Enhancing Multi-Center Generalization of Machine Learning-Based Depression Diagnosis From Resting-State fMRI.
Ontology highlight
ABSTRACT:
SUBMITTER: Nakano T
PROVIDER: S-EPMC7270328 | biostudies-literature | 2020
REPOSITORIES: biostudies-literature
Publications
Enhancing Multi-Center Generalization of Machine Learning-Based Depression Diagnosis From Resting-State fMRI.
Nakano Takashi T Takamura Masahiro M Ichikawa Naho N Okada Go G Okamoto Yasumasa Y Yamada Makiko M Suhara Tetsuya T Yamawaki Shigeto S Yoshimoto Junichiro J
Frontiers in psychiatry 20200528
Resting-state fMRI has the potential to help doctors detect abnormal behavior in brain activity and to diagnose patients with depression. However, resting-state fMRI has a bias depending on the scanner site, which makes it difficult to diagnose depression at a new site. In this paper, we propose methods to improve the performance of the diagnosis of major depressive disorder (MDD) at an independent site by reducing the site bias effects using regression. For this, we used a subgroup of healthy s ...[more]
Similar Datasets
Project description:Resting-state fMRI (R-fMRI) has shown considerable promise in providing potential biomarkers for diagnosis, prognosis and drug response across a range of diseases. Incorporating R-fMRI into multi-center studies is becoming increasingly popular, imposing technical challenges on data acquisition and analysis, as fMRI data is particularly sensitive to structured noise resulting from hardware, software, and environmental differences. Here, we investigated whether a novel clean up tool for structured noise was capable of reducing center-related R-fMRI differences between healthy subjects. We analyzed three Tesla R-fMRI data from 72 subjects, half of whom were scanned with eyes closed in a Philips Achieva system in The Netherlands, and half of whom were scanned with eyes open in a Siemens Trio system in the UK. After pre-statistical processing and individual Independent Component Analysis (ICA), FMRIB's ICA-based X-noiseifier (FIX) was used to remove noise components from the data. GICA and dual regression were run and non-parametric statistics were used to compare spatial maps between groups before and after applying FIX. Large significant differences were found in all resting-state networks between study sites before using FIX, most of which were reduced to non-significant after applying FIX. The between-center difference in the medial/primary visual network, presumably reflecting a between-center difference in protocol, remained statistically significant. FIX helps facilitate multi-center R-fMRI research by diminishing structured noise from R-fMRI data. In doing so, it improves combination of existing data from different centers in new settings and comparison of rare diseases and risk genes for which adequate sample size remains a challenge.
| S-EPMC4621866 | biostudies-literature
Project description:Multi-center MRI studies can enhance power, generalizability, and discovery for clinical neuroimaging research in brain disorders. Here, we sought to establish the utility of a clustering algorithm as an alternative to more traditional intra-class correlation coefficient approaches in a longitudinal multi-center human phantom study. We completed annual reliability scans on 'travelling human phantoms'. Acquisitions across sites were harmonized prospectively. Twenty-seven MRI sessions were available across four participants, scanned on five scanners, across three years. For each scan, three metrics were extracted: cortical thickness (CT), white matter fractional anisotropy (FA), and resting state functional connectivity (FC). For each metric, hierarchical clustering (Ward's method) was performed. The cluster solutions were compared to participant and scanner using the adjusted Rand index (ARI). For all metrics, data clustered by participant rather than by scanner (ARI > 0.8 comparing clusters to participants, ARI < 0.2 comparing clusters to scanners). These results demonstrate that hierarchical clustering can reliably identify structural and functional scans from different participants imaged on different scanners across time. With increasing interest in data-driven approaches in psychiatric and neurologic brain imaging studies, our findings provide a framework for multi-center analytic approaches aiming to identify subgroups of participants based on brain structure or function.
| S-EPMC6482446 | biostudies-literature
Project description:Machine learning classifiers have become increasingly popular tools to generate single-subject inferences from fMRI data. With this transition from the traditional group level difference investigations to single-subject inference, the application of machine learning methods can be seen as a considerable step forward. Existing studies, however, have given scarce or no information on the generalizability to other subject samples, limiting the use of such published classifiers in other research projects. We conducted a simulation study using publicly available resting-state fMRI data from the 1000 Functional Connectomes and COBRE projects to examine the generalizability of classifiers based on regional homogeneity of resting-state time series. While classification accuracies of up to 0.8 (using sex as the target variable) could be achieved on test datasets drawn from the same study as the training dataset, the generalizability of classifiers to different study samples proved to be limited albeit above chance. This shows that on the one hand a certain amount of generalizability can robustly be expected, but on the other hand this generalizability should not be overestimated. Indeed, this study substantiates the need to include data from several sites in a study investigating machine learning classifiers with the aim of generalizability.
| S-EPMC4114329 | biostudies-literature
Project description:Disrupted striatal functional connectivity is proposed to play a critical role in the development of psychotic symptoms. Previous resting-state functional magnetic resonance imaging (rs-fMRI) studies typically reported disrupted striatal connectivity in patients with psychosis and in individuals at clinical and genetic high risk of the disorder relative to healthy controls. This has not been widely studied in healthy individuals with subclinical psychotic-like experiences (schizotypy). Here we applied the emerging technology of multi-echo rs-fMRI to examine corticostriatal connectivity in this group, which is thought to drastically maximize physiological noise removal and increase BOLD contrast-to-noise ratio. Multi-echo rs-fMRI data (echo times, 12, 28, 44, 60 ms) were acquired from healthy individuals with low (LS, n = 20) and high (HS, n = 19) positive schizotypy as determined with the Oxford-Liverpool Inventory of Feelings and Experiences (O-LIFE). After preprocessing to ensure optimal contrast and removal of non-BOLD signal components, whole-brain functional connectivity from six striatal seeds was compared between the HS and LS groups. Effects were considered significant at cluster-level p < .05 family-wise error correction. Compared to LS, HS subjects showed lower rs-fMRI connectivity between ventromedial prefrontal regions and ventral striatal regions. Lower connectivity was also observed between the dorsal putamen and the hippocampus, occipital regions, as well as the cerebellum. These results demonstrate that subclinical positive psychotic-like experiences in healthy individuals are associated with striatal hypoconnectivity as detected using multi-echo rs-fMRI. Further application of this approach may aid in characterizing functional connectivity abnormalities across the extended psychosis phenotype.
| S-EPMC6413302 | biostudies-literature
Project description:We present a technique for predicting cardiac and respiratory phase on a time point by time point basis, from fMRI image data. These predictions have utility in attempts to detrend effects of the physiological cycles from fMRI image data. We demonstrate the technique both in the case where it can be trained on a subject's own data, and when it cannot. The prediction scheme uses a multiclass support vector machine algorithm. Predictions are demonstrated to have a close fit to recorded physiological phase, with median Pearson correlation scores between recorded and predicted values of 0.99 for the best case scenario (cardiac cycle trained on a subject's own data) down to 0.83 for the worst case scenario (respiratory predictions trained on group data), as compared to random chance correlation score of 0.70. When predictions were used with RETROICOR--a popular physiological noise removal tool--the effects are compared to using recorded phase values. Using Fourier transforms and seed based correlation analysis, RETROICOR is shown to produce similar effects whether recorded physiological phase values are used, or they are predicted using this technique. This was seen by similar levels of noise reduction noise in the same regions of the Fourier spectra, and changes in seed based correlation scores in similar regions of the brain. This technique has a use in situations where data from direct monitoring of the cardiac and respiratory cycles are incomplete or absent, but researchers still wish to reduce this source of noise in the image data.
| S-EPMC6870181 | biostudies-literature
Project description:BackgroundSome studies suggest that as much as 40% of all causes of death in a group of patients with schizophrenia can be attributed to suicides and compared with the general population, patients with schizophrenia have an 8.5-fold greater suicide risk (SR). There is a vital need for accurate and reliable methods to predict the SR among patients with schizophrenia based on biological measures. However, it is unknown whether the suicidal risk in schizophrenia can be related to alterations in spontaneous brain activity, or if the resting-state functional magnetic resonance imaging (rsfMRI) measures can be used alongside machine learning (ML) algorithms in order to identify patients with SR.MethodsFifty-nine participants including patients with schizophrenia with and without SR as well as age and gender-matched healthy underwent 13 min resting-state functional magnetic resonance imaging. Both static and dynamic indexes of the amplitude of low-frequency fluctuation (ALFF), the fractional amplitude of low-frequency fluctuations (fALFF), regional homogeneity as well as functional connectivity (FC) were calculated and used as an input for five machine learning algorithms: Gradient boosting (GB), LASSO, Logistic Regression (LR), Random Forest and Support Vector Machine.ResultsAll groups revealed different intra-network functional connectivity in ventral DMN and anterior SN. The best performance was reached for the LASSO applied to FC with an accuracy of 70% and AUROC of 0.76 (p < 0.05). Significant classification ability was also reached for GB and LR using fALFF and ALFF measures.ConclusionOur findings suggest that SR in schizophrenia can be seen on the level of DMN and SN functional connectivity alterations. ML algorithms were able to significantly differentiate SR patients. Our results could be useful in developing neuromarkers of SR in schizophrenia based on non-invasive rsfMRI.
| S-EPMC7829970 | biostudies-literature
Project description:BACKGROUND:Previous studies have demonstrated that patients with treatment-resistant depression (TRD) and treatment-sensitive depression (TSD) differed at neural level. However, it remains unclear if these two subtypes of depression differ in the interhemispheric coordination. This study was undertaken for two purposes: (1) to explore the differences in interhemispheric coordination between these two subtypes by using the voxel-mirrored homotopic connectivity (VMHC) method; and (2) to determine if the difference of interhemispheric coordination can be used as a biomarker(s) to differentiate TRD from both TSD and healthy subjects (HS). METHODS:Twenty-three patients with TRD, 22 with TSD, and 19 HS participated in the study. Data of these participants were analyzed with the VMHC and seed-based functional connectivity (FC) approaches. RESULTS:Compared to the TSD group, the TRD group showed significantly lower VMHC values in the calcarine cortex, fusiform gyrus, hippocampus, superior temporal gyrus, middle cingulum, and precentral gyrus. Lower VMHC values were also observed in the TRD group in the calcarine cortex relative to the HS group. However, the TSD group had no significant change in VMHC value in any brain region compared to the HS group. Receiver operating characteristic curves (ROC) analysis revealed that the VMHC values in the calcarine cortex had discriminatory function distinguishing patients with TRD from patients with TSD as well as those participants in the HS group. CONCLUSIONS:Lower VMHC values of patients with TRD relative to those with TSD and those in the HS group in the calcarine cortex appeared to be a unique feature for patients with TRD and it may be used as an imaging biomarker to separate patients with TRD from those with TSD or HS.
| S-EPMC3732240 | biostudies-literature
Project description:Resting-state fMRI has been widely used in investigating the pathophysiology of late-life depression (LLD). Unlike the conventional linear approach, cross-sample entropy (CSE) analysis shows the nonlinear property in fMRI signals between brain regions. Moreover, recent advances in deep learning, such as convolutional neural networks (CNNs), provide a timely application for understanding LLD. Accurate and prompt diagnosis is essential in LLD; hence, this study aimed to combine CNN and CSE analysis to discriminate LLD patients and non-depressed comparison older adults based on brain resting-state fMRI signals. Seventy-seven older adults, including 49 patients and 28 comparison older adults, were included for fMRI scans. Three-dimensional CSEs with volumes corresponding to 90 seed regions of interest of each participant were developed and fed into models for disease classification and depression severity prediction. We obtained a diagnostic accuracy > 85% in the superior frontal gyrus (left dorsolateral and right orbital parts), left insula, and right middle occipital gyrus. With a mean root-mean-square error (RMSE) of 2.41, three separate models were required to predict depressive symptoms in the severe, moderate, and mild depression groups. The CSE volumes in the left inferior parietal lobule, left parahippocampal gyrus, and left postcentral gyrus performed best in each respective model. Combined complexity analysis and deep learning algorithms can classify patients with LLD from comparison older adults and predict symptom severity based on fMRI data. Such application can be utilized in precision medicine for disease detection and symptom monitoring in LLD.
| S-EPMC9922223 | biostudies-literature
Project description:Resting-state functional MRI (rs-fMRI) is widely used to noninvasively study human brain networks. Network functional connectivity is often estimated by calculating the timeseries correlation between blood-oxygen-level dependent (BOLD) signal from different regions of interest (ROIs). However, standard correlation cannot characterize the direction of information flow between regions. In this paper, we introduce and test a new concept, prediction correlation, to estimate effective connectivity in functional brain networks from rs-fMRI. In this approach, the correlation between two BOLD signals is replaced by a correlation between one BOLD signal and a prediction of this signal via a causal system driven by another BOLD signal. Three validations are described: (1) Prediction correlation performed well on simulated data where the ground truth was known, and outperformed four other methods. (2) On simulated data designed to display the "common driver" problem, prediction correlation did not introduce false connections between non-interacting driven ROIs. (3) On experimental data, prediction correlation recovered the previously identified network organization of human brain. Prediction correlation scales well to work with hundreds of ROIs, enabling it to assess whole brain interregional connectivity at the single subject level. These results provide an initial validation that prediction correlation can capture the direction of information flow and estimate the duration of extended temporal delays in information flow between regions of interest ROIs based on BOLD signal. This approach not only maintains the high sensitivity to network connectivity provided by the correlation analysis, but also performs well in the estimation of causal information flow in the brain.
| S-EPMC5433247 | biostudies-literature
Project description:Early diagnosis is critical for individuals with Alzheimer's disease (AD) in clinical practice because its progress is irreversible. In the existing literature, support vector machine (SVM) has always been applied to distinguish between AD and healthy controls (HC) based on neuroimaging data. But previous studies have only used a single SVM to classify AD and HC, and the accuracy is not very high and generally less than 90%. The method of random support vector machine cluster was proposed to classify AD and HC in this paper. From the Alzheimer's Disease Neuroimaging Initiative database, the subjects including 25 AD individuals and 35 HC individuals were obtained. The classification accuracy could reach to 94.44% in the results. Furthermore, the method could also be used for feature selection and the accuracy could be maintained at the level of 94.44%. In addition, we could also find out abnormal brain regions (inferior frontal gyrus, superior frontal gyrus, precentral gyrus and cingulate cortex). It is worth noting that the proposed random support vector machine cluster could be a new insight to help the diagnosis of AD.
| S-EPMC5865739 | biostudies-literature