An automated machine learning approach to predict brain age from cortical anatomical measures

A perspective on brain-age estimation and its clinical promise

Brain-age estimation has gained increased attention in the neuroscientific community owing to its potential use as a biomarker of brain health. The difference between estimated and chronological age based on neuroimaging data enables a unique perspective on brain development and aging, with multiple open questions still remaining in the brain-age research field. This Perspective presents an overview of current advancements in the field and envisions the future evolution of the brain-age framework before its potential deployment in hospital settings.

Predicting brain age gap with radiomics and automl: A Promising approach for age-Related brain degeneration biomarkers

The Brain Age Gap (BAG), which refers to the difference between chronological age and predicted neuroimaging age, is proposed as a potential biomarker for age-related brain degeneration. However, existing brain age prediction models usually rely on a single marker and can not discover meaningful hidden information in radiographic images. This study focuses on the application of radiomics, an advanced imaging analysis technique, combined with automated machine learning to predict BAG. Our methods achieve a promising result with a mean absolute error of 1.509 using the Alzheimer's Disease Neuroimaging Initiative dataset. Furthermore, we find that the hippocampus and parahippocampal gyrus play a significant role in predicting age with interpretable method called SHapley Additive exPlanations. Additionally, our investigation of age prediction discrepancies between patients with Alzheimer's disease (AD) and those with mild cognitive impairment (MCI) reveals a notable correlation with clinical cognitive assessment scale scores. This suggests that BAG has the potential to serve as a biomarker to support the diagnosis of AD and MCI. Overall, this study presents valuable insights into the application of neuroimaging models in the diagnosis of neurodegenerative diseases.

MRI-based brain age prediction model for children under 3 years old using deep residual network

Early identification and intervention of abnormal brain development individual subjects are of great significance, especially during the earliest and most active stage of brain development in children aged under 3. Neuroimage-based brain's biological age has been associated with health, ability, and remaining life. However, the existing brain age prediction models based on neuroimage are predominantly adult-oriented. Here, we collected 658 T1-weighted MRI scans from 0 to 3 years old healthy controls and developed an accurate brain age prediction model for young children using deep learning techniques with high accuracy in capturing age-related changes. The performance of the deep learning-based model is comparable to that of the SVR-based model, showcasing remarkable precision and yielding a noteworthy correlation of 91% between the predicted brain age and the chronological age. Our results demonstrate the accuracy of convolutional neural network (CNN) brain-predicted age using raw T1-weighted MRI data with minimum preprocessing necessary. We also applied our model to children with low birth weight, premature delivery history, autism, and ADHD, and discovered that the brain age was delayed in children with extremely low birth weight (less than 1000 g) while ADHD may cause accelerated aging of the brain. Our child-specific brain age prediction model can be a valuable quantitative tool to detect abnormal brain development and can be helpful in the early identification and intervention of age-related brain disorders.

Use Test of Automated Machine Learning in Cancer Diagnostics

Our aim is to investigate the added value of automated machine learning (AutoML) for potential future applications in cancer diagnostics. Using two important diagnostic questions, the non-invasive determination of IDH mutation status and ATRX status, we analyze whether it is possible to use AutoML to develop models that are comparable in performance to conventional machine learning models (ML) developed by experts. For this purpose, we develop AutoML models using different feature preselection methods and compare the results with previously developed conventional ML models. The cohort used for our study comprises T2-weighted MRI images of 124 patients with histologically confirmed gliomas. Using AutoML, we were able to develop sophisticated models in a very short time with only a few lines of computer code. In predicting IDH mutation status, we obtained a mean AUC of 0.7400 and a mean AUPRC of 0.8582. ATRX mutation status was predicted with very similar discriminatory power, with a mean AUC of 0.7810 and a mean AUPRC of 0.8511. In both cases, AutoML was even able to achieve a discriminatory power slightly above that of the respective conventionally developed models in a very short computing time, thus making such methods accessible to non-experts in the near future.

Biological brain age prediction using machine learning on structural neuroimaging data: Multi-cohort validation against biomarkers of Alzheimer's disease and neurodegeneration stratified by sex

Brain-age can be inferred from structural neuroimaging and compared to chronological age (brain-age delta) as a marker of biological brain aging. Accelerated aging has been found in neurodegenerative disorders like Alzheimer's disease (AD), but its validation against markers of neurodegeneration and AD is lacking. Here, imaging-derived measures from the UK Biobank dataset (N=22,661) were used to predict brain-age in 2,314 cognitively unimpaired (CU) individuals at higher risk of AD and mild cognitive impaired (MCI) patients from four independent cohorts with available biomarker data: ALFA+, ADNI, EPAD, and OASIS. Brain-age delta was associated with abnormal amyloid-β, more advanced stages (AT) of AD pathology and APOE-ε4 status. Brain-age delta was positively associated with plasma neurofilament light, a marker of neurodegeneration, and sex differences in the brain effects of this marker were found. These results validate brain-age delta as a non-invasive marker of biological brain aging in non-demented individuals with abnormal levels of biomarkers of AD and axonal injury.

Natural Language Processing in Electronic Health Records in relation to healthcare decision-making: A systematic review

BACKGROUND

Natural Language Processing (NLP) is widely used to extract clinical insights from Electronic Health Records (EHRs). However, the lack of annotated data, automated tools, and other challenges hinder the full utilisation of NLP for EHRs. Various Machine Learning (ML), Deep Learning (DL) and NLP techniques are studied and compared to understand the limitations and opportunities in this space comprehensively.

METHODOLOGY

After screening 261 articles from 11 databases, we included 127 papers for full-text review covering seven categories of articles: (1) medical note classification, (2) clinical entity recognition, (3) text summarisation, (4) deep learning (DL) and transfer learning architecture, (5) information extraction, (6) Medical language translation and (7) other NLP applications. This study follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

RESULT AND DISCUSSION

EHR was the most commonly used data type among the selected articles, and the datasets were primarily unstructured. Various ML and DL methods were used, with prediction or classification being the most common application of ML or DL. The most common use cases were: the International Classification of Diseases, Ninth Revision (ICD-9) classification, clinical note analysis, and named entity recognition (NER) for clinical descriptions and research on psychiatric disorders.

CONCLUSION

We find that the adopted ML models were not adequately assessed. In addition, the data imbalance problem is quite important, yet we must find techniques to address this underlining problem. Future studies should address key limitations in studies, primarily identifying Lupus Nephritis, Suicide Attempts, perinatal self-harmed and ICD-9 classification.

Using machine learning to estimate the calendar age based on autonomic cardiovascular function

Introduction

Aging is accompanied by physiological changes in cardiovascular regulation that can be evaluated using a variety of metrics. In this study, we employ machine learning on autonomic cardiovascular indices in order to estimate participants' age.

Methods

We analyzed a database including resting state electrocardiogram and continuous blood pressure recordings of healthy volunteers. A total of 884 data sets met the inclusion criteria. Data of 72 other participants with an BMI indicating obesity (>30 kg/m²) were withheld as an evaluation sample. For all participants, 29 different cardiovascular indices were calculated including heart rate variability, blood pressure variability, baroreflex function, pulse wave dynamics, and QT interval characteristics. Based on cardiovascular indices, sex and device, four different approaches were applied in order to estimate the calendar age of healthy subjects, i.e., relevance vector regression (RVR), Gaussian process regression (GPR), support vector regression (SVR), and linear regression (LR). To estimate age in the obese group, we drew normal-weight controls from the large sample to build a training set and a validation set that had an age distribution similar to the obesity test sample.

Results

In a five-fold cross validation scheme, we found the GPR model to be suited best to estimate calendar age, with a correlation of r=0.81 and a mean absolute error of MAE=5.6 years. In men, the error (MAE=5.4 years) seemed to be lower than that in women (MAE=6.0 years). In comparison to normal-weight subjects, GPR and SVR significantly overestimated the age of obese participants compared with controls. The highest age gap indicated advanced cardiovascular aging by 5.7 years in obese participants.

Discussion

In conclusion, machine learning can be used to estimate age on cardiovascular function in a healthy population when considering previous models of biological aging. The estimated age might serve as a comprehensive and readily interpretable marker of cardiovascular function. Whether it is a useful risk predictor should be investigated in future studies.

Brain Age Prediction in Developing Childhood with Multimodal Magnetic Resonance Images

It is well known that brain development is very fast and complex in the early childhood with age-based neurological and physiological changes of brain structure and function. The brain maturity is an important indicator for evaluating the normal development of children. In this paper, we propose a multimodal regression framework to combine the features from structural magnetic resonance imaging (sMRI) and diffusion tensor imaging (DTI) data for age prediction of children. First, three types of features are extracted from sMRI and DTI data. Second, we propose to combine the sparse coding and Q-Learning for feature selection from each modality. Finally, the ensemble regression is performed by random forest based on proximity measures to fuse multimodal features for age prediction. The proposed method is evaluated on 212 participants, including 76 young children less than 2 years old and 136 children aged from 2-15 years old recruited from Shanghai Children's Hospital. The results show that integrating multimodal features has achieved the highest accuracies with the root mean squared error (RMSE) of 0.208 years and mean absolute error (MAE) of 0.150 years for age prediction of young children (0-2), and RMSE of 1.666 years and MAE of 1.087 years for older children (2-15). We have shown that the selected features by Q-Learning can consistently improve the prediction accuracy. The comparison of prediction results demonstrates that the proposed method performs better than other competing methods.

Multimodal brain age prediction fusing morphometric and imaging data and association with cardiovascular risk factors

Introduction

The difference between the chronological and biological brain age, called the brain age gap (BAG), has been identified as a promising biomarker to detect deviation from normal brain aging and to indicate the presence of neurodegenerative diseases. Moreover, the BAG has been shown to encode biological information about general health, which can be measured through cardiovascular risk factors. Current approaches for biological brain age estimation, and therefore BAG estimation, either depend on hand-crafted, morphological measurements extracted from brain magnetic resonance imaging (MRI) or on direct analysis of brain MRI images. The former can be processed with traditional machine learning models while the latter is commonly processed with convolutional neural networks (CNNs). Using a multimodal setting, this study aims to compare both approaches in terms of biological brain age prediction accuracy and biological information captured in the BAG.

Methods

T1-weighted MRI, containing brain tissue information, and magnetic resonance angiography (MRA), providing information about brain arteries, from 1,658 predominantly healthy adults were used. The volumes, surface areas, and cortical thickness of brain structures were extracted from the T1-weighted MRI data, while artery density and thickness within the major blood flow territories and thickness of the major arteries were extracted from MRA data. Independent multilayer perceptron and CNN models were trained to estimate the brain age from the hand-crafted features and image data, respectively. Next, both approaches were fused to assess the benefits of combining image data and hand-crafted features for brain age prediction.

Results

The combined model achieved a mean absolute error of 4 years between the chronological and predicted biological brain age. Among the independent models, the lowest mean absolute error was observed for the CNN using T1-weighted MRI data (4.2 years). When evaluating the BAGs obtained using the different approaches and imaging modalities, diverging associations between cardiovascular risk factors were found. For example, BAGs obtained from the CNN models showed an association with systolic blood pressure, while BAGs obtained from hand-crafted measurements showed greater associations with obesity markers.

Discussion

In conclusion, the use of more diverse sources of data can improve brain age estimation modeling and capture more diverse biological deviations from normal aging.

Radiomic study on preoperative multi-modal magnetic resonance images identifies IDH-mutant TERT promoter-mutant gliomas

OBJECTIVES

Gliomas with comutations of isocitrate dehydrogenase (IDH) genes and telomerase reverse transcriptase (TERT) gene promoter (IDHmut pTERTmut) show distinct biological features and respond to first-line treatment differently in comparison with other gliomas. This study aimed to characterize the IDHmut pTERTmut gliomas in multimodal MRI using the radiomic method and establish a precise diagnostic model identifying this group of gliomas.

METHODS

A total of 140 patients with untreated primary gliomas were admitted between 2016 and 2020 to West China Hospital as a discovery cohort, including 22 IDHmut pTERTmut patients. Thirty-four additional cases from a different hospital were included in the study as an independent validation cohort. A total of 3654 radiomic features were extracted from the preoperative multimodal MRI images (T1c, FLAIR, and ADC maps) and filtered in a data-driven approach. The discovery cohort was split into training and test sets by a 4:1 ratio. A diagnostic model (multilayer perceptron classifier) for detecting the IDHmut pTERTmut gliomas was trained using an automatic machine-learning algorithm named tree-based pipeline optimization tool (TPOT). The most critical radiomic features in the model were identified and visualized.

RESULTS

The model achieved an area under the receiver-operating curve (AUROC) of 0.971 (95% CI, 0.902-1.000), the sensitivity of 0.833 (95% CI, 0.333-1.000), and the specificity of 0.966 (95% CI, 0.931-1.000) in the test set. The area under the precision-recall curve (AUCPR) was 0.754 (95% CI, 0.572-0.833) and the F1 score was 0.833 (95% CI, 0.500-1.000). In the independent validation set, the model reached 0.952 AUROC, 0.714 sensitivity, 0.963 specificity, 0.841 AUCPR, and 0.769 F1 score. MR radiomic features of the IDHmut pTERTmut gliomas represented homogenous low-complexity texture in three modalities.

CONCLUSIONS

An accurate diagnostic model was constructed for detecting IDHmut pTERTmut gliomas using multimodal radiomic features. The most important features were associated with the homogenous simple texture of IDHmut pTERTmut gliomas in MRI images transformed using Laplacian of Gaussian and wavelet filters.

Accelerated functional brain aging in major depressive disorder: evidence from a large scale fMRI analysis of Chinese participants

Major depressive disorder (MDD) is one of the most common mental health conditions that has been intensively investigated for its association with brain atrophy and mortality. Recent studies suggest that the deviation between the predicted and the chronological age can be a marker of accelerated brain aging to characterize MDD. However, current conclusions are usually drawn based on structural MRI information collected from Caucasian participants. The universality of this biomarker needs to be further validated by subjects with different ethnic/racial backgrounds and by different types of data. Here we make use of the REST-meta-MDD, a large scale resting-state fMRI dataset collected from multiple cohort participants in China. We develop a stacking machine learning model based on 1101 healthy controls, which estimates a subject's chronological age from fMRI with promising accuracy. The trained model is then applied to 1276 MDD patients from 24 sites. We observe that MDD patients exhibit a +4.43 years (p < 0.0001, Cohen's d = 0.31, 95% CI: 2.23-3.88) higher brain-predicted age difference (brain-PAD) compared to controls. In the MDD subgroup, we observe a statistically significant +2.09 years (p < 0.05, Cohen's d = 0.134525) brain-PAD in antidepressant users compared to medication-free patients. The statistical relationship observed is further checked by three different machine learning algorithms. The positive brain-PAD observed in participants in China confirms the presence of accelerated brain aging in MDD patients. The utilization of functional brain connectivity for age estimation verifies existing findings from a new dimension.

Automated Classification of Atherosclerotic Radiomics Features in Coronary Computed Tomography Angiography (CCTA)

Radiomics is the process of extracting useful quantitative features of high-dimensional data that allows for automated disease classification, including atherosclerotic disease. Hence, this study aimed to quantify and extract the radiomic features from Coronary Computed Tomography Angiography (CCTA) images and to evaluate the performance of automated machine learning (AutoML) model in classifying the atherosclerotic plaques. In total, 202 patients who underwent CCTA examination at Institut Jantung Negara (IJN) between September 2020 and May 2021 were selected as they met the inclusion criteria. Three primary coronary arteries were segmented on axial sectional images, yielding a total of 606 volume of interest (VOI). Subsequently, the first order, second order, and shape order of radiomic characteristics were extracted for each VOI. Model 1, Model 2, Model 3, and Model 4 were constructed using AutoML-based Tree-Pipeline Optimization Tools (TPOT). The heatmap confusion matrix, recall (sensitivity), precision (PPV), F1 score, accuracy, receiver operating characteristic (ROC), and area under the curve (AUC) were analysed. Notably, Model 1 with the first-order features showed superior performance in classifying the normal coronary arteries (F1 score: 0.88; Inverse F1 score: 0.94), as well as in classifying the calcified (F1 score: 0.78; Inverse F1 score: 0.91) and mixed plaques (F1 score: 0.76; Inverse F1 score: 0.86). Moreover, Model 2 consisting of second-order features was proved useful, specifically in classifying the non-calcified plaques (F1 score: 0.63; Inverse F1 score: 0.92) which are a key point for prediction of cardiac events. Nevertheless, Model 3 comprising the shape-based features did not contribute to the classification of atherosclerotic plaques. Overall, TPOT shown promising capabilities in terms of finding the best pipeline and tailoring the model using CCTA-based radiomic datasets.

Image-Based Differentiation of Intracranial Metastasis From Glioblastoma Using Automated Machine Learning

Purpose

The majority of solitary brain metastases appear similar to glioblastomas (GBMs) on magnetic resonance imaging (MRI). This study aimed to develop and validate an MRI-based model to differentiate intracranial metastases from GBMs using automated machine learning.

Materials and Methods

Radiomics features from 354 patients with brain metastases and 354 with GBMs were used to build prediction algorithms based on T2-weighted images, contrast-enhanced (CE) T1-weighted images, or both. The data of these subjects were subjected to a nested 10-fold split in the training and testing groups to build the best algorithms using the tree-based pipeline optimization tool (TPOT). The algorithms were independently validated using data from 124 institutional patients with solitary brain metastases and 103 patients with GBMs from the cancer genome atlas.

Results

Three groups of models were developed. The average areas under the receiver operating characteristic curve (AUCs) were 0.856 for CE T1-weighted images, 0.976 for T2-weighted images, and 0.988 for a combination in the testing groups, and the AUCs of the groups of models in the independent validation were 0.687, 0.831, and 0.867, respectively. A total of 149 radiomics features were considered as the most valuable features for the differential diagnosis of GBMs and metastases.

Conclusion

The models established by TPOT can distinguish glioblastoma from solitary brain metastases well, and its non-invasiveness, convenience, and robustness make it potentially useful for clinical applications.

Implementing Critical Machine Learning (ML) Approaches for Generating Robust Discriminative Neuroimaging Representations Using Structural Equation Model (SEM)

Critical ML or CML is a critical approach development of the standard ML (SML) procedure. Conventional ML (ML) is being used in radiology departments where complex neuroimages are discriminated using ML technology. Radiologists and researchers found that sole decision by the ML algorithms is not accurate enough to implement the treatment procedure. Thus, an intelligent decision is required further by the radiologists after evaluating the ML outcomes. The current research is based on the critical ML, where radiologists' critical thinking ability, IQ (intelligence quotient), and experience in radiology have been examined to understand how these factors affect the accuracy of neuroimaging discrimination. A primary quantitative survey has been carried out, and the data were analysed in IBM SPSS. The results showed that experience in works has a positive impact on neuroimaging discrimination accuracy. IQ and trained ML are also responsible for improving the accuracy as well. Thus, radiologists with more experience in that field are able to improve the discriminative and diagnostic capability of CML.

Automatic machine learning based on native T1 mapping can identify myocardial fibrosis in patients with hypertrophic cardiomyopathy

OBJECTIVES

To investigate the feasibility of automatic machine learning (autoML) based on native T1 mapping to predict late gadolinium enhancement (LGE) status in hypertrophic cardiomyopathy (HCM).

METHODS

Ninety-one HCM patients and 44 healthy controls who underwent cardiovascular MRI were enrolled. The native T1 maps of HCM patients were classified as LGE ( +) or LGE (-) based on location-matched LGE images. An autoML pipeline was implemented using the tree-based pipeline optimization tool (TPOT) for 3 binary classifications: LGE ( +) and LGE (-), LGE (-) and control, and HCM and control. TPOT modeling was repeated 10 times to obtain the optimal model for each classification. The diagnostic performance of the best models by slice and by case was evaluated using sensitivity, specificity, accuracy, and microaveraged area under the curve (AUC).

RESULTS

Ten prediction models were generated by TPOT for each of the 3 binary classifications. The diagnostic accuracy obtained with the best pipeline in detecting LGE status in the testing cohort of HCM patients was 0.80 by slice and 0.79 by case. In addition, the TPOT model also showed discriminability between LGE (-) patients and control (accuracy: 0.77 by slice; 0.78 by case) and for all HCM patients and controls (accuracy: 0.88 for both).

CONCLUSIONS

Native T1 map analysis based on autoML correlates with LGE ( +) or (-) status. The TPOT machine learning algorithm could be a promising method for predicting myocardial fibrosis, as reflected by the presence of LGE in HCM patients without the need for late contrast-enhanced MRI sequences.

KEY POINTS

• The tree-based pipeline optimization tool (TPOT) is a machine learning algorithm that could help predict late gadolinium enhancement (LGE) status in patients with hypertrophic cardiomyopathy. • The TPOT could serve as an adjuvant method to detect LGE by using information from native T1 maps, thus avoiding the need for contrast agent. • The TPOT also detects native T1 map alterations in LGE-negative patients with hypertrophic cardiomyopathy.

Automated Machine Learning for Healthcare and Clinical Notes Analysis

Machine learning (ML) has been slowly entering every aspect of our lives and its positive impact has been astonishing. To accelerate embedding ML in more applications and incorporating it in real-world scenarios, automated machine learning (AutoML) is emerging. The main purpose of AutoML is to provide seamless integration of ML in various industries, which will facilitate better outcomes in everyday tasks. In healthcare, AutoML has been already applied to easier settings with structured data such as tabular lab data. However, there is still a need for applying AutoML for interpreting medical text, which is being generated at a tremendous rate. For this to happen, a promising method is AutoML for clinical notes analysis, which is an unexplored research area representing a gap in ML research. The main objective of this paper is to fill this gap and provide a comprehensive survey and analytical study towards AutoML for clinical notes. To that end, we first introduce the AutoML technology and review its various tools and techniques. We then survey the literature of AutoML in the healthcare industry and discuss the developments specific to clinical settings, as well as those using general AutoML tools for healthcare applications. With this background, we then discuss challenges of working with clinical notes and highlight the benefits of developing AutoML for medical notes processing. Next, we survey relevant ML research for clinical notes and analyze the literature and the field of AutoML in the healthcare industry. Furthermore, we propose future research directions and shed light on the challenges and opportunities this emerging field holds. With this, we aim to assist the community with the implementation of an AutoML platform for medical notes, which if realized can revolutionize patient outcomes.

A Comparison of Quantitative R1 and Cortical Thickness in Identifying Age, Lifespan Dynamics, and Disease States of the Human Cortex

Brain development and aging are complex processes that unfold in multiple brain regions simultaneously. Recently, models of brain age prediction have aroused great interest, as these models can potentially help to understand neurological diseases and elucidate basic neurobiological mechanisms. We test whether quantitative magnetic resonance imaging can contribute to such age prediction models. Using R1, the longitudinal rate of relaxation, we explore lifespan dynamics in cortical gray matter. We compare R1 with cortical thickness, a well-established biomarker of brain development and aging. Using 160 healthy individuals (6-81 years old), we found that R1 and cortical thickness predicted age similarly, but the regions contributing to the prediction differed. Next, we characterized R1 development and aging dynamics. Compared with anterior regions, in posterior regions we found an earlier R1 peak but a steeper postpeak decline. We replicate these findings: firstly, we tested a subset (N = 10) of the original dataset for whom we had additional scans at a lower resolution; and second, we verified the results on an independent dataset (N = 34). Finally, we compared the age prediction models on a subset of 10 patients with multiple sclerosis. The patients are predicted older than their chronological age using R1 but not with cortical thickness.

Automated machine learning to predict the co-occurrence of isocitrate dehydrogenase mutations and O6 -methylguanine-DNA methyltransferase promoter methylation in patients with gliomas

Combining isocitrate dehydrogenase mutation (IDHmut) with O⁶ -methylguanine-DNA methyltransferase promoter methylation (MGMTmet) has been identified as a critical prognostic molecular marker for gliomas. The aim of this study was to determine the ability of glioma radiomics features from magnetic resonance imaging (MRI) to predict the co-occurrence of IDHmut and MGMTmet by applying the tree-based pipeline optimization tool (TPOT), an automated machine learning (autoML) approach. This was a retrospective study, in which 162 patients with gliomas were evaluated, including 58 patients with co-occurrence of IDHmut and MGMTmet and 104 patients with other status comprising: IDH wildtype and MGMT unmethylated (n = 67), IDH wildtype and MGMTmet (n = 36), and IDHmut and MGMT unmethylated (n = 1). Three-dimensional (3D) T1-weighted images, gadolinium-enhanced 3D T1-weighted images (Gd-3DT1WI), T2-weighted images, and fluid-attenuated inversion recovery (FLAIR) images acquired at 3.0 T were used. Radiomics features were extracted from FLAIR and Gd-3DT1WI images. The TPOT was employed to generate the best machine learning pipeline, which contains both feature selector and classifier, based on input feature sets. A 4-fold cross-validation was used to evaluate the performance of automatically generated models. For each iteration, the training set included 121 subjects, while the test set included 41 subjects. Student's t-test or a chi-square test was applied on different clinical characteristics between two groups. Sensitivity, specificity, accuracy, kappa score, and AUC were used to evaluate the performance of TPOT-generated models. Finally, we compared the above metrics of TPOT-generated models to identify the best-performing model. Patients' ages and grades between two groups were significantly different (p = 0.002 and p = 0.000, respectively). The 4-fold cross-validation showed that gradient boosting classifier trained on shape and textual features from the Laplacian-of-Gaussian-filtered Gd-3DT1 achieved the best performance (average sensitivity = 81.1%, average specificity = 94%, average accuracy = 89.4%, average kappa score = 0.76, average AUC = 0.951). Using autoML based on radiomics features from MRI, a high discriminatory accuracy was achieved for predicting co-occurrence of IDHmut and MGMTmet in gliomas. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 3.

An automated machine learning approach to predict brain age from cortical anatomical measures

The use of machine learning (ML) algorithms has significantly increased in neuroscience. However, from the vast extent of possible ML algorithms, which one is the optimal model to predict the target variable? What are the hyperparameters for such a model? Given the plethora of possible answers to these questions, in the last years, automated ML (autoML) has been gaining attention. Here, we apply an autoML library called Tree-based Pipeline Optimisation Tool (TPOT) which uses a tree-based representation of ML pipelines and conducts a genetic programming-based approach to find the model and its hyperparameters that more closely predicts the subject's true age. To explore autoML and evaluate its efficacy within neuroimaging data sets, we chose a problem that has been the focus of previous extensive study: brain age prediction. Without any prior knowledge, TPOT was able to scan through the model space and create pipelines that outperformed the state-of-the-art accuracy for Freesurfer-based models using only thickness and volume information for anatomical structure. In particular, we compared the performance of TPOT (mean absolute error [MAE]: 4.612 ± .124 years) and a relevance vector regression (MAE 5.474 ± .140 years). TPOT also suggested interesting combinations of models that do not match the current most used models for brain prediction but generalise well to unseen data. AutoML showed promising results as a data-driven approach to find optimal models for neuroimaging applications.

Brain-Age Prediction Using Shallow Machine Learning: Predictive Analytics Competition 2019

As we age, our brain structure changes and our cognitive capabilities decline. Although brain aging is universal, rates of brain aging differ markedly, which can be associated with pathological mechanism of psychiatric and neurological diseases. Predictive models have been applied to neuroimaging data to learn patterns associated with this variability and develop a neuroimaging biomarker of the brain condition. Aiming to stimulate the development of more accurate brain-age predictors, the Predictive Analytics Competition (PAC) 2019 provided a challenge that included a dataset of 2,640 participants. Here, we present our approach which placed between the top 10 of the challenge. We developed an ensemble of shallow machine learning methods (e.g., Support Vector Regression and Decision Tree-based regressors) that combined voxel-based and surface-based morphometric data. We used normalized brain volume maps (i.e., gray matter, white matter, or both) and features of cortical regions and anatomical structures, like cortical thickness, volume, and mean curvature. In order to fine-tune the hyperparameters of the machine learning methods, we combined the use of genetic algorithms and grid search. Our ensemble had a mean absolute error of 3.7597 years on the competition, showing the potential that shallow methods still have in predicting brain-age.

Brain Age Prediction With Morphological Features Using Deep Neural Networks: Results From Predictive Analytic Competition 2019

Morphological changes in the brain over the lifespan have been successfully described by using structural magnetic resonance imaging (MRI) in conjunction with machine learning (ML) algorithms. International challenges and scientific initiatives to share open access imaging datasets also contributed significantly to the advance in brain structure characterization and brain age prediction methods. In this work, we present the results of the predictive model based on deep neural networks (DNN) proposed during the Predictive Analytic Competition 2019 for brain age prediction of 2638 healthy individuals. We used FreeSurfer software to extract some morphological descriptors from the raw MRI scans of the subjects collected from 17 sites. We compared the proposed DNN architecture with other ML algorithms commonly used in the literature (RF, SVR, Lasso). Our results highlight that the DNN models achieved the best performance with MAE = 4.6 on the hold-out test, outperforming the other ML strategies. We also propose a complete ML framework to perform a robust statistical evaluation of feature importance for the clinical interpretability of the results.