Dataset Information

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes.

ABSTRACT: There is an increasing expectation that advanced, computationally expensive machine learning (ML) techniques, when applied to large population-wide neuroimaging datasets, will help to uncover key differences in the human brain in health and disease. We take a comprehensive approach to explore how multiple aspects of brain structural connectivity can predict sex, age, general cognitive function and general psychopathology, testing different ML algorithms from deep learning (DL) model (BrainNetCNN) to classical ML methods. We modelled N = 8183 structural connectomes from UK Biobank using six different structural network weightings obtained from diffusion MRI. Streamline count generally provided the highest prediction accuracies in all prediction tasks. DL did not improve on prediction accuracies from simpler linear models. Further, high correlations between gradient attribution coefficients from DL and model coefficients from linear models suggested the models ranked the importance of features in similar ways, which indirectly suggested the similarity in models' strategies for making predictive decision to some extent. This highlights that model complexity is unlikely to improve detection of associations between structural connectomes and complex phenotypes with the current sample size.

SUBMITTER: Yeung HW

PROVIDER: S-EPMC9980898 | biostudies-literature | 2023 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes.

Yeung Hon Wah HW Stolicyn Aleks A Buchanan Colin R CR Tucker-Drob Elliot M EM Bastin Mark E ME Luz Saturnino S McIntosh Andrew M AM Whalley Heather C HC Cox Simon R SR Smith Keith K

Human brain mapping 20221221 5

There is an increasing expectation that advanced, computationally expensive machine learning (ML) techniques, when applied to large population-wide neuroimaging datasets, will help to uncover key differences in the human brain in health and disease. We take a comprehensive approach to explore how multiple aspects of brain structural connectivity can predict sex, age, general cognitive function and general psychopathology, testing different ML algorithms from deep learning (DL) model (BrainNetCNN ...[more]

PMID: 36541441

Similar Datasets

Project description:BackgroundSelf-perceived general health (SPGH) is a general health indicator commonly used in epidemiological research and is associated with a wide range of exposures from different domains. However, most studies on SPGH only investigated a limited set of exposures and did not take the entire external exposome into account. We aimed to develop predictive models for SPGH based on exposome datasets using machine learning techniques and identify the most important predictors of poor SPGH status.MethodsRandom forest (RF) was used on two datasets based on personal characteristics from the 2012 and 2016 editions of the Dutch national health survey, enriched with environmental and neighborhood characteristics. Model performance was determined using the area under the curve (AUC) score. The most important predictors were identified using a variable importance procedure and individual effects of exposures using partial dependence and accumulated local effect plots. The final 2012 dataset contained information on 199,840 individuals and 81 variables, whereas the final 2016 dataset had 244,557 individuals with 91 variables.ResultsOur RF models had overall good predictive performance (2012: AUC = 0.864 (CI: 0.852-0.876); 2016: AUC = 0.890 (CI: 0.883-0.896)) and the most important predictors were "Control of own life", "Physical activity", "Loneliness" and "Making ends meet". Subjects who felt insufficiently in control of their own life, scored high on the De Jong-Gierveld loneliness scale or had difficulty in making ends meet were more likely to have poor SPGH status, whereas increased physical activity per week reduced the probability of poor SPGH. We observed associations between some neighborhood and environmental characteristics, but these variables did not contribute to the overall predictive strength of the models.ConclusionsThis study identified that within an external exposome dataset, the most important predictors for SPGH status are related to mental wellbeing, physical exercise, loneliness, and financial status.

Project description:BackgroundVery preterm infants are at elevated risk for neurodevelopmental delays. Earlier prediction of delays allows timelier intervention and improved outcomes. Machine learning (ML) was used to predict mental and psychomotor delay at 25 months.MethodsWe applied RandomForest classifier to data from 1109 very preterm infants recruited over 20 years. ML selected key predictors from 52 perinatal and 16 longitudinal variables (1-22 mo assessments). SHapley Additive exPlanations provided model interpretability.ResultsBalanced accuracy with perinatal variables was 62%/61% (mental/psychomotor). Top predictors of mental and psychomotor delay overlapped and included: birth year, days in hospital, antenatal MgSO4, days intubated, birth weight, abnormal cranial ultrasound, gestational age, mom's age and education, and intrauterine growth restriction. Highest balanced accuracy was achieved with 19-month follow-up scores and perinatal variables (72%/73%).ConclusionsCombining perinatal and longitudinal data, ML modeling predicted 24 month mental/psychomotor delay in very preterm infants ½ year early, allowing intervention to start that much sooner. Modeling using only perinatal features fell short of clinical application. Birth year's importance reflected a linear decline in predicting delay as birth year became more recent.ImpactCombining perinatal and longitudinal data, ML modeling was able to predict 24 month mental/psychomotor delay in very preterm infants ½ year early (25% of their lives) potentially advancing implementation of intervention services. Although cognitive/verbal and fine/gross motor delays require separate interventions, in very preterm infants there is substantial overlap in the risk factors that can be used to predict these delays. Birth year has an important effect on ML prediction of delay in very preterm infants, with those born more recently (1989-2009) being increasing less likely to be delayed, perhaps reflecting advances in medical practice.

Project description:BackgroundIn the older general population, neurodegenerative diseases (NDs) are associated with increased disability, decreased physical and cognitive function. Detecting risk factors can help implement prevention measures. Using deep neural networks (DNNs), a machine-learning algorithm could be an alternative to Cox regression in tabular datasets with many predictive features. We aimed to compare the performance of different types of DNNs with regularized Cox proportional hazards models to predict NDs in the older general population.MethodsWe performed a longitudinal analysis with participants of the English Longitudinal Study of Ageing. We included men and women with no NDs at baseline, aged 60 years and older, assessed every 2 years from 2004 to 2005 (wave2) to 2016-2017 (wave 8). The features were a set of 91 epidemiological and clinical baseline variables. The outcome was new events of Parkinson's, Alzheimer or dementia. After applying multiple imputations, we trained three DNN algorithms: Feedforward, TabTransformer, and Dense Convolutional (Densenet). In addition, we trained two algorithms based on Cox models: Elastic Net regularization (CoxEn) and selected features (CoxSf).Results5433 participants were included in wave 2. During follow-up, 12.7% participants developed NDs. Although the five models predicted NDs events, the discriminative ability was superior using TabTransformer (Uno's C-statistic (coefficient (95% confidence intervals)) 0.757 (0.702, 0.805). TabTransformer showed superior time-dependent balanced accuracy (0.834 (0.779, 0.889)) and specificity (0.855 (0.0.773, 0.909)) than the other models. With the CoxSf (hazard ratio (95% confidence intervals)), age (10.0 (6.9, 14.7)), poor hearing (1.3 (1.1, 1.5)) and weight loss 1.3 (1.1, 1.6)) were associated with a higher DNN risk. In contrast, executive function (0.3 (0.2, 0.6)), memory (0, 0, 0.1)), increased gait speed (0.2, (0.1, 0.4)), vigorous physical activity (0.7, 0.6, 0.9)) and higher BMI (0.4 (0.2, 0.8)) were associated with a lower DNN risk.ConclusionTabTransformer is promising for prediction of NDs with heterogeneous tabular datasets with numerous features. Moreover, it can handle censored data. However, Cox models perform well and are easier to interpret than DNNs. Therefore, they are still a good choice for NDs.

Dataset Information

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes.

Publications

Predicting sex, age, general cognition and mental health with machine learning on brain structural connectomes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets