Alzheimer Disease: Frackowiak RS

Good CD, Scahill RI, Fox NC, Ashburner J, Friston KJ, Chan D, Crum WR, Rossor MN, Frackowiak RS. · Wellcome Department of Cognitive Neurology, Institute of Neurology, University College London, 12 Queen Square, London WC1N 3BG, United Kingdom. · Neuroimage. · Pubmed #12482066 No free full text.

Abstract: We compared voxel-based morphometry (VBM) with independent accurate region-of-interest (ROI) measurements of temporal lobe structures in order to validate the usefulness of this fully automated and unbiased technique in Alzheimer's disease (AD) and semantic dementia (SD). In AD, ROI analyses appear more sensitive to volume loss in the amygdalae, whereas VBM analyses appear more sensitive to right middle temporal gyrus and regional hippocampal volume loss. In SD, ROI analyses appear more sensitive to left middle and inferior temporal gyrus volume loss, whereas VBM appears more sensitive to regional hippocampal volume loss. In addition the significance of volume reductions was generally less in VBM owing to more stringent corrections for multiple comparisons. In conclusion, the automated technique detects a general trend of atrophy similar to that of expertly labeled ROI measurements in AD and SD, although there are discrepancies in the ranking of severity and in the significance of volume reductions that are more marked in AD.

Klöppel S, Stonnington CM, Barnes J, Chen F, Chu C, Good CD, Mader I, Mitchell LA, Patel AC, Roberts CC, Fox NC, Jack CR, Ashburner J, Frackowiak RS. · Department of Psychiatry and Psychotherapy and Freiburg Brain Imaging, University Clinic Freiburg, Freiburg, Germany. · Brain. · Pubmed #18835868 links to  free full text

Abstract: There has been recent interest in the application of machine learning techniques to neuroimaging-based diagnosis. These methods promise fully automated, standard PC-based clinical decisions, unbiased by variable radiological expertise. We recently used support vector machines (SVMs) to separate sporadic Alzheimer's disease from normal ageing and from fronto-temporal lobar degeneration (FTLD). In this study, we compare the results to those obtained by radiologists. A binary diagnostic classification was made by six radiologists with different levels of experience on the same scans and information that had been previously analysed with SVM. SVMs correctly classified 95% (sensitivity/specificity: 95/95) of sporadic Alzheimer's disease and controls into their respective groups. Radiologists correctly classified 65-95% (median 89%; sensitivity/specificity: 88/90) of scans. SVM correctly classified another set of sporadic Alzheimer's disease in 93% (sensitivity/specificity: 100/86) of cases, whereas radiologists ranged between 80% and 90% (median 83%; sensitivity/specificity: 80/85). SVMs were better at separating patients with sporadic Alzheimer's disease from those with FTLD (SVM 89%; sensitivity/specificity: 83/95; compared to radiological range from 63% to 83%; median 71%; sensitivity/specificity: 64/76). Radiologists were always accurate when they reported a high degree of diagnostic confidence. The results show that well-trained neuroradiologists classify typical Alzheimer's disease-associated scans comparable to SVMs. However, SVMs require no expert knowledge and trained SVMs can readily be exchanged between centres for use in diagnostic classification. These results are encouraging and indicate a role for computerized diagnostic methods in clinical practice.

Klöppel S, Stonnington CM, Chu C, Draganski B, Scahill RI, Rohrer JD, Fox NC, Jack CR, Ashburner J, Frackowiak RS. · Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London, London, UK. · Brain. · Pubmed #18202106 links to  free full text

Abstract: To be diagnostically useful, structural MRI must reliably distinguish Alzheimer's disease (AD) from normal aging in individual scans. Recent advances in statistical learning theory have led to the application of support vector machines to MRI for detection of a variety of disease states. The aims of this study were to assess how successfully support vector machines assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centres could be used to obtain effective classification of scans. We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centres with different scanning equipment. Because the clinical diagnosis of mild AD is difficult we also tested the ability of support vector machines to differentiate control scans from patients without post-mortem confirmation. Finally we sought to use these methods to differentiate scans between patients suffering from AD from those with frontotemporal lobar degeneration. Up to 96% of pathologically verified AD patients were correctly classified using whole brain images. Data from different centres were successfully combined achieving comparable results from the separate analyses. Importantly, data from one centre could be used to train a support vector machine to accurately differentiate AD and normal ageing scans obtained from another centre with different subjects and different scanner equipment. Patients with mild, clinically probable AD and age/sex matched controls were correctly separated in 89% of cases which is compatible with published diagnosis rates in the best clinical centres. This method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or frontotemporal lobar degeneration to their respective group. Our study leads to three conclusions: Firstly, support vector machines successfully separate patients with AD from healthy aging subjects. Secondly, they perform well in the differential diagnosis of two different forms of dementia. Thirdly, the method is robust and can be generalized across different centres. This suggests an important role for computer based diagnostic image analysis for clinical practice.

Stonnington CM, Tan G, Klöppel S, Chu C, Draganski B, Jack CR, Chen K, Ashburner J, Frackowiak RS. · Wellcome Trust Centre for Neuroimaging, Institute of Neurology, UCL, London, UK. · Neuroimage. · Pubmed #18032068 links to  free full text

Abstract: Large, multi-site studies utilizing MRI-derived measures from multiple scanners present an opportunity to advance research by pooling data. On the other hand, it remains unclear whether or not the potential confound introduced by different scanners and upgrades will devalue the integrity of any results. Although there are studies of scanner differences for the purpose of calibration and quality control, the current literature is devoid of studies that describe the analysis of multi-scanner data with regard to the interaction of scanner(s) with effects of interest. We investigated a data-set of 136 subjects, 62 patients with mild to moderate Alzheimer's disease and 74 cognitively normal elderly controls, with MRI scans from one center that were acquired over 10 years with 6 different scanners and multiple upgrades over time. We used a whole-brain voxel-wise analysis to evaluate the effect of scanner, effect of disease, and the interaction of scanner and disease for the 6 different scanners. The effect of disease in patients showed the expected significant reduction of grey matter in the medial temporal lobe. Scanner differences were substantially less than the group differences and only significant in the thalamus. There was no significant interaction of scanner with disease group. We describe the rationale for concluding that our results were not confounded by scanner differences. Similar analyses in other multi-scanner data-sets could be used to justify the pooling of data when needed, such as in studies of rare disorders or in multi-center designs.

Pariente J, Cole S, Henson R, Clare L, Kennedy A, Rossor M, Cipoloti L, Puel M, Demonet JF, Chollet F, Frackowiak RS. · Wellcome Department of Imaging Neuroscience, London, United Kingdom. · Ann Neurol. · Pubmed #16315273 No free full text.

Abstract: We conducted an event-related functional magnetic resonance imaging experiment to better understand the potentially compensatory alternative brain networks activated by a clinically relevant face-name association task in Alzheimer's disease (AD) patients and matched control subjects. We recruited 17 healthy subjects and 12 AD patients at an early stage of the disease. They underwent functional magnetic resonance imaging scanning in four sessions. Each of the sessions combined a "study" phase and a "test" phase. Face/name pairs were presented in each study phase, and subjects were asked to associate faces with names. In the test phase, a recognition task, faces seen in the study phase were presented each with four different names. The task required selection of appropriate previously associated names from the study phase. Responses were recorded for post hoc classification into those successfully or unsuccessfully encoded. There were significant differences between the groups in accuracy and reaction time. Comparison of correctly versus incorrectly encoded and recognized pairs in the two groups indicated bilateral hippocampal hypoactivation both when encoding and recognizing in the AD group. Moreover, patients showed bilateral hyperactivation of parts of the parietal and frontal lobes. We discuss whether hyperactivation of a frontoparietal network reflects compensatory strategies for failing associative memory in AD patients.

Signorini M, Paulesu E, Friston K, Perani D, Colleluori A, Lucignani G, Grassi F, Bettinardi V, Frackowiak RS, Fazio F. · Istituto Scientifico H San Raffaele, Universita' di Milano via Olgeltina 60, 20132, Milano, Italia. · Neuroimage. · Pubmed #9918728 No free full text.

Abstract: The [18F]fluorodeoxyglucose ([18F]FDG) method for measuring brain metabolism has not the wide clinical application that one might expect, partly because of its high cost and the complexity of the quantification procedure, but also because of reporting techniques based on region of interest (ROI) analysis, which are time-consuming and not fully objective. In this paper we report a clinical validation of statistical parametric mapping (SPM) using rCMRglc (quantitative) and radioactivity distribution (nonquantitative) [18F]FDG PET data. We show that a 10-min noninteractive voxel-based SPM analysis on a standard workstation enables objective assessment, including localization in stereotactic space, of regional glucose consumption abnormalities, whose reliability can be assessed on statistical and clinical grounds. Clinical validity was established using a small series of patients with degenerative or developmental disorders, including probable Alzheimer's disease, progressive aphasia, multiple sclerosis, developmental specific language impairment, and epilepsy. Analysis of quantitative and nonquantitative data showed the same pattern of results, suggesting that, for clinical purposes, quantitation and invasive arterial cannulation can be avoided. This should facilitate a wider application of the technique and the extension of SPM clinical analysis to H215O PET or high resolution SPECT perfusion studies.