Dataset Information

Optimising network modelling methods for fMRI.

ABSTRACT: A major goal of neuroimaging studies is to develop predictive models to analyze the relationship between whole brain functional connectivity patterns and behavioural traits. However, there is no single widely-accepted standard pipeline for analyzing functional connectivity. The common procedure for designing functional connectivity based predictive models entails three main steps: parcellating the brain, estimating the interaction between defined parcels, and lastly, using these integrated associations between brain parcels as features fed to a classifier for predicting non-imaging variables e.g., behavioural traits, demographics, emotional measures, etc. There are also additional considerations when using correlation-based measures of functional connectivity, resulting in three supplementary steps: utilising Riemannian geometry tangent space parameterization to preserve the geometry of functional connectivity; penalizing the connectivity estimates with shrinkage approaches to handle challenges related to short time-series (and noisy) data; and removing confounding variables from brain-behaviour data. These six steps are contingent on each-other, and to optimise a general framework one should ideally examine these various methods simultaneously. In this paper, we investigated strengths and short-comings, both independently and jointly, of the following measures: parcellation techniques of four kinds (categorized further depending upon number of parcels), five measures of functional connectivity, the decision of staying in the ambient space of connectivity matrices or in tangent space, the choice of applying shrinkage estimators, six alternative techniques for handling confounds and finally four novel classifiers/predictors. For performance evaluation, we have selected two of the largest datasets, UK Biobank and the Human Connectome Project resting state fMRI data, and have run more than 9000 different pipeline variants on a total of ∼14000 individuals to determine the optimum pipeline. For independent performance validation, we have run some best-performing pipeline variants on ABIDE and ACPI datasets (∼1000 subjects) to evaluate the generalisability of proposed network modelling methods.

SUBMITTER: Pervaiz U

PROVIDER: S-EPMC7086233 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundEarly surgical intervention in children with drug resistant epilepsy has benefits but requires using tolerable and minimally invasive tests. EEG-fMRI studies have demonstrated good sensitivity for the localization of epileptic focus but a poor yield although the reasons for this have not been systematically addressed. While adults EEG-fMRI studies are performed in the "resting state"; children are commonly sedated however, this has associated risks and potential confounds. In this study, we assessed the impact of the following factors on the tolerability and results of EEG-fMRI in children: viewing a movie inside the scanner; movement; occurrence of interictal epileptiform discharges (IED); scan duration and design efficiency. This work's motivation is to optimize EEG-fMRI parameters to make this test widely available to paediatric population.MethodsForty-six children with focal epilepsy and 20 controls (6-18) underwent EEG-fMRI. For two 10 minutes sessions subjects were told to lie still with eyes closed, as it is classically performed in adult studies ("rest sessions"), for another two sessions, subjects watched a child friendly stimulation i.e. movie ("movie sessions"). IED were mapped with EEG-fMRI for each session and across sessions. The resulting maps were classified as concordant/discordant with the presumed epileptogenic focus for each subject.FindingsMovement increased with scan duration, but the movie reduced movement by ~40% when played within the first 20 minutes. There was no effect of movie on the occurrence of IED, nor in the concordance of the test. Ability of EEG-fMRI to map the epileptogenic region was similar for the 20 and 40 minute scan durations. Design efficiency was predictive of concordance.ConclusionsA child friendly natural stimulus improves the tolerability of EEG-fMRI and reduces in-scanner movement without having an effect on IED occurrence and quality of EEG-fMRI maps. This allowed us to scan children as young as 6 and obtain localising information without sedation. Our data suggest that ~20 minutes is the optimal length of scanning for EEG-fMRI studies in children with frequent IED. The efficiency of the fMRI design derived from spontaneous IED generation is an important factor for producing concordant results.

Project description:BackgroundStatistical models are regularly used in the forecasting and surveillance of infectious diseases to guide public health. Variable selection assists in determining factors associated with disease transmission, however, often overlooked in this process is the evaluation and suitability of the statistical model used in forecasting disease transmission and outbreaks. Here we aim to evaluate several modelling methods to optimise predictive modelling of Ross River virus (RRV) disease notifications and outbreaks in epidemiological important regions of Victoria and Western Australia.Methodology/principal findingsWe developed several statistical methods using meteorological and RRV surveillance data from July 2000 until June 2018 in Victoria and from July 1991 until June 2018 in Western Australia. Models were developed for 11 Local Government Areas (LGAs) in Victoria and seven LGAs in Western Australia. We found generalised additive models and generalised boosted regression models, and generalised additive models and negative binomial models to be the best fit models when predicting RRV outbreaks and notifications, respectively. No association was found with a model's ability to predict RRV notifications in LGAs with greater RRV activity, or for outbreak predictions to have a higher accuracy in LGAs with greater RRV notifications. Moreover, we assessed the use of factor analysis to generate independent variables used in predictive modelling. In the majority of LGAs, this method did not result in better model predictive performance.Conclusions/significanceWe demonstrate that models which are developed and used for predicting disease notifications may not be suitable for predicting disease outbreaks, or vice versa. Furthermore, poor predictive performance in modelling disease transmissions may be the result of inappropriate model selection methods. Our findings provide approaches and methods to facilitate the selection of the best fit statistical model for predicting mosquito-borne disease notifications and outbreaks used for disease surveillance.

Project description:IntroductionTo reduce the burden of colorectal cancer (CRC) in Shanghai, China, a CRC screening programme was commenced in 2013 inviting those aged 50-74 years to triennial screening with a faecal immunochemical test (FIT) and risk assessment. However, it is unknown whether this is the optimal screening strategy for this population. We aimed to determine the optimal CRC screening programme for Shanghai in terms of benefits, burden, harms and cost-effectiveness.MethodsUsing Microsimulation Screening Analysis-Colon (MISCAN-Colon), we estimated the costs and effects of the current screening programme compared with a situation without screening. Subsequently, we estimated the benefits (life years gained (LYG)), burden (number of screening events, colonoscopies and false-positive tests), harms (number of colonoscopy complications) and costs (Renminb (¥)) of screening for 324 alternative screening strategies. We compared several different age ranges, screening modalities, intervals and FIT cut-off levels. An incremental cost-effectiveness analysis determined the optimal strategy assuming a willingness-to-pay of ¥193 931 per LYG.ResultsCompared with no screening, the current screening programme reduced CRC incidence by 40% (19 cases per 1000 screened individuals) and CRC mortality by 67% (7 deaths). This strategy gained 32 additional life years, increased colonoscopy demand to 1434 per 1000 individuals and cost an additional ¥199 652. The optimal screening strategy was annual testing using a validated one-sample FIT, with a cut-off of 10 µg haemoglobin per gram from ages 45 to 80 years (incremental cost-effectiveness ratio, ¥62 107). This strategy increased LY by 0.18% and costs by 27%. Several alternative cost-effective strategies using a validated FIT offered comparable benefits to the current programme but lower burden and costs.ConclusionsAlthough the current screening programme in Shanghai is effective at reducing CRC incidence and mortality, the programme could be optimised using a validated FIT. When implementing CRC screening, jurisdictions with limited health resources should use a validated test.

Dataset Information

Optimising network modelling methods for fMRI.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets