcome was 14/19 for SPM, 15/19 for 3D-SSP, 14/19 for visual assessment, and 9/19 for MRI (P > 0.05). A favorable Engel outcome (class I/II) was found in 16 of 19 cases (84%), and failure of seizure control was found in 3 of 19 patients (class III/IV).

Conclusion

Voxel-based ¹⁸F-FDG-PET brain mapping analysis using SPM or 3D-SSP can improve the detection rate of the epileptic focus compared to visual assessment and MRI. Consistency with PEZ was similar between SPM and 3D-SSP; according to their own characteristics, 3D-SSP is recommended for primary evaluation due to greater efficiency and operability of the software, while SPM is recommended for high-accuracy localization of complex lesions. Therefore, joint application of both software packages may be the best solution for FDG-PET analysis of epileptic focus localization.

Brain fluorodeoxyglucose (FDG) PET in dementia

The purpose of this article is to present a selective and concise summary of fluorodeoxyglucose (FDG) positron emission tomography (PET) in dementia imaging. FDG PET is used to visualize a downstream topographical marker that indicates the distribution of neural injury or synaptic dysfunction, and can identify distinct phenotypes of dementia due to Alzheimer's disease (AD), Lewy bodies, and frontotemporal lobar degeneration. AD dementia shows hypometabolism in the parietotemporal association area, posterior cingulate, and precuneus. Hypometabolism in the inferior parietal lobe and posterior cingulate/precuneus is a predictor of cognitive decline from mild cognitive impairment (MCI) to AD dementia. FDG PET may also predict conversion of cognitively normal individuals to those with MCI. Age-related hypometabolism is observed mainly in the anterior cingulate and anterior temporal lobe, along with regional atrophy. Voxel-based statistical analyses, such as statistical parametric mapping or three-dimensional stereotactic surface projection, improve the diagnostic performance of imaging of dementias. The potential of FDG PET in future clinical and methodological studies should be exploited further.

Brain imaging in the differential diagnosis of young-onset dementias

Young-onset dementia is a broad category of diseases that affect adults before the age of 65, with devastating effects on individuals and families. Neuroimaging plays a clear and ever-expanding role in the workup of these diseases. MRI demonstrates classic patterns of atrophy that help to confirm the clinical diagnosis and may predict the underlying disease. Functional nuclear imaging, such as PET, demonstrates areas of brain dysfunction even in the absence of visible atrophy. These techniques can inform important aspects of the care of young-onset dementia, such as the underlying pathologic condition, treatment, and prognosis.

Volume of interest-based [18F]fluorodeoxyglucose PET discriminates MCI converting to Alzheimer's disease from healthy controls. A European Alzheimer's Disease Consortium (EADC) study

An emerging issue in neuroimaging is to assess the diagnostic reliability of PET and its application in clinical practice. We aimed at assessing the accuracy of brain FDG-PET in discriminating patients with MCI due to Alzheimer's disease and healthy controls. Sixty-two patients with amnestic MCI and 109 healthy subjects recruited in five centers of the European AD Consortium were enrolled. Group analysis was performed by SPM8 to confirm metabolic differences. Discriminant analyses were then carried out using the mean FDG uptake values normalized to the cerebellum computed in 45 anatomical volumes of interest (VOIs) in each hemisphere (90 VOIs) as defined in the Automated Anatomical Labeling (AAL) Atlas and on 12 meta-VOIs, bilaterally, obtained merging VOIs with similar anatomo-functional characteristics. Further, asymmetry indexes were calculated for both datasets. Accuracy of discrimination by a Support Vector Machine (SVM) and the AAL VOIs was tested against a validated method (PALZ). At the voxel level SMP8 showed a relative hypometabolism in the bilateral precuneus, and posterior cingulate, temporo-parietal and frontal cortices. Discriminant analysis classified subjects with an accuracy ranging between .91 and .83 as a function of data organization. The best values were obtained from a subset of 6 meta-VOIs plus 6 asymmetry values reaching an area under the ROC curve of .947, significantly larger than the one obtained by the PALZ score. High accuracy in discriminating MCI converters from healthy controls was reached by a non-linear classifier based on SVM applied on predefined anatomo-functional regions and inter-hemispheric asymmetries. Data pre-processing was automated and simplified by an in-house created Matlab-based script encouraging its routine clinical use. Further validation toward nonconverter MCI patients with adequately long follow-up is needed.

•

¹⁸F-FDG-PET/CT analysis of metabolic differences between MCI converting to AD and HC

•

Large and very well controlled cohorts from EADC-Consortium were investigated.

•

Data were analyzed by a friendly-to-use Matlab-based script and Support Vector Machine.

•

Excellent discrimination between MCI and HC (sensitivity 92%; specificity 91%)

•

Highest accuracy reported so far in MCI and promising implementation in clinical routine

Quantitative assessment of rest and acetazolamide CBF using quantitative SPECT reconstruction and sequential administration of (123)I-iodoamphetamine: comparison among data acquired at three institutions

PURPOSE

A recently developed technique which reconstructs quantitative images from original projection data acquired using existing single-photon emission computed tomography (SPECT) devices enabled quantitative assessment of cerebral blood flow (CBF) at rest and after acetazolamide challenge. This study was intended to generate a normal database and to investigate its inter-institutional consistency.

METHODS

The three institutions carried out a series of SPECT scanning on 32 healthy volunteers, following a recently proposed method that involved dual administration of (123)I-iodoamphetamine during a single SPECT scan. Intra-institute and inter-institutional variations of regional CBF values were evaluated both at rest and after acetazolamide challenge. Functional images were pooled for both rest and acetazolamide CBF, and inter-institutional difference was evaluated among these images using two independent software programs.

RESULTS

Quantitative assessment of CBF images at rest and after acetazolamide was successfully achieved with the given protocol in all institutions. Intra-institutional variation of CBF values at rest and after acetazolamide was consistent with previously reported values. Quantitative CBF values showed no significant difference among institutions in all regions, except for a posterior cerebral artery region after acetazolamide challenge in one institution which employed SPECT device with lowest spatial resolution. Pooled CBF images at rest and after acetazolamide generated using two software programs showed no institutional differences after equalization of the spatial resolution.

CONCLUSIONS

SPECT can provide reproducible images from projection data acquired using different SPECT devices. A common database acquired at different institutions may be shared among institutions, if images are reconstructed using a quantitative reconstruction program, and acquired by following a standardized protocol.

Evaluation of the difference-correction effect of the gamma camera systems used by easy Z-score Imaging System (eZIS) analysis

Objective

We examined the difference of the effect by data to revise a gamma camera difference. The difference-correction method of the camera is incorporated in eZIS analysis.

Methods

We acquired single photon emission computed tomography (SPECT) data from the three-dimensional (3D) Hoffman brain phantom (Hoffman), the three-dimensional brain phantom (3D-Brain), Pool phantom (pool) and from normal subjects (Normal-SPECT) to investigate compensating for a difference in gamma camera systems. We compared SPECT counts of standard camera with the SPECT counts that revised the difference of the gamma camera system (camera). Furthermore, we compared the “Z-score map (Z-score)”. To verify the effect of the compensation, we examined digitally simulated data designed to represent a patient with Alzheimer’s dementia. We carried out both eZIS analysis and “Specific Volume of interest Analysis (SVA)”.

Results

There was no great difference between the correction effect using Hoffman phantom data and that using 3D-Brain phantom data. Furthermore, a good compensation effect was obtained only over a limited area. The compensation based on the pool was found to be less satisfactory than any of the other compensations according to all results of the measurements examined in the study. The compensation based on the Normal-SPECT data resulted in a Z-score map (Z-score) for the result that approximated that from the standard camera. Therefore, we concluded that the effect of the compensation based on Normal-SPECT data was the best of the four methods tested.

Conclusions

Based on eZIS analysis, the compensation using the pool data was inferior to the compensations using the other methods tested. Based on the results of the SAV analysis, the effect of the compensation using the Hoffman data was better than the effect of the compensation using the 3D-Brain data. By all end-point measures, the compensation based on the Normal-SPECT data was more accurate than the compensation based on any of the other three phantoms.

PET approaches for diagnosis of dementia

There is increasing use of neuroimaging modalities, including PET, for diagnosing dementia. For example, FDG-PET demonstrates hypometabolic regions in the posterior cingulate gyri, precuneus, and parietotemporal association cortices, while amyloid PET indicates amyloid deposition in Alzheimer disease and mild cognitive impairment due to Alzheimer disease. Furthermore, the use of combination PET with structural MR imaging can improve the diagnostic accuracy of dementia. In other neurodegenerative dementias, each disease exhibits a specific metabolic reduction pattern. In dementia with Lewy bodies, occipital glucose metabolism is decreased, while in frontotemporal dementia, frontal and anterior temporal metabolism is predominantly decreased. These FDG-PET findings and positive or negative amyloid deposits are important biomarkers for various neurodegenerative dementias.

Bilateral symmetry aspects in computer-aided Alzheimer's disease diagnosis by single-photon emission-computed tomography imaging

OBJECTIVE

This paper explores the importance of the latent symmetry of the brain in computer-aided systems for diagnosing Alzheimer's disease (AD). Symmetry and asymmetry are studied from two points of view: (i) the development of an effective classifier within the scope of machine learning techniques, and (ii) the assessment of its relevance to the AD diagnosis in the early stages of the disease.

METHODS

The proposed methodology is based on eigenimage decomposition of single-photon emission-computed tomography images, using an eigenspace extension to accommodate odd and even eigenvectors separately. This feature extraction technique allows for support-vector-machine classification and image analysis.

RESULTS

Identification of AD patterns is improved when the latent symmetry of the brain is considered, with an estimated 92.78% accuracy (92.86% sensitivity, 92.68% specificity) using a linear kernel and a leave-one-out cross validation strategy. Also, asymmetries may be used to define a test for AD that is very specific (90.24% specificity) but not especially sensitive.

CONCLUSIONS

Two main conclusions are derived from the analysis of the eigenimage spectrum. Firstly, the recognition of AD patterns is improved when considering only the symmetric part of the spectrum. Secondly, asymmetries in the hypo-metabolic patterns, when present, are more pronounced in subjects with AD.

NMF-SVM based CAD tool applied to functional brain images for the diagnosis of Alzheimer's disease

This paper presents a novel computer-aided diagnosis (CAD) technique for the early diagnosis of the Alzheimer's disease (AD) based on nonnegative matrix factorization (NMF) and support vector machines (SVM) with bounds of confidence. The CAD tool is designed for the study and classification of functional brain images. For this purpose, two different brain image databases are selected: a single photon emission computed tomography (SPECT) database and positron emission tomography (PET) images, both of them containing data for both Alzheimer's disease (AD) patients and healthy controls as a reference. These databases are analyzed by applying the Fisher discriminant ratio (FDR) and nonnegative matrix factorization (NMF) for feature selection and extraction of the most relevant features. The resulting NMF-transformed sets of data, which contain a reduced number of features, are classified by means of a SVM-based classifier with bounds of confidence for decision. The proposed NMF-SVM method yields up to 91% classification accuracy with high sensitivity and specificity rates (upper than 90%). This NMF-SVM CAD tool becomes an accurate method for SPECT and PET AD image classification.

[Availability of normal database by single photon emission computed tomography system with use of 3 dimensional-stereotactic surface projections]

PURPOSE

The present study aims to quantitatively investigate a normal database (NDB) created under the same acquisition and reconstruction conditions for three gamma camera systems (four types of collimator systems) with use of three-dimensional stereotactic surface projections (3D-SSP). We rebuilt a NDB with use of the N-isopropyl-p-(123)I-iodoamphetamine ((123)I-IMP) SPECT data derived from 30 healthy individuals at 20 institutions nationwide. We standardized the acquisition and reconstruction conditions, evaluated Z scores using patient data (PD) and examined each compensation effect.

RESULTS

Z scores determined using the advanced NDB were the same value. Artifacts were often generated in Z score maps derived from the conventional NDB (CONDB). The Z score of the own site NDB (OWNDB) was 70% of that calculated based on the CONDB. The combinatorial difference in compensation (scatter and attenuation) resulted in many artifacts being generated in Z score map images.

DISCUSSIONS

More artifacts were generated in Z score map images using the novel NDB compared with the CONDB. The novel NDB was comparable to the performance of OWNB. The accuracy of brain function image analysis can be improved the reconstruction conditions and correcting for scatter and attenuation on both the novel NDB and PD.

Observation of cortical activity during speech stimulation in prelingually deafened adults with cochlear implantation by positron emission tomography-computed tomography

OBJECTIVES

We evaluated the cortical activity of 2 successful prelingually deafened adult cochlear implant (CI) users who have been trained by auditory-verbal/oral communication since childhood.

METHODS

Changes in regional cerebral blood flow were measured by positron emission tomography using '8F-fluorodeoxyglucose while the subjects were receiving auditory language stimuli by listening to a story. Ten normal-hearing volunteers were observed as age-matched control subjects.

RESULTS

In both cases, the auditory-related regions, when compared to same regions in the control subjects, showed hypermetabolism in the left dorsolateral prefrontal cortex and the left precentral gyrus--similar to that in successful CI users who are prelingually deafened children or postlingually deafened adults. Both subjects had the ability to activate these areas, and this ability might be one of the reasons that accounts for such exceptionally good performance in older prelingually deaf CI users. As for the visual-related regions, hypometabolism was observed in Brodmann areas 18 and 19, and this finding might be related to the intensive auditory-verbal/oral education that the subjects had received since childhood.

CONCLUSIONS

Despite the limits imposed by the small sample size and the spatial resolution of positron emission tomography, this study yielded insights into the nature of the brain plasticity in prelingually deafened adults who are successful CI users.

Perfusion SPECT and FDG-PET

Both perfusion SPECT and FDG-PET provide images that closely reflect neuronal activity. There is a characteristic regional impairment in Alzheimer’s disease (AD) that involves mainly the temporo-parietal association cortices, mesial temporal structures and to a more variable degree also the frontal association cortex. This pattern of functional impairment can provide a biomarker for diagnosis of AD and other neurodegenerative dementias at the clinical stage of mild cognitive impairment, and for monitoring of progression. FDG-PET is quantitatively more accurate and thus better suited to multicenter studies than perfusion SPECT. Regional metabolic and blood flow changes are closely related to clinical symptoms, and most areas involved in these changes will also develop significant cortical atrophy. FDG-PET is complementary to amyloid PET, which targets a molecular marker that does not have a close relation to current symptoms. Current restrictions in the availability and cost of FDG-PET are being reduced, as oncological FDG-PET is being adopted as a standard clinical service in most countries. Limitations in the availability of trained staff should be overcome by training programs set up by professional organizations. Against the background of the development of new criteria for diagnosing AD before the onset of dementia, FDG-PET is expected to play an increasing role in diagnosing patients at an early stage of AD and in clinical trials of drugs aimed at preventing or delaying the onset of dementia.

Posterior cingulate atrophy and metabolic decline in early stage Alzheimer's disease

To test the hypothesis that Alzheimer's disease (AD) patients with posterior cingulate/precuneus (PCP) atrophy would be a distinct disease form in view of metabolic decline. Eighty-one AD patients underwent (18)F-fluorodeoxyglucose (FDG) positron emission tomography (PET) and structural magnetic resonance imaging (MRI). Positron emission tomography and voxel-based morphometry (VBM) Z-score maps were generated for the individual patients using age-specific normal databases. The patients were classified into 3 groups based on atrophic patterns (no-Hipp-PCP, atrophy in neither hippocampus nor PCP; Hipp, hippocampal atrophy; PCP, PCP atrophy). There were 16 patients classified as no-Hipp-PCP, 55 as Hipp, and 10 as PCP. The Mini Mental State Examination (MMSE) score was similar among the groups. The greater FDG decline than atrophy was observed in all groups, including the no-Hipp-PCP. The PCP group was younger, and was associated with a greater degree of FDG decline in PCP than the others. There are diverse atrophic patterns in a spectrum of AD. In particular, a subset of patients show PCP atrophy, which is associated with greater metabolic burden.

Computer-assisted system for diagnosis of Alzheimer disease using data base- independent estimation and fluorodeoxyglucose- positron-emission tomography and 3D-stereotactic surface projection

BACKGROUND AND PURPOSE

Recently, voxel-based statistical parametric images have been developed as additional diagnostic tools for AD. However these methods require the generation of a data base of healthy brain images. The purpose of this study was to produce and evaluate an automatic method using a data base-independent estimation system for the diagnosis of mild AD.

MATERIALS AND METHODS

We retrospectively selected 66 subjects, including 33 patients with early AD and 33 age-matched healthy volunteers. Individual brain metabolic images were obtained by using FDG-PET. These were transformed by using 3D-SSP. We then produced CADDIES, which compares the parietal and sensorimotor metabolic counts by using t tests. If parietal metabolism was significantly lower than the sensorimotor metabolism, the subject was automatically diagnosed as having AD. The FDG-PET images were also analyzed by using a previous automatic diagnosis system (CAAD) that is dependent on the construction of a "normal data base" of healthy brain images. Diagnostic performance was compared between the 2 methods.

RESULTS

The CADDIES demonstrated a sensitivity of 88%, specificity of 79%, and accuracy of 85%, while the CAAD system demonstrated a sensitivity of 70%, specificity of 94%, and accuracy of 82%. The area under the ROC curve of CADDIES was 0.964. The areas under ROC curves of the CAAD method in the parietal and posterior cingulate gyri were 0.843 and 0.939, respectively.

CONCLUSIONS

The CADDIES method demonstrated a diagnostic accuracy similar to that of the CAAD system. Our results indicate that this method can be applied to the detection of patients with early AD in routine clinical examinations, with the benefit of not requiring the generation of a normal data base.

Comparative study of anatomical normalization errors in SPM and 3D-SSP using digital brain phantom

OBJECTIVE

In single photon emission computed tomography (SPECT) cerebral blood flow studies, two major algorithms are widely used statistical parametric mapping (SPM) and three-dimensional stereotactic surface projections (3D-SSP). The aim of this study is to compare an SPM algorithm-based easy Z score imaging system (eZIS) and a 3D-SSP system in the errors of anatomical standardization using 3D-digital brain phantom images.

METHODS

We developed a 3D-brain digital phantom based on MR images to simulate the effects of head tilt, perfusion defective region size, and count value reduction rate on the SPECT images. This digital phantom was used to compare the errors of anatomical standardization by the eZIS and the 3D-SSP algorithms.

RESULTS

While the eZIS allowed accurate standardization of the images of the phantom simulating a head in rotation, lateroflexion, anteflexion, or retroflexion without angle dependency, the standardization by 3D-SSP was not accurate enough at approximately 25° or more head tilt. When the simulated head contained perfusion defective regions, one of the 3D-SSP images showed an error of 6.9% from the true value. Meanwhile, one of the eZIS images showed an error as large as 63.4%, revealing a significant underestimation.

CONCLUSION

When required to evaluate regions with decreased perfusion due to such causes as hemodynamic cerebral ischemia, the 3D-SSP is desirable. In a statistical image analysis, we must reconfirm the image after anatomical standardization by all means.

Analysis of SPECT brain images for the diagnosis of Alzheimer's disease based on NMF for feature extraction

This letter presents a novel computer-aided diagnosis (CAD) technique for the early diagnosis of Alzheimer's disease (AD) based on non-negative matrix factorization (NMF) analysis applied to single photon emission computed tomography (SPECT) images. A baseline normalized SPECT database containing normalized data for both AD patients and healthy reference patients is selected for this study. The SPECT database is analyzed by applying the Fisher discriminant ratio (FDR) for feature selection and NMF for feature extraction of relevant components of each subject. The main goal of these preprocessing steps is to reduce the large dimensionality of the input data and to relieve the so called "curse of dimensionality" problem. The resulting NMF-transformed set of data, which contains a reduced number of features, is classified by means of a support vector machines based classification technique (SVM). The proposed NMF + SVM method yields up to 94% classification accuracy, with high sensitivity and specificity values (upper than 90%), becoming an accurate method for SPECT image classification. For the sake of completeness, comparison between another recently developed principal component analysis (PCA) plus SVM method and the proposed method is also provided, yielding results for the NMF + SVM approach that outperform the behavior of the reference PCA + SVM or conventional voxel-as-feature (VAF) plus SVM methods.

Computer aided diagnosis system for the Alzheimer's disease based on partial least squares and random forest SPECT image classification

This letter shows a computer aided diagnosis (CAD) technique for the early detection of the Alzheimer's disease (AD) by means of single photon emission computed tomography (SPECT) image classification. The proposed method is based on partial least squares (PLS) regression model and a random forest (RF) predictor. The challenge of the curse of dimensionality is addressed by reducing the large dimensionality of the input data by downscaling the SPECT images and extracting score features using PLS. A RF predictor then forms an ensemble of classification and regression tree (CART)-like classifiers being its output determined by a majority vote of the trees in the forest. A baseline principal component analysis (PCA) system is also developed for reference. The experimental results show that the combined PLS-RF system yields a generalization error that converges to a limit when increasing the number of trees in the forest. Thus, the generalization error is reduced when using PLS and depends on the strength of the individual trees in the forest and the correlation between them. Moreover, PLS feature extraction is found to be more effective for extracting discriminative information from the data than PCA yielding peak sensitivity, specificity and accuracy values of 100%, 92.7%, and 96.9%, respectively. Moreover, the proposed CAD system outperformed several other recently developed AD CAD systems.

Performance of FDG PET for detection of Alzheimer's disease in two independent multicentre samples (NEST-DD and ADNI)

AIM

We investigated the performance of FDG PET using an automated procedure for discrimination between Alzheimer's disease (AD) and controls, and studied the influence of demographic and technical factors.

METHODS

FDG PET data were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) [102 controls (76.0 +/- 4.9 years) and 89 AD patients (75.7 +/- 7.6 years, MMSE 23.5 +/- 2.1) and the Network for Standardisation of Dementia Diagnosis (NEST-DD) [36 controls (62.2 +/- 5.0 years) and 237 AD patients (70.8 +/- 8.3 years, MMSE 20.9 +/- 4.4). The procedure created t-maps of abnormal voxels. The sum of t-values in predefined areas that are typically affected by AD (AD t-sum) provided a measure of scan abnormality associated with a preset threshold for discrimination between patients and controls.

RESULTS

AD patients had much higher AD t-sum scores compared to controls (p < 0.01), which were significantly related to dementia severity (ADNI: r = -0.62, p < 0.01; NEST-DD: r = -0.59, p < 0.01). Early-onset AD patients had significantly higher AD t-sum scores than late-onset AD patients (p < 0.01). Differences between databases were mainly due to different age distributions. The predefined AD t-sum threshold yielded a sensitivity and specificity of 83 and 78% in ADNI and 78 and 94% in NEST-DD, respectively.

CONCLUSION

The automated FDG PET analysis procedure provided good discrimination power, and was most accurate for early-onset AD.

Diagnostic performance of Tc-99m HMPAO SPECT for early and late onset Alzheimer's disease: a clinical evaluation of linearization correction

OBJECTIVE

This study examined the influence of linearization correction (LC) on brain perfusion single-photon emission computed tomography (SPECT) for the diagnosis of Alzheimer's disease (AD).

METHODS

The early onset group (<65 years old) consisted of 10 patients with AD, and the late onset group (>/=65 years old) of 13 patients with AD. Age-matched controls included seven younger and seven older normal volunteers. Tc-99m hexamethyl propyleneamine oxine (HMPAO) SPECT images were reconstructed with or without LC [LC (+) or LC (-)] and a statistical analysis was performed using a three-dimensional stereotactic surface projection (3D-SSP). In addition, a fully automatic diagnostic system was developed, which calculated the proportion of the number of abnormal pixels in the superior and inferior parietal lobule, as well as in the precuneus and posterior cingulate gyrus.

RESULTS

The areas under the receiver-operating characteristic curve (AUCs) of the early onset group for conventional axial SPECT images, SPECT + 3D-SSP images and the fully automatic diagnostic system were 0.71, 0.88, and 0.92 in LC (-) and 0.67, 0.85, and 0.91 in LC (+), respectively. The AUCs of the late onset group were 0.50, 0.61, and 0.79 in LC (-) and 0.49, 0.67, and 0.85 in LC (+), respectively.

CONCLUSION

LC on Tc-99m HMPAO SPECT did not significantly influence the diagnostic performance for differentiating between AD and normal controls in either early or late onset AD. Further examination with individuals suffering from very mild dementia is, therefore, expected to elucidate the effect of LC on minimally hypoperfused areas.

Computer-assisted diagnostic system for neurodegenerative dementia using brain SPECT and 3D-SSP

PURPOSE

To develop a computer-assisted automated diagnostic system to distinguish among Alzheimer disease (AD), dementia with Lewy bodies (DLB), and other degenerative disorders in patients with mild dementia.

METHODS

Single photon emission computed tomography (SPECT) images with injection of N-Isopropyl-p-[(123)I]iodoamphetamine (IMP) were obtained from patients with mild degenerative dementia. First, datasets from 20 patients mild AD, 15 patients with dementia with DLB, and 17 healthy controls were used to develop an automated diagnosing system based on three-dimensional stereotactic surface projections (3D-SSP). AD- and DLB-specific regional templates were created using 3D-SSP, and critical Z scores in the templates were established. Datasets from 50 AD patients, 8 DLB patients, and 10 patients with non-AD/DLB type degenerative dementia (5 with frontotemporal dementia and 5 with progressive supranuclear palsy) were then used to test the diagnostic accuracy of the optimized automated system in comparison to the diagnostic interpretation of conventional IMP-SPECT images. These comparisons were performed to differentiate AD and DLB from non-AD/DLB and to distinguish AD from DLB. A receiver operating characteristic (ROC) analysis was performed.

RESULTS

The area under the ROC curve (Az) and the accuracy of the automated diagnosis system were 0.89 and 82%, respectively, for AD/DLB vs. non-AD/DLB patients, and 0.70 and 65%, respectively, for AD vs. DLB patients. The mean Az and the accuracy of the visual inspection were 0.84 and 77%, respectively, for AD/DLB vs. non-AD/DLB patients, and 0.70 and 65%, respectively, for AD vs. DLB patients. The mean Az and the accuracy of the combination of visual inspection and this system were 0.96 and 91%, respectively, for AD/DLB vs. non-AD/DLB patients, and 0.70 and 66%, respectively, for AD vs. DLB patients.

CONCLUSION

The system developed in the present study achieved as good discrimination of AD, DLB, and other degenerative disorders in patients with mild dementia as the commonly performed visual inspection of conventional SPECT images. A combination of visual inspection and this system is helpful in the differential diagnosis of patients with mild dementia.

Regional cerebral blood flow correlates of visuospatial tasks in Alzheimer's disease

This study investigated the role of visuospatial tasks in identifying cognitive decline in patients with Alzheimer's disease (AD), by correlating neuropsychological performance with cerebral perfusion measures. There were 157 participants: 29 neurologically healthy controls (age: 70.3 +/- 6.6, MMSE > or = 27), 86 patients with mild AD (age: 69.18 +/- 8.28, MMSE > or = 21) and 42 patients moderate/severe AD (age: 68.86 +/- 10.69, MMSE 8-20). Single Photon Emission Computerized Tomography (SPECT) was used to derive regional perfusion ratios, and correlated using partial least squares (PLS) with neuropsychological test scores from the Benton Line Orientation (BLO) and the Rey-Osterrieth Complex Figure (RO). Cross-sectional analysis demonstrated that mean scores differed in accordance with disease status: control group (BLO 25.5, RO 33.3); mild AD (BLO 20.1, RO 25.5); moderate/severe AD (BLO 10.7, RO 16). Correlations were observed between BLO/RO and right parietal SPECT regions in the AD groups. Visuospatial performance, often undersampled in cognitive batteries for AD, is clearly impaired even in mild AD and correlates with functional deficits as indexed by cerebral perfusion ratios on SPECT implicating right hemisphere circuits. Furthermore, PLS reveals that usual spatial tasks probe a distributed brain network in both hemispheres including many areas targeted by early AD pathology.

Effect of sample size for normal database on diagnostic performance of brain FDG PET for the detection of Alzheimer's disease using automated image analysis

OBJECTIVE

To investigate the relationship between the sample size for a normal database (NDB) and diagnostic performance of FDG PET using three-dimensional stereotactic surface projection for the detection of Alzheimer's disease.

METHODS

We generated nine NDB sets consisting of 4, 6, 8, 10, 20, 30, 40, 50 and 60 normal subjects. In order to assess the diagnostic performance using these NDBs to distinguish Alzheimer's disease patients from normal subjects, we recruited 52 patients with probable Alzheimer's disease (25 males, 27 females; mean age, 66.8+/-8.1 years) and 50 normal subjects (24 males, 26 females; mean age, 65.7+/-9.4 years). A receiver operating characteristic (ROC) analysis was performed for comparison of diagnostic accuracy among NDB sets.

RESULTS

Small NDBs (n< or =10) yielded poor quality of mean and SD images as compared with large NDBs (n> or =20). The ROC curves of the smaller group varied inconsistently, whereas those of the larger group were nearly superimposable. The area under the ROC curve (AUC) of the NDBs with sample size 6 (0.911) or 8 (0.929) was significantly smaller than that of the largest NDB (n=60, 0.956). The AUCs of the larger group did not fall below 0.950, whereas AUCs of the smaller subgroup never exceeded 0.950.

CONCLUSIONS

Our data indicate that the sample size for an NDB affects the diagnostic performance of FDG PET using automated statistical approach, and that inclusion of at least 10 subjects is recommended, and 20 seems to be preferable for generating NDBs, although even a small NDB may provide clinically relevant results.

Direct comparison study between FDG-PET and IMP-SPECT for diagnosing Alzheimer's disease using 3D-SSP analysis in the same patients

PURPOSE

The purpose of this study was to evaluate and compare the diagnostic ability of 2-[(18)F]-fluoro-2-deoxy-D: -glucose (FDG) positron emission tomography (PET) and N-isopropyl-p-(123)I iodoamphetamine single photon emission computed tomography (IMP-SPECT) using three-dimensional stereotactic surface projections (3D-SSP) in patients with moderate Alzheimer's disease (AD).

MATERIALS AND METHODS

FDG-PET and IMP-SPECT were performed within 3 months in 14 patients with probable moderate AD. Z-score maps of FDG-PET and IMP-SPECT images of a patient were obtained by comparison with data obtained from control subjects. Four expert physicians evaluated and graded the glucose hypometabolism and regional cerebral blood flow (rCBF), focusing in particular on the posterior cingulate gyri/precunei and parietotemporal regions, and determined the reliability for AD. Receiver operating characteristic (ROC) curves were applied to the results for clarification. To evaluate the correlation between two modalities, the regions of interest (ROIs) were set in the posterior cingulate gyri/precunei and parietotemporal region on 3D-SSP images, and mean Z-values were calculated.

CONCLUSION

No significant difference was observed in the area under the ROC curve (AUC) between FDG-PET and IMP-SPECT images (FDG-PET 0.95, IMP-SPECT 0.94). However, a significant difference (P < 0.05) was observed in the AUC for the posterior cingulate gyri/precuneus (FDG-PET 0.94, IMP-SPECT 0.81). The sensitivity and specificity of each modality were 86%, and 97% for FDG-PET and 70% and 100% for IMP-SPECT. We could find no significant difference between FDG-PET and IMP-SPECT in terms of diagnosing moderate AD using 3D-SSP. There was a high correlation between the two modalities in the parietotemporal region (Spearman's r = 0.82, P < 0.001). The correlation in the posterior cingulate gyri/precunei region was lower than that in the parietotemporal region (Spearman's r = 0.63, P < 0.016).

Imaging biomarkers and their role in dementia clinical trials

There are five potential major roles for neuroimaging with respect to dementia; (1) as a cognitive neuroscience research tool, (2) for prediction of which normal or slightly impaired individuals will develop dementia and over what time frame, (3) for early diagnosis of Alzheimer's disease (AD) in demented individuals, (sensitivity) and separation of AD from other forms of dementia (specificity), (4) for monitoring of disease progression, and (5) for monitoring response to therapies. Focusing on the last role, no single imaging approach is yet ideal, as all trade-off speed, cost, and accuracy. Functional imaging (SPECT and PET) is best suited to tracking symptomatic therapy response, and anatomic (MRI volumetric) imaging or amyloid PET are more suited to reflect dementia modulation studies. The potential for imaging with respect to pharmacological studies of dementia--to provide surrogate markers for drug studies, to improve diagnosis, to speed evaluation of outcomes, and to decrease sample sizes--is huge. At the present time, however, no single measure has sufficient proven reliability, replicability, or robustness, to replace clinical primary outcome measures.

Fully automatic differential diagnosis system for dementia with Lewy bodies and Alzheimer’s disease using FDG-PET and 3D-SSP

PURPOSE

To evaluate a fully automatic computer-assisted diagnostic system for mild dementia with Lewy bodies (DLB), permitting distinction from mild Alzheimer's disease (AD).

METHODS

Using 18F-fluorodeoxyglucose and positron emission tomography (FDG-PET), glucose metabolic images were obtained from mild DLB and mild AD patients. Two groups consisting of 16 mild DLB patients and 21 mild AD patients were recruited for diagnostic evaluation between mild DLB and mild AD. The mean age+/-SD of the mild DLB group and the mild AD group was 74.3+/-4.9 and 71.7+/-2.1 years, respectively, and the mean scores of the MMSE for the mild DLB and the mild AD group were 21.7+/-1.9 and 23.1+/-2.1, respectively. A receiver operating characteristic (ROC) analysis was performed to compare the diagnostic performance, in terms of discrimination between DLB and AD, of conventional axial FDG-PET images inspected visually by experts and beginners with that of our fully automatic diagnosis system using the statistical brain mapping method and Z scores obtained with the DLB template.

RESULTS

The diagnostic performance of the automatic system was comparable to that of visual inspection by experts. The area under the ROC curve for the automatic diagnosis system was 0.77. The mean area under the ROC curve for visual inspection by experts and beginners was 0.76 and 0.65, respectively.

CONCLUSION

The fully automatic differential diagnosis system for distinction between mild DLB and AD showed a similar diagnostic accuracy to visual inspection by experts. It would be a useful diagnostic tool to distinguish mild DLB from mild AD in clinical practice.