Background: FMRI resting state networks (RSNs) are used to characterize brain disorders. They also show extensive heterogeneity across patients. Identifying systematic differences between RSNs in... Show moreBackground: FMRI resting state networks (RSNs) are used to characterize brain disorders. They also show extensive heterogeneity across patients. Identifying systematic differences between RSNs in patients, i.e. discovering neurofunctional subtypes, may further increase our understanding of disease heterogeneity. Currently, no methodology is available to estimate neurofunctional subtypes and their associated RSNs simultaneously.New method: We present an unsupervised learning method for fMRI data, called Clusterwise Independent Component Analysis (C-ICA). This enables the clustering of patients into neurofunctional subtypes based on differences in shared ICA-derived RSNs. The parameters are estimated simultaneously, which leads to an improved estimation of subtypes and their associated RSNs.Results: In five simulation studies, the C-ICA model is successfully validated using both artificially and realistically simulated data (N = 30-40). The successful performance of the C-ICA model is also illustrated on an empirical data set consisting of Alzheimer's disease patients and elderly control subjects (N = 250). C-ICA is able to uncover a meaningful clustering that partially matches (balanced accuracy = .72) the diagnostic labels and identifies differences in RSNs between the Alzheimer and control cluster. Comparison with other methods: Both in the simulation study and the empirical application, C-ICA yields better results compared to competing clustering methods (i.e., a two step clustering procedure based on single subject ICA's and a Group ICA plus dual regression variant thereof) that do not simultaneously estimate a clustering and associated RSNs. Indeed, the overall mean adjusted Rand Index, a measure for cluster recovery, equals 0.65 for C-ICA and ranges from 0.27 to 0.46 for competing methods.Conclusions: The successful performance of C-ICA indicates that it is a promising method to extract neuro-functional subtypes from multi-subject resting state-fMRI data. This method can be applied on fMRI scans of patient groups to study (neurofunctional) subtypes, which may eventually further increase understanding of disease heterogeneity. Show less
Durieux, J.; Rombouts, S.A.R.B.; Vos, F. de; Koini, M.; Wilderjans, T.F. 2022
Background: FMRI resting state networks (RSNs) are used to characterize brain disorders. They also show extensive heterogeneity across patients. Identifying systematic differences between RSNs in... Show moreBackground: FMRI resting state networks (RSNs) are used to characterize brain disorders. They also show extensive heterogeneity across patients. Identifying systematic differences between RSNs in patients, i.e. discovering neurofunctional subtypes, may further increase our understanding of disease heterogeneity. Currently, no methodology is available to estimate neurofunctional subtypes and their associated RSNs simultaneously. New method: We present an unsupervised learning method for fMRI data, called Clusterwise Independent Component Analysis (C-ICA). This enables the clustering of patients into neurofunctional subtypes based on differences in shared ICA-derived RSNs.The parameters are estimated simultaneously, which leads to an improved estimation of subtypes and their associated RSNs. Results: In five simulation studies, the C-ICA model is successfully validated using both artificially and realistically simulated data (N = 30-40). The successful performance of the C-ICA model is also illustrated on an empirical data set consisting of Alzheimer's disease patients and elderly control subjects (N = 250). C-ICA is able to uncover a meaningful clustering that partially matches (balanced accuracy = .72) the diagnostic labels and identifies differences in RSNs between the Alzheimer and control cluster. Comparison with other methods: Both in the simulation study and the empirical application, C-ICA yields better results compared to competing clustering methods (i.e., a two step clustering procedure based on single subject ICA's and a Group ICA plus dual regression variant thereof) that do not simultaneously estimate a clustering and associated RSNs. Indeed, the overall mean adjusted Rand Index, a measure for cluster recovery, equals 0.65 for C-ICA and ranges from 0.27 to 0.46 for competing methods. Conclusions: The successful performance of C-ICA indicates that it is a promising method to extract neuro-functional subtypes from multi-subject resting state-fMRI data. This method can be applied on fMRI scans of patient groups to study (neurofunctional) subtypes, which may eventually further increase understanding of disease heterogeneity. Show less
Plomp, J.J.; Huijbers, M.G.M.; Verschuuren, J.J.G.M.; Borodovsky, A. 2022
Background: Myasthenia gravis (MG) is an autoimmune neuromuscular disorder hallmarked by fluctuating fatigable muscle weakness. Most patients have autoantibodies against acetylcholine receptors ... Show moreBackground: Myasthenia gravis (MG) is an autoimmune neuromuscular disorder hallmarked by fluctuating fatigable muscle weakness. Most patients have autoantibodies against acetylcholine receptors (AChRs) at the neuromuscular junction (NMJ). These are thought to have three possible pathogenic mode-of-actions: 1) cross linking and endocytosis of AChRs, 2) direct block of AChRs and 3) complement activation. The relative contributions of these mechanisms to synaptic block and muscle weakness of individual patients cannot be determined. It likely varies between patients and perhaps also with disease course, depending on the nature of the circulating AChR antibodies.New method: We developed a new bioassay which specifically enables functional characterization and quantification of complement-mediated synaptic damage at NMJs, without interference of the other pathogenic mechanisms. To this end, we pre-incubated mouse hemi-diaphragm muscle-nerve preparations with mAb35-hG1, a humanized rat AChR monoclonal and subsequently exposed the preparation to normal human serum as a complement source. NMJ-restricted effects were studied.Results: Clearly NMJ-restricted damage occurred. With immunohistology we showed complement deposition at NMJs, and synaptic electrophysiological measurements demonstrated transmission block. In whole-muscle contraction experiments we quantified the effect and characterized its onset and progression during the incubation with normal human serum. Comparison with existing methods: With this new assay the complement-mediated component of myasthenic NMJ pathology can be studied separately. Conclusions: Our assay will be of importance in detailed mechanistic studies of local complement activation at NMJs, investigations of new complement inhibitors, and laboratory pre-screening of therapeutic efficacy for individual MG patients to optimize care with clinically approved complement inhibitors. Show less
Background: The combination of EEG and ultra-high-field (7T and above) fMRI holds the promise to relate electrophysiology and hemodynamics with greater signal to noise level and at higher spatial... Show moreBackground: The combination of EEG and ultra-high-field (7T and above) fMRI holds the promise to relate electrophysiology and hemodynamics with greater signal to noise level and at higher spatial resolutions than conventional field strengths. Technical and safety restrictions have so far resulted in compromises in terms of MRI coil selection, resulting in reduced, signal quality, spatial coverage and resolution in EEG-fMRI studies at 7 T.New method: We adapted a 64-channel MRI-compatible EEG cap so that it could be used with a closed 32-channel MRI head coil thus avoiding several of these compromises. We compare functional and anatomical as well as the EEG quality recorded with this adapted setup with those recorded with a setup that uses an open-ended 8-channel head-coil.Results: Our set-up with the adapted EEG cap inside the closed 32 channel coil resulted in the recording of good quality EEG and (f)MRI data. Both functional and anatomical MRI images show no major effects of the adapted EEG cap on MR signal quality. We demonstrate the ability to compute ERPs and changes in alpha and gamma oscillations from the recorded EEG data.Comparison with existing methods: Compared to MRI recordings with an 8-channel open-ended head-coil, the loss in signal quality of the MRI images related to the adapted EEG cap is considerably reduced.Conclusions: The adaptation of the EEG cap permits the simultaneous recording of good quality whole brain (f) MRI data using a 32 channel receiver coil, while maintaining the quality of the EEG data. Show less
The shape, structure and connectivity of nerve cells are important aspects of neuronal function. Genetic and epigenetic factors that alter neuronal morphology or synaptic localization of pre- and... Show moreThe shape, structure and connectivity of nerve cells are important aspects of neuronal function. Genetic and epigenetic factors that alter neuronal morphology or synaptic localization of pre- and post-synaptic proteins contribute significantly to neuronal output and may underlie clinical states. To assess the impact of individual genes and disease-causing mutations on neuronal morphology, reliable methods are needed. Unfortunately, manual analysis of immuno-fluorescence images of neurons to quantify neuronal shape and synapse number, size and distribution is labor-intensive, time-consuming and subject to human bias and error. We have developed an automated image analysis routine using steerable filters and deconvolutions to automatically analyze dendrite and synapse characteristics in immuno-fluorescence images. Our approach reports dendrite morphology, synapse size and number but also synaptic vesicle density and synaptic accumulation of proteins as a function of distance from the soma as consistent as expert observers while reducing analysis time considerably. In addition, the routine can be used to detect and quantify a wide range of neuronal organelles and is capable of batch analysis of a large number of images enabling high-throughput analysis. (C) 2010 Elsevier B.V. All rights reserved. Show less
