Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis

Exploring the potential of representation and transfer learning for anatomical neuroimaging: Application to psychiatry

The perspective of personalized medicine for brain disorders requires efficient learning models for anatomical neuroimaging-based prediction of clinical conditions. There is now a consensus on the benefit of deep learning (DL) in addressing many medical imaging tasks, such as image segmentation. However, for single-subject prediction problems, recent studies yielded contradictory results when comparing DL with Standard Machine Learning (SML) on top of classical feature extraction. Most existing comparative studies were limited in predicting phenotypes of little clinical interest, such as sex and age, and using a single dataset. Moreover, they conducted a limited analysis of the employed image pre-processing and feature selection strategies. This paper extensively compares DL and SML prediction capacity on five multi-site problems, including three increasingly complex clinical applications in psychiatry namely schizophrenia, bipolar disorder, and Autism Spectrum Disorder (ASD) diagnosis. To compensate for the relative scarcity of neuroimaging data on these clinical datasets, we also evaluate three pre-training strategies for transfer learning from brain imaging of the general healthy population: self-supervised learning, generative modeling and supervised learning with age. Overall, we find similar performance between randomly initialized DL and SML for the three clinical tasks and a similar scaling trend for sex prediction. This was replicated on an external dataset. We also show highly correlated discriminative brain regions between DL and linear ML models in all problems. Nonetheless, we demonstrate that self-supervised pre-training on large-scale healthy population imaging datasets (N≈10k), along with Deep Ensemble, allows DL to learn robust and transferable representations to smaller-scale clinical datasets (N≤1k). It largely outperforms SML on 2 out of 3 clinical tasks both in internal and external test sets. These findings suggest that the improvement of DL over SML in anatomical neuroimaging mainly comes from its capacity to learn meaningful and useful abstract representations of the brain anatomy, and it sheds light on the potential of transfer learning for personalized medicine in psychiatry.

The Contribution of Explainable Machine Learning Algorithms Using ROI-based Brain Surface Morphology Parameters in Distinguishing Early-onset Schizophrenia From Bipolar Disorder

RATIONALE AND OBJECTIVES

To differentiate early-onset schizophrenia (EOS) from early-onset bipolar disorder (EBD) using surface-based morphometry measurements and brain volumes using machine learning (ML) algorithms.

METHOD

High-resolution T1-weighted images were obtained to measure cortical thickness (CT), gyrification, gyrification index (GI), sulcal depth (SD), fractal dimension (FD), and brain volumes. After the feature selection step, ML classifiers were applied for each feature set and the combination of them. The SHapley Additive exPlanations (SHAP) technique was implemented to interpret the contribution of each feature.

FINDINGS

144 adolescents (16.2 ± 1.4 years, female=39%) with EOS (n = 81) and EBD (n = 63) were included. The Adaptive Boosting (AdaBoost) algorithm had the highest accuracy (82.75%) in the whole dataset that includes all variables from Destrieux atlas. The best-performing algorithms were K-nearest neighbors (KNN) for FD subset, support vector machine (SVM) for SD subset, and AdaBoost for GI subset. The KNN algorithm had the highest accuracy (accuracy=79.31%) in the whole dataset from the Desikan-Killiany-Tourville atlas.

CONCLUSION

This study demonstrates the use of ML in the differential diagnosis of EOS and EBD using surface-based morphometry measurements. Future studies could focus on multicenter data for the validation of these results.

Classification of suicidality by training supervised machine learning models with brain MRI findings: A systematic review

BACKGROUND

Suicide is a global public health issue causing around 700,000 deaths worldwide each year. Therefore, identifying suicidal thoughts and behaviors in patients can help lower the suicide-related mortality rate. This review aimed to investigate the feasibility of suicidality identification by applying supervised Machine Learning (ML) methods to Magnetic Resonance Imaging (MRI) data.

METHODS

We conducted a systematic search on PubMed, Scopus, and Web of Science to identify studies examining suicidality by applying ML methods to MRI features. Also, the Prediction Model Risk of Bias Assessment Tool (PROBAST) was employed for the quality assessment.

RESULTS

23 studies met the inclusion criteria. Of these, 20 developed prediction models without external validation and 3 developed prediction models with external validation. The performance of ML models varied among the reviewed studies, with the highest reported values of accuracies and Area Under the Curve (AUC) ranging from 51.7 % to 100 % and 0.52 to 1, respectively. Over half of the studies that reported accuracy (12/21) or AUC (13/16) achieved values of ≥0.8. Our comparative analysis indicated that deep learning exhibited the highest predictive performance compared to other ML models. The most commonly identified discriminative imaging features were resting-state functional connectivity and grey matter volume within prefrontal-limbic structures.

LIMITATIONS

Small sample sizes, lack of external validation, heterogeneous study designs, and ML model development.

CONCLUSIONS

Most of the studies developed ML models capable of ML-based suicide identification, although ML models' predictive performance varied across the reviewed studies. Thus, further well-designed is necessary to uncover the true potential of different ML models in this field.

Improved estimation of the risk of manic relapse by combining clinical and brain scan data

INTRODUCTION

Estimating the risk of manic relapse could help the psychiatrist individually adjust the treatment to the risk. Some authors have attempted to estimate this risk from baseline clinical data. Still, no studies have assessed whether the estimation could improve by adding structural magnetic resonance imaging (MRI) data. We aimed to evaluate it.

MATERIAL AND METHODS

We followed a cohort of 78 patients with a manic episode without mixed symptoms (bipolar type I or schizoaffective disorder) at 2-4-6-9-12-15-18 months and up to 10 years. Within a cross-validation scheme, we created and evaluated a Cox lasso model to estimate the risk of manic relapse using both clinical and MRI data.

RESULTS

The model successfully estimated the risk of manic relapse (Cox regression of the time to relapse as a function of the estimated risk: hazard ratio (HR)=2.35, p=0.027; area under the curve (AUC)=0.65, expected calibration error (ECE)<0.2). The most relevant variables included in the model were the diagnosis of schizoaffective disorder, poor impulse control, unusual thought content, and cerebellum volume decrease. The estimations were poorer when we used clinical or MRI data separately.

CONCLUSION

Combining clinical and MRI data may improve the risk of manic relapse estimation after a manic episode. We provide a website that estimates the risk according to the model to facilitate replication by independent groups before translation to clinical settings.

A meta-analysis and systematic review of single vs. multimodal neuroimaging techniques in the classification of psychosis

BACKGROUND

Psychotic disorders are characterized by structural and functional abnormalities in brain networks. Neuroimaging techniques map and characterize such abnormalities using unique features (e.g., structural integrity, coactivation). However, it is unclear if a specific method, or a combination of modalities, is particularly effective in identifying differences in brain networks of someone with a psychotic disorder.

METHODS

A systematic meta-analysis evaluated machine learning classification of schizophrenia spectrum disorders in comparison to healthy control participants using various neuroimaging modalities (i.e., T1-weighted imaging (T1), diffusion tensor imaging (DTI), resting state functional connectivity (rs-FC), or some combination (multimodal)). Criteria for manuscript inclusion included whole-brain analyses and cross-validation to provide a complete picture regarding the predictive ability of large-scale brain systems in psychosis. For this meta-analysis, we searched Ovid MEDLINE, PubMed, PsychInfo, Google Scholar, and Web of Science published between inception and March 13th 2023. Prediction results were averaged for studies using the same dataset, but parallel analyses were run that included studies with pooled sample across many datasets. We assessed bias through funnel plot asymmetry. A bivariate regression model determined whether differences in imaging modality, demographics, and preprocessing methods moderated classification. Separate models were run for studies with internal prediction (via cross-validation) and external prediction.

RESULTS

93 studies were identified for quantitative review (30 T1, 9 DTI, 40 rs-FC, and 14 multimodal). As a whole, all modalities reliably differentiated those with schizophrenia spectrum disorders from controls (OR = 2.64 (95%CI = 2.33 to 2.95)). However, classification was relatively similar across modalities: no differences were seen across modalities in the classification of independent internal data, and a small advantage was seen for rs-FC studies relative to T1 studies in classification in external datasets. We found large amounts of heterogeneity across results resulting in significant signs of bias in funnel plots and Egger's tests. Results remained similar, however, when studies were restricted to those with less heterogeneity, with continued small advantages for rs-FC relative to structural measures. Notably, in all cases, no significant differences were seen between multimodal and unimodal approaches, with rs-FC and unimodal studies reporting largely overlapping classification performance. Differences in demographics and analysis or denoising were not associated with changes in classification scores.

CONCLUSIONS

The results of this study suggest that neuroimaging approaches have promise in the classification of psychosis. Interestingly, at present most modalities perform similarly in the classification of psychosis, with slight advantages for rs-FC relative to structural modalities in some specific cases. Notably, results differed substantially across studies, with suggestions of biased effect sizes, particularly highlighting the need for more studies using external prediction and large sample sizes. Adopting more rigorous and systematized standards will add significant value toward understanding and treating this critical population.

An Overview of Bipolar Disorder Diagnosis Using Machine Learning Approaches: Clinical Opportunities and Challenges

Objective: Automatic diagnosis of psychiatric disorders such as bipolar disorder (BD) through machine learning techniques has attracted substantial attention from psychiatric and artificial intelligence communities. These approaches mostly rely on various biomarkers extracted from electroencephalogram (EEG) or magnetic resonance imaging (MRI)/functional MRI (fMRI) data. In this paper, we provide an updated overview of existing machine learning-based methods for bipolar disorder (BD) diagnosis using MRI and EEG data. Method : This study is a short non-systematic review with the aim of describing the current situation in automatic diagnosis of BD using machine learning methods. Therefore, an appropriate literature search was conducted via relevant keywords for original EEG/MRI studies on distinguishing BD from other conditions, particularly from healthy peers, in PubMed, Web of Science, and Google Scholar databases. Results: We reviewed 26 studies, including 10 EEG studies and 16 MRI studies (including structural and functional MRI), that used traditional machine learning methods and deep learning algorithms to automatically detect BD. The reported accuracies for EEG studies is about 90%, while the reported accuracies for MRI studies remains below the minimum level for clinical relevance, i.e. about 80% of the classification outcome for traditional machine learning methods. However, deep learning techniques have generally achieved accuracies higher than 95%. Conclusion: Research utilizing machine learning applied to EEG signals and brain images has provided proof of concept for how this innovative technique can help psychiatrists distinguish BD patients from healthy people. However, the results have been somewhat contradictory and we must keep away from excessive optimistic interpretations of the findings. Much progress is still needed to reach the level of clinical practice in this field.

Machine learning approaches linking brain function to behavior in the ABCD STOP task

The stop-signal task (SST) is one of the most common fMRI tasks of response inhibition, and its performance measure, the stop-signal reaction-time (SSRT), is broadly used as a measure of cognitive control processes. The neurobiology underlying individual or clinical differences in response inhibition remain unclear, consistent with the general pattern of quite modest brain-behavior associations that have been recently reported in well-powered large-sample studies. Here, we investigated the potential of multivariate, machine learning (ML) methods to improve the estimation of individual differences in SSRT with multimodal structural and functional region of interest-level neuroimaging data from 9- to 11-year-olds children in the ABCD Study. Six ML algorithms were assessed across modalities and fMRI tasks. We verified that SST activation performed best in predicting SSRT among multiple modalities including morphological MRI (cortical surface area/thickness), diffusion tensor imaging, and fMRI task activations, and then showed that SST activation explained 12% of the variance in SSRT using cross-validation and out-of-sample lockbox data sets (n = 7298). Brain regions that were more active during the task and that showed more interindividual variation in activation were better at capturing individual differences in performance on the task, but this was only true for activations when successfully inhibiting. Cortical regions outperformed subcortical areas in explaining individual differences but the two hemispheres performed equally well. These results demonstrate that the detection of reproducible links between brain function and performance can be improved with multivariate approaches and give insight into a number of brain systems contributing to individual differences in this fundamental cognitive control process.

Application of machine learning methods in predicting schizophrenia and bipolar disorders: A systematic review

Background and Aim

Schizophrenia and bipolar disorder (BD) are critical and high-risk inherited mental disorders with debilitating symptoms. Worldwide, 3% of the population suffers from these disorders. The mortality rate of these patients is higher compared to other people. Current procedures cannot effectively diagnose these disorders because it takes an average of 10 years from the onset of the first symptoms to the definitive diagnosis of the disease. Machine learning (ML) techniques are used to meet this need. This study aimed to summarize information on the use of ML techniques for predicting schizophrenia and BD to help early and timely diagnosis of the disease.

Methods

A systematic literature search included articles published until January 19, 2020 in 3 databases. Two reviewers independently assessed original papers to determine eligibility for inclusion in this review. PRISMA guidelines were followed to conduct the study, and the Prediction Model Risk of Bias Assessment Tool (PROBAST) to assess included papers.

Results

In this review, 1243 papers were retrieved through database searches, of which 15 papers were included based on full-text assessment. ML techniques were used to predict schizophrenia and BDs. The main algorithms applied were support vector machine (SVM) (10 studies), random forests (RF) (5 studies), and gradient boosting (GB) (3 studies). Input and output characteristics were very diverse and have been kept to enable future research. RFs algorithms demonstrated significantly higher accuracy and sensitivity than SVM and GB. GB demonstrated significantly higher specificity than SVM and RF. We found no significant difference between RF and SVM in terms of specificity.

Conclusion

ML can precisely predict results and assist in making clinical decisions-concerning schizophrenia and BD. RF often performed better than other algorithms in supervised learning tasks. This study identified gaps in the literature and opportunities for future psychological ML research.

Clinical predictors of antipsychotic treatment resistance: Development and internal validation of a prognostic prediction model by the STRATA-G consortium

INTRODUCTION

Our aim was to, firstly, identify characteristics at first-episode of psychosis that are associated with later antipsychotic treatment resistance (TR) and, secondly, to develop a parsimonious prediction model for TR.

METHODS

We combined data from ten prospective, first-episode psychosis cohorts from across Europe and categorised patients as TR or non-treatment resistant (NTR) after a mean follow up of 4.18 years (s.d. = 3.20) for secondary data analysis. We identified a list of potential predictors from clinical and demographic data recorded at first-episode. These potential predictors were entered in two models: a multivariable logistic regression to identify which were independently associated with TR and a penalised logistic regression, which performed variable selection, to produce a parsimonious prediction model. This model was internally validated using a 5-fold, 50-repeat cross-validation optimism-correction.

RESULTS

Our sample consisted of N = 2216 participants of which 385 (17 %) developed TR. Younger age of psychosis onset and fewer years in education were independently associated with increased odds of developing TR. The prediction model selected 7 out of 17 variables that, when combined, could quantify the risk of being TR better than chance. These included age of onset, years in education, gender, BMI, relationship status, alcohol use, and positive symptoms. The optimism-corrected area under the curve was 0.59 (accuracy = 64 %, sensitivity = 48 %, and specificity = 76 %).

IMPLICATIONS

Our findings show that treatment resistance can be predicted, at first-episode of psychosis. Pending a model update and external validation, we demonstrate the potential value of prediction models for TR.

Improving across-dataset brain tissue segmentation for MRI imaging using transformer

Brain tissue segmentation has demonstrated great utility in quantifying MRI data by serving as a precursor to further post-processing analysis. However, manual segmentation is highly labor-intensive, and automated approaches, including convolutional neural networks (CNNs), have struggled to generalize well due to properties inherent to MRI acquisition, leaving a great need for an effective segmentation tool. This study introduces a novel CNN-Transformer hybrid architecture designed to improve brain tissue segmentation by taking advantage of the increased performance and generality conferred by Transformers for 3D medical image segmentation tasks. We first demonstrate the superior performance of our model on various T1w MRI datasets. Then, we rigorously validate our model's generality applied across four multi-site T1w MRI datasets, covering different vendors, field strengths, scan parameters, and neuropsychiatric conditions. Finally, we highlight the reliability of our model on test-retest scans taken in different time points. In all situations, our model achieved the greatest generality and reliability compared to the benchmarks. As such, our method is inherently robust and can serve as a valuable tool for brain related T1w MRI studies. The code for the TABS network is available at: https://github.com/raovish6/TABS.

Combining MRI and clinical data to detect high relapse risk after the first episode of psychosis

Detecting patients at high relapse risk after the first episode of psychosis (HRR-FEP) could help the clinician adjust the preventive treatment. To develop a tool to detect patients at HRR using their baseline clinical and structural MRI, we followed 227 patients with FEP for 18–24 months and applied MRIPredict. We previously optimized the MRI-based machine-learning parameters (combining unmodulated and modulated gray and white matter and using voxel-based ensemble) in two independent datasets. Patients estimated to be at HRR-FEP showed a substantially increased risk of relapse (hazard ratio = 4.58, P < 0.05). Accuracy was poorer when we only used clinical or MRI data. We thus show the potential of combining clinical and MRI data to detect which individuals are more likely to relapse, who may benefit from increased frequency of visits, and which are unlikely, who may be currently receiving unnecessary prophylactic treatments. We also provide an updated version of the MRIPredict software.

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

The diagnosis based on clinical assessment of pediatric bipolar disorder (PBD) may sometimes lead to misdiagnosis in clinical practice. For the past several years, machine learning (ML) methods were introduced for the classification of bipolar disorder (BD), which were helpful in the diagnosis of BD. In this study, brain cortical thickness and subcortical volume of 33 PBD-I patients and 19 age-sex matched healthy controls (HCs) were extracted from the magnetic resonance imaging (MRI) data and set as features for classification. The dimensionality reduced feature subset, which was filtered by Lasso or f_classif, was sent to the six classifiers (logistic regression (LR), support vector machine (SVM), random forest classifier, naïve Bayes, k-nearest neighbor, and AdaBoost algorithm), and the classifiers were trained and tested. Among all the classifiers, the top two classifiers with the highest accuracy were LR (84.19%) and SVM (82.80%). Feature selection was performed in the six algorithms to obtain the most important variables including the right middle temporal gyrus and bilateral pallidum, which is consistent with structural and functional anomalous changes in these brain regions in PBD patients. These findings take the computer-aided diagnosis of BD a step forward.

A morphological study of schizophrenia with magnetic resonance imaging, advanced analytics, and machine learning

We have performed a morphological analysis of patients with schizophrenia and compared them with healthy controls. Our analysis includes the use of publicly available automated extraction tools to assess regional cortical thickness (inclusive of within region cortical thickness variability) from structural magnetic resonance imaging (MRI), to characterize group-wise abnormalities associated with schizophrenia based on a publicly available dataset. We have also performed a correlation analysis between the automatically extracted biomarkers and a variety of patient clinical variables available. Finally, we also present the results of a machine learning analysis. Results demonstrate regional cortical thickness abnormalities in schizophrenia. We observed a correlation (rho = 0.474) between patients’ depression and the average cortical thickness of the right medial orbitofrontal cortex. Our leading machine learning technology evaluated was the support vector machine with stepwise feature selection, yielding a sensitivity of 92% and a specificity of 74%, based on regional brain measurements, including from the insula, superior frontal, caudate, calcarine sulcus, gyrus rectus, and rostral middle frontal regions. These results imply that advanced analytic techniques combining MRI with automated biomarker extraction can be helpful in characterizing patients with schizophrenia.

Structural and diffusion MRI based schizophrenia classification using 2D pretrained and 3D naive Convolutional Neural Networks

The ability of automatic feature learning makes Convolutional Neural Network (CNN) potentially suitable to uncover the complex and widespread brain changes in schizophrenia. Despite that, limited studies have been done on schizophrenia identification using interpretable deep learning approaches on multimodal neuroimaging data. Here, we developed a deep feature approach based on pre-trained 2D CNN and naive 3D CNN models trained from scratch for schizophrenia classification by integrating 3D structural and diffusion magnetic resonance imaging (MRI) data. We found that the naive 3D CNN models outperformed the pretrained 2D CNN models and the handcrafted feature-based machine learning approach using support vector machine during both cross-validation and testing on an independent dataset. Multimodal neuroimaging-based models accomplished performance superior to models based on a single modality. Furthermore, we identified brain grey matter and white matter regions critical for illness classification at the individual- and group-level which supported the salience network and striatal dysfunction hypotheses in schizophrenia. Our findings underscore the potential of CNN not only to automatically uncover and integrate multimodal 3D brain imaging features for schizophrenia identification, but also to provide relevant neurobiological interpretations which are crucial for developing objective and interpretable imaging-based probes for prognosis and diagnosis in psychiatric disorders.

A greedy regression algorithm with coarse weights offers novel advantages

Regularized regression analysis is a mature analytic approach to identify weighted sums of variables predicting outcomes. We present a novel Coarse Approximation Linear Function (CALF) to frugally select important predictors and build simple but powerful predictive models. CALF is a linear regression strategy applied to normalized data that uses nonzero weights + 1 or - 1. Qualitative (linearly invariant) metrics to be optimized can be (for binary response) Welch (Student) t-test p-value or area under curve (AUC) of receiver operating characteristic, or (for real response) Pearson correlation. Predictor weighting is critically important when developing risk prediction models. While counterintuitive, it is a fact that qualitative metrics can favor CALF with ± 1 weights over algorithms producing real number weights. Moreover, while regression methods may be expected to change most or all weight values upon even small changes in input data (e.g., discarding a single subject of hundreds) CALF weights generally do not so change. Similarly, some regression methods applied to collinear or nearly collinear variables yield unpredictable magnitude or the direction (in p-space) of the weights as a vector. In contrast, with CALF if some predictors are linearly dependent or nearly so, CALF simply chooses at most one (the most informative, if any) and ignores the others, thus avoiding the inclusion of two or more collinear variables in the model.

Machine learning approaches for prediction of bipolar disorder based on biological, clinical and neuropsychological markers: a systematic review and meta-analysis

Applying machine learning (ML) to objective markers may overcome prognosis uncertainty due to the subjective nature of the diagnosis of bipolar disorder (BD). This PRISMA-compliant meta-analysis provides new systematic evidence of the BD classification accuracy reached by different markers and ML algorithms. We focused on neuroimaging, electrophysiological techniques, peripheral biomarkers, genetic data, neuropsychological or clinical measures, and multimodal approaches. PubMed, Embase and Scopus were searched through 3^rd December 2020. Meta-analyses were performed using random-effect models. Overall, 81 studies were included in this systematic review and 65 in the meta-analysis (11,336 participants, 3,903 BD). The overall pooled classification accuracy was 0.77 (95%CI[0.75;0.80]). Despite subgroup analyses for diagnostic comparison group, psychiatric disorders, marker, ML algorithm, and validation procedure were not significant, linear discriminant analysis significantly outperformed support vector machine for peripheral biomarkers (p=0.03). Sample size was inversely related to accuracy. Evidence of publication bias was detected. Ultimately, although ML reached a high accuracy in differentiating BD from other psychiatric disorders, best practices in methodology are needed for the advancement of future studies.

Distinguish bipolar and major depressive disorder in adolescents based on multimodal neuroimaging: Results from the Adolescent Brain Cognitive Development study®

Background

Major depressive disorder and bipolar disorder in adolescents are prevalent and are associated with cognitive impairment, executive dysfunction, and increased mortality. Early intervention in the initial stages of major depressive disorder and bipolar disorder can significantly improve personal health.

Methods

We collected 309 samples from the Adolescent Brain Cognitive Development study, including 116 adolescents with bipolar disorder, 64 adolescents with major depressive disorder, and 129 healthy adolescents, and employed a support vector machine to develop classification models for identification. We developed a multimodal model, which combined functional connectivity of resting-state functional magnetic resonance imaging and four anatomical measures of structural magnetic resonance imaging (cortical thickness, area, volume, and sulcal depth). We measured the performances of both multimodal and single modality classifiers.

Results

The multimodal classifiers showed outstanding performance compared with all five single modalities, and they are 100% for major depressive disorder versus healthy controls, 100% for bipolar disorder versus healthy control, 98.5% (95% CI: 95.4–100%) for major depressive disorder versus bipolar disorder, 100% for major depressive disorder versus depressed bipolar disorder and the leave-one-site-out analysis results are 77.4%, 63.3%, 79.4%, and 81.7%, separately.

Conclusions

The study shows that multimodal classifiers show high classification performances. Moreover, cuneus may be a potential biomarker to differentiate major depressive disorder, bipolar disorder, and healthy adolescents. Overall, this study can form multimodal diagnostic prediction workflows for clinically feasible to make more precise diagnose at the early stage and potentially reduce loss of personal pain and public society.

Searching for Imaging Biomarkers of Psychotic Dysconnectivity

BACKGROUND

Progress in precision psychiatry is predicated on identifying reliable individual-level diagnostic biomarkers. For psychosis, measures of structural and functional connectivity could be promising biomarkers given consistent reports of dysconnectivity across psychotic disorders using magnetic resonance imaging.

METHODS

We leveraged data from four independent cohorts of patients with psychosis and control subjects with observations from approximately 800 individuals. We used group-level analyses and two supervised machine learning algorithms (support vector machines and ridge regression) to test within-, between-, and across-sample classification performance of white matter and resting-state connectivity metrics.

RESULTS

Although we replicated group-level differences in brain connectivity, individual-level classification was suboptimal. Classification performance within samples was variable across folds (highest area under the curve [AUC] range = 0.30) and across datasets (average support vector machine AUC range = 0.50; average ridge regression AUC range = 0.18). Classification performance between samples was similarly variable or resulted in AUC values of approximately 0.65, indicating a lack of model generalizability. Furthermore, collapsing across samples (resting-state functional magnetic resonance imaging, N = 888; diffusion tensor imaging, N = 860) did not improve model performance (maximal AUC = 0.67). Ridge regression models generally outperformed support vector machine models, although classification performance was still suboptimal in terms of clinical relevance. Adjusting for demographic covariates did not greatly affect results.

CONCLUSIONS

Connectivity measures were not suitable as diagnostic biomarkers for psychosis as assessed in this study. Our results do not negate that other approaches may be more successful, although it is clear that a systematic approach to individual-level classification with large independent validation samples is necessary to properly vet neuroimaging features as diagnostic biomarkers.

Effects of Brain Atlases and Machine Learning Methods on the Discrimination of Schizophrenia Patients: A Multimodal MRI Study

Recently, machine learning techniques have been widely applied in discriminative studies of schizophrenia (SZ) patients with multimodal magnetic resonance imaging (MRI); however, the effects of brain atlases and machine learning methods remain largely unknown. In this study, we collected MRI data for 61 first-episode SZ patients (FESZ), 79 chronic SZ patients (CSZ) and 205 normal controls (NC) and calculated 4 MRI measurements, including regional gray matter volume (GMV), regional homogeneity (ReHo), amplitude of low-frequency fluctuation and degree centrality. We systematically analyzed the performance of two classifications (SZ vs NC; FESZ vs CSZ) based on the combinations of three brain atlases, five classifiers, two cross validation methods and 3 dimensionality reduction algorithms. Our results showed that the groupwise whole-brain atlas with 268 ROIs outperformed the other two brain atlases. In addition, the leave-one-out cross validation was the best cross validation method to select the best hyperparameter set, but the classification performances by different classifiers and dimensionality reduction algorithms were quite similar. Importantly, the contributions of input features to both classifications were higher with the GMV and ReHo features of brain regions in the prefrontal and temporal gyri. Furthermore, an ensemble learning method was performed to establish an integrated model, in which classification performance was improved. Taken together, these findings indicated the effects of these factors in constructing effective classifiers for psychiatric diseases and showed that the integrated model has the potential to improve the clinical diagnosis and treatment evaluation of SZ.

Detection of functional and structural brain alterations in female schizophrenia using elastic net logistic regression

Neuroimaging technique is a powerful tool to characterize the abnormality of brain networks in schizophrenia. However, the neurophysiological substrate of schizophrenia is still unclear. Here we investigated the patterns of brain functional and structural changes in female patients with schizophrenia using elastic net logistic regression analysis of resting-state functional magnetic resonance imaging data. Data from 52 participants (25 female schizophrenia patients and 27 healthy controls) were obtained. Using an elastic net penalty, the brain regions most relevant to schizophrenia pathology were defined in the models using the amplitude of low-frequency fluctuations (ALFF) and gray matter, respectively. The receiver operating characteristic analysis showed reliable classification accuracy with 85.7% in ALFF analysis, and 77.1% in gray matter analysis. Notably, our results showed eight common regions between the ALFF and gray matter analyses, including the Frontal-Inf-Orb-R, Rolandic-Oper-R, Olfactory-R, Angular-L, Precuneus-L, Precuenus-R, Heschl-L, and Temporal-Pole-Mid-R. In addition, the severity of symptoms was found positively associated with the ALFF within the Rolandic-Oper-R and Frontal-Inf-Orb-R. Our findings indicated that elastic net logistic regression could be a useful tool to identify the characteristics of schizophrenia -related brain deterioration, which provides novel insights into schizophrenia diagnosis and prediction.

Characterizing Brain Tumor Regions Using Texture Analysis in Magnetic Resonance Imaging

Purpose

To extract texture features from magnetic resonance imaging (MRI) scans of patients with brain tumors and use them to train a classification model for supporting an early diagnosis.

Methods

Two groups of regions (control and tumor) were selected from MRI scans of 40 patients with meningioma or glioma. These regions were analyzed to obtain texture features. Statistical analysis was conducted using SPSS (version 20.0), including the Shapiro-Wilk test and Wilcoxon signed-rank test, which were used to test significant differences in each feature between the tumor and healthy regions. T-distributed stochastic neighbor embedding (t-SNE) was used to visualize the data distribution so as to avoid tumor selection bias. The Gini impurity index in random forests (RFs) was used to select the top five out of all features. Based on the five features, three classification models were built respectively with three machine learning classifiers: RF, support vector machine (SVM), and back propagation (BP) neural network.

Results

Sixteen of the 25 features were significantly different between the tumor and healthy areas. Through the Gini impurity index in RFs, standard deviation, first-order moment, variance, third-order absolute moment, and third-order central moment were selected to build the classification model. The classification model trained using the SVM classifier achieved the best performance, with sensitivity, specificity, and area under the curve of 94.04%, 92.3%, and 0.932, respectively.

Conclusion

Texture analysis with an SVM classifier can help differentiate between brain tumor and healthy areas with high speed and accuracy, which would facilitate its clinical application.

Machine learning modelling of blood lipid biomarkers in familial hypercholesterolaemia versus polygenic/environmental dyslipidaemia

Familial hypercholesterolaemia increases circulating LDL-C levels and leads to premature cardiovascular disease when undiagnosed or untreated. Current guidelines support genetic testing in patients complying with clinical diagnostic criteria and cascade screening of their family members. However, most of hyperlipidaemic subjects do not present pathogenic variants in the known disease genes, and most likely suffer from polygenic hypercholesterolaemia, which translates into a relatively low yield of genetic screening programs. This study aims to identify new biomarkers and develop new approaches to improve the identification of individuals carrying monogenic causative variants. Using a machine-learning approach in a paediatric dataset of individuals, tested for disease causative genes and with an extended lipid profile, we developed new models able to classify familial hypercholesterolaemia patients with a much higher specificity than currently used methods. The best performing models incorporated parameters absent from the most common FH clinical criteria, namely apoB/apoA-I, TG/apoB and LDL1. These parameters were found to contribute to an improved identification of monogenic individuals. Furthermore, models using only TC and LDL-C levels presented a higher specificity of classification when compared to simple cut-offs. Our results can be applied towards the improvement of the yield of genetic screening programs and corresponding costs.

Artificial Intelligence in Clinical Neuroscience: Methodological and Ethical Challenges

Clinical neuroscience is increasingly relying on the collection of large volumes of differently structured data and the use of intelligent algorithms for data analytics. In parallel, the ubiquitous collection of unconventional data sources (e.g. mobile health, digital phenotyping, consumer neurotechnology) is increasing the variety of data points. Big data analytics and approaches to Artificial Intelligence (AI) such as advanced machine learning are showing great potential to make sense of these larger and heterogeneous data flows. AI provides great opportunities for making new discoveries about the brain, improving current preventative and diagnostic models in both neurology and psychiatry and developing more effective assistive neurotechnologies. Concurrently, it raises many new methodological and ethical challenges. Given their transformative nature, it is still largely unclear how AI-driven approaches to the study of the human brain will meet adequate standards of scientific validity and affect normative instruments in neuroethics and research ethics. This manuscript provides an overview of current AI-driven approaches to clinical neuroscience and an assessment of the associated key methodological and ethical challenges. In particular, it will discuss what ethical principles are primarily affected by AI approaches to human neuroscience, and what normative safeguards should be enforced in this domain.

Towards a brain-based predictome of mental illness

Neuroimaging-based approaches have been extensively applied to study mental illness in recent years and have deepened our understanding of both cognitively healthy and disordered brain structure and function. Recent advancements in machine learning techniques have shown promising outcomes for individualized prediction and characterization of patients with psychiatric disorders. Studies have utilized features from a variety of neuroimaging modalities, including structural, functional, and diffusion magnetic resonance imaging data, as well as jointly estimated features from multiple modalities, to assess patients with heterogeneous mental disorders, such as schizophrenia and autism. We use the term "predictome" to describe the use of multivariate brain network features from one or more neuroimaging modalities to predict mental illness. In the predictome, multiple brain network-based features (either from the same modality or multiple modalities) are incorporated into a predictive model to jointly estimate features that are unique to a disorder and predict subjects accordingly. To date, more than 650 studies have been published on subject-level prediction focusing on psychiatric disorders. We have surveyed about 250 studies including schizophrenia, major depression, bipolar disorder, autism spectrum disorder, attention-deficit hyperactivity disorder, obsessive-compulsive disorder, social anxiety disorder, posttraumatic stress disorder, and substance dependence. In this review, we present a comprehensive review of recent neuroimaging-based predictomic approaches, current trends, and common shortcomings and share our vision for future directions.

Support Vector Machine-Based Schizophrenia Classification Using Morphological Information from Amygdaloid and Hippocampal Subregions

Structural changes in the hippocampus and amygdala have been demonstrated in schizophrenia patients. However, whether morphological information from these subcortical regions could be used by machine learning algorithms for schizophrenia classification were unknown. The aim of this study was to use volume of the amygdaloid and hippocampal subregions for schizophrenia classification. The dataset consisted of 57 patients with schizophrenia and 69 healthy controls. The volume of 26 hippocampal and 20 amygdaloid subregions were extracted from T1 structural MRI images. Sequential backward elimination (SBE) algorithm was used for feature selection, and a linear support vector machine (SVM) classifier was configured to explore the feasibility of hippocampal and amygdaloid subregions in the classification of schizophrenia. The proposed SBE-SVM model achieved a classification accuracy of 81.75% on 57 patients and 69 healthy controls, with a sensitivity of 84.21% and a specificity of 81.16%. AUC was 0.8241 (p < 0.001 tested with 1000-times permutation). The results demonstrated evidence of hippocampal and amygdaloid structural changes in schizophrenia patients, and also suggested that morphological features from the amygdaloid and hippocampal subregions could be used by machine learning algorithms for the classification of schizophrenia.

No evidence for a relationship between social closeness and similarity in resting-state functional brain connectivity in schoolchildren

Previous research suggests that the proximity of individuals in a social network predicts how similarly their brains respond to naturalistic stimuli. However, the relationship between social connectedness and brain connectivity in the absence of external stimuli has not been examined. To investigate whether neural homophily between friends exists at rest we collected resting-state functional magnetic resonance imaging (fMRI) data from 68 school-aged girls, along with social network information from all pupils in their year groups (total 5,066 social dyads). Participants were asked to rate the amount of time they voluntarily spent with each person in their year group, and directed social network matrices and community structure were then determined from these data. No statistically significant relationships between social distance, community homogeneity and similarity of global-level resting-state connectivity were observed. Nor were we able to predict social distance using a regularised regression technique (i.e. elastic net regression based on the local-level similarities in resting-state whole-brain connectivity between participants). Although neural homophily between friends exists when viewing naturalistic stimuli, this finding did not extend to functional connectivity at rest in our population. Instead, resting-state connectivity may be less susceptible to the influences of a person's social environment.

Multivariate patterns of gray matter volume in thalamic nuclei are associated with positive schizotypy in healthy individuals

BACKGROUND

Previous models suggest biological and behavioral continua among healthy individuals (HC), at-risk condition, and full-blown schizophrenia (SCZ). Part of these continua may be captured by schizotypy, which shares subclinical traits and biological phenotypes with SCZ, including thalamic structural abnormalities. In this regard, previous findings have suggested that multivariate volumetric patterns of individual thalamic nuclei discriminate HC from SCZ. These results were obtained using machine learning, which allows case-control classification at the single-subject level. However, machine learning accuracy is usually unsatisfactory possibly due to phenotype heterogeneity. Indeed, a source of misclassification may be related to thalamic structural characteristics of those HC with high schizotypy, which may resemble structural abnormalities of SCZ. We hypothesized that thalamic structural heterogeneity is related to schizotypy, such that high schizotypal burden would implicate misclassification of those HC whose thalamic patterns resemble SCZ abnormalities.

METHODS

Following a previous report, we used Random Forests to predict diagnosis in a case-control sample (SCZ = 131, HC = 255) based on thalamic nuclei gray matter volumes estimates. Then, we investigated whether the likelihood to be classified as SCZ (π-SCZ) was associated with schizotypy in 174 HC, evaluated with the Schizotypal Personality Questionnaire.

RESULTS

Prediction accuracy was 72.5%. Misclassified HC had higher positive schizotypy scores, which were correlated with π-SCZ. Results were specific to thalamic rather than whole-brain structural features.

CONCLUSIONS

These findings strengthen the relevance of thalamic structural abnormalities to SCZ and suggest that multivariate thalamic patterns are correlates of the continuum between schizotypy in HC and the full-blown disease.

A Systematic Characterization of Structural Brain Changes in Schizophrenia

A systematic characterization of the similarities and differences among different methods for detecting structural brain abnormalities in schizophrenia, such as voxel-based morphometry (VBM), tensor-based morphometry (TBM), and projection-based thickness (PBT), is important for understanding the brain pathology in schizophrenia and for developing effective biomarkers for a diagnosis of schizophrenia. However, such studies are still lacking. Here, we performed VBM, TBM, and PBT analyses on T1-weighted brain MR images acquired from 116 patients with schizophrenia and 116 healthy controls. We found that, although all methods detected wide-spread structural changes, different methods captured different information - only 10.35% of the grey matter changes in cortex were detected by all three methods, and VBM only detected 11.36% of the white matter changes detected by TBM. Further, pattern classification between patients and controls revealed that combining different measures improved the classification accuracy (81.9%), indicating that fusion of different structural measures serves as a better neuroimaging marker for the objective diagnosis of schizophrenia.

Will machine learning applied to neuroimaging in bipolar disorder help the clinician? A critical review and methodological suggestions

OBJECTIVES

The existence of anatomofunctional brain abnormalities in bipolar disorder (BD) is now well established by magnetic resonance imaging (MRI) studies. To create diagnostic and prognostic tools, as well as identifying biologically valid subtypes of BD, research has recently turned towards the use of machine learning (ML) techniques. We assessed both supervised ML and unsupervised ML studies in BD to evaluate their robustness, reproducibility and the potential need for improvement.

METHOD

We systematically searched for studies using ML algorithms based on MRI data of patients with BD until February 2019.

RESULT

We identified 47 studies, 45 using supervised ML techniques and 2 including unsupervised ML analyses. Among supervised studies, 43 focused on diagnostic classification. The reported accuracies for classification of BD ranged between (a) 57% and 100%, for BD vs healthy controls; (b) 49.5% and 93.1% for BD vs patients with major depressive disorder; and (c) 50% and 96.2% for BD vs patients with schizophrenia. Reported accuracies for discriminating subjects genetically at risk for BD (either from control or from patients with BD) ranged between 64.3% and 88.93%.

CONCLUSIONS

Although there are strong methodological limitations in previous studies and an important need for replication in large multicentric samples, the conclusions of our review bring hope of future computer-aided diagnosis of BD and pave the way for other applications, such as treatment response prediction. To reinforce the reliability of future results we provide methodological suggestions for good practice in conducting and reporting MRI-based ML studies in BD.

Precision psychiatry with immunological and cognitive biomarkers: a multi-domain prediction for the diagnosis of bipolar disorder or schizophrenia using machine learning

Precision psychiatry is attracting increasing attention lately as a recognized priority. One of the goals of precision psychiatry is to develop tools capable of aiding a clinically informed psychiatric diagnosis objectively. Cognitive, inflammatory and immunological factors are altered in both bipolar disorder (BD) and schizophrenia (SZ), however, most of these alterations do not respect diagnostic boundaries from a phenomenological perspective and possess great variability in different individuals with the same phenotypic diagnosis and, consequently, none so far has proven to have the ability of reliably aiding in the differential diagnosis of BD and SZ. We developed a probabilistic multi-domain data integration model consisting of immune and inflammatory biomarkers in peripheral blood and cognitive biomarkers using machine learning to predict diagnosis of BD and SZ. A total of 416 participants, being 323, 372, and 279 subjects for blood, cognition and combined biomarkers analysis, respectively. Our multi-domain model performances for the BD vs. control (sensitivity 80% and specificity 71%) and for the SZ vs. control (sensitivity 84% and specificity 81%) pairs were high in general, however, our multi-domain model had only moderate performance for the differential diagnosis of BD and SZ (sensitivity 71% and specificity 73%). In conclusion, our results show that the diagnosis of BD and of SZ, and that the differential diagnosis of BD and SZ can be predicted with possible clinical utility by a computational machine learning algorithm employing blood and cognitive biomarkers, and that their integration in a multi-domain outperforms algorithms based in only one domain. Independent studies are needed to validate these findings.

Using Machine Learning and Structural Neuroimaging to Detect First Episode Psychosis: Reconsidering the Evidence

Despite the high level of interest in the use of machine learning (ML) and neuroimaging to detect psychosis at the individual level, the reliability of the findings is unclear due to potential methodological issues that may have inflated the existing literature. This study aimed to elucidate the extent to which the application of ML to neuroanatomical data allows detection of first episode psychosis (FEP), while putting in place methodological precautions to avoid overoptimistic results. We tested both traditional ML and an emerging approach known as deep learning (DL) using 3 feature sets of interest: (1) surface-based regional volumes and cortical thickness, (2) voxel-based gray matter volume (GMV) and (3) voxel-based cortical thickness (VBCT). To assess the reliability of the findings, we repeated all analyses in 5 independent datasets, totaling 956 participants (514 FEP and 444 within-site matched controls). The performance was assessed via nested cross-validation (CV) and cross-site CV. Accuracies ranged from 50% to 70% for surfaced-based features; from 50% to 63% for GMV; and from 51% to 68% for VBCT. The best accuracies (70%) were achieved when DL was applied to surface-based features; however, these models generalized poorly to other sites. Findings from this study suggest that, when methodological precautions are adopted to avoid overoptimistic results, detection of individuals in the early stages of psychosis is more challenging than originally thought. In light of this, we argue that the current evidence for the diagnostic value of ML and structural neuroimaging should be reconsidered toward a more cautious interpretation.

Using structural MRI to identify bipolar disorders - 13 site machine learning study in 3020 individuals from the ENIGMA Bipolar Disorders Working Group

Bipolar disorders (BDs) are among the leading causes of morbidity and disability. Objective biological markers, such as those based on brain imaging, could aid in clinical management of BD. Machine learning (ML) brings neuroimaging analyses to individual subject level and may potentially allow for their diagnostic use. However, fair and optimal application of ML requires large, multi-site datasets. We applied ML (support vector machines) to MRI data (regional cortical thickness, surface area, subcortical volumes) from 853 BD and 2167 control participants from 13 cohorts in the ENIGMA consortium. We attempted to differentiate BD from control participants, investigated different data handling strategies and studied the neuroimaging/clinical features most important for classification. Individual site accuracies ranged from 45.23% to 81.07%. Aggregate subject-level analyses yielded the highest accuracy (65.23%, 95% CI = 63.47-67.00, ROC-AUC = 71.49%, 95% CI = 69.39-73.59), followed by leave-one-site-out cross-validation (accuracy = 58.67%, 95% CI = 56.70-60.63). Meta-analysis of individual site accuracies did not provide above chance results. There was substantial agreement between the regions that contributed to identification of BD participants in the best performing site and in the aggregate dataset (Cohen's Kappa = 0.83, 95% CI = 0.829-0.831). Treatment with anticonvulsants and age were associated with greater odds of correct classification. Although short of the 80% clinically relevant accuracy threshold, the results are promising and provide a fair and realistic estimate of classification performance, which can be achieved in a large, ecologically valid, multi-site sample of BD participants based on regional neurostructural measures. Furthermore, the significant classification in different samples was based on plausible and similar neuroanatomical features. Future multi-site studies should move towards sharing of raw/voxelwise neuroimaging data.

Selective Review of Neuroimaging Findings in Youth at Clinical High Risk for Psychosis: On the Path to Biomarkers for Conversion

First episode psychosis (FEP), and subsequent diagnosis of schizophrenia or schizoaffective disorder, predominantly occurs during late adolescence, is accompanied by a significant decline in function and represents a traumatic experience for patients and families alike. Prior to first episode psychosis, most patients experience a prodromal period of 1-2 years, during which symptoms first appear and then progress. During that time period, subjects are referred to as being at Clinical High Risk (CHR), as a prodromal period can only be designated in hindsight in those who convert. The clinical high-risk period represents a critical window during which interventions may be targeted to slow or prevent conversion to psychosis. However, only one third of subjects at clinical high risk will convert to psychosis and receive a formal diagnosis of a primary psychotic disorder. Therefore, in order for targeted interventions to be developed and applied, predicting who among this population will convert is of critical importance. To date, a variety of neuroimaging modalities have identified numerous differences between CHR subjects and healthy controls. However, complicating attempts at predicting conversion are increasingly recognized co-morbidities, such as major depressive disorder, in a significant number of CHR subjects. The result of this is that phenotypes discovered between CHR subjects and healthy controls are likely non-specific to psychosis and generalized for major mental illness. In this paper, we selectively review evidence for neuroimaging phenotypes in CHR subjects who later converted to psychosis. We then evaluate the recent landscape of machine learning as it relates to neuroimaging phenotypes in predicting conversion to psychosis.

Structural abnormality in schizophrenia versus bipolar disorder: A whole brain cortical thickness, surface area, volume and gyrification analyses

OBJECTIVES

The profiles of cortical abnormalities in schizophrenia and bipolar disorder, and how far they resemble each other, have only been studied to a limited extent. The aim of this study was to identify and compare the changes in cortical morphology associated with these pathologies.

METHODS

A total of 384 subjects, including 128 patients with schizophrenia, 128 patients with bipolar disorder and 127 sex-age-matched healthy subjects, were examined using cortical surface-based morphology. Four cortical structural measures were studied: cortical volume (CV), cortical thickness (CT), surface area (SA) and gyrification index (GI). Group comparisons for each separate cortical measure were conducted.

RESULTS

At a threshold of P = 0.05 corrected, both patient groups showed significant widespread CV and CT reductions in similar areas compared to healthy subjects. However, the changes in schizophrenia were more pronounced. While CV decrease in bipolar disorder was exclusively explained by cortical thinning, in schizophrenia it was driven by changes in CT and partially by SA. Reduced GI was only found in schizophrenia. The direct comparison between both disorders showed significant reductions in all measures in patients with schizophrenia.

CONCLUSIONS

Cortical volume and cortical thickness deficits are shared between patients with schizophrenia and bipolar disorder, suggesting that both pathologies may be affected by similar environmental and neurodegenerative factors. However, the exclusive alteration in schizophrenia of metrics related to the geometry and curvature of the brain cortical surface (SA, GI) suggests that this group is influenced by additional neurodevelopmental and genetic factors.

Effective connectivity within a triple network brain system discriminates schizophrenia spectrum disorders from psychotic bipolar disorder at the single-subject level

OBJECTIVE

Schizophrenia spectrum disorders (SSD) and psychotic bipolar disorder share a number of genetic and neurobiological features, despite a divergence in clinical course and outcome trajectories. We studied the diagnostic classification potential that can be achieved on the basis of the structure and connectivity within a triple network system (the default mode, salience and central executive network) in patients with SSD and psychotic bipolar disorder.

METHODS

Directed static connectivity and its dynamic variance was estimated among 8 nodes of the three large-scale networks. Multivariate autoregressive models of deconvolved resting state functional magnetic resonance imaging time series were obtained from 57 patients (38 with SSD and 19 with bipolar disorder and psychosis). We used 2/3 of the patients for training and validation of the classifier and the remaining 1/3 as an independent hold-out test data for performance estimation.

RESULTS

A high level of discrimination between bipolar disorder with psychosis and SSD (combined balanced accuracy = 96.2%; class accuracies 100% for bipolar and 92.3% for SSD) was achieved when effective connectivity and morphometry of the triple network nodes was combined with symptom scores. Patients with SSD were discriminated from patients with bipolar disorder and psychosis as showing higher clinical severity of disorganization and higher variability in the effective connectivity between salience and executive networks.

CONCLUSIONS

Our results support the view that the study of network-level connectivity patterns can not only clarify the pathophysiology of SSD but also provide a measure of excellent clinical utility to identify discrete diagnostic/prognostic groups among individuals with psychosis.

Clinical Strategies and Technical Challenges in Psychoradiology

Psychoradiology is an emerging discipline at the intersection between radiology and psychiatry. It holds promise for playing a role in clinical diagnosis, evaluation of treatment response and prognosis, and illness risk prediction for patients with psychiatric disorders. Addressing complex issues, such as the biological heterogeneity of psychiatric syndromes and unclear neurobiological mechanisms underpinning radiological abnormalities, is a challenge that needs to be resolved. With the advance of multimodal imaging and more efforts in standardization of image acquisition and analysis, psychoradiology is becoming a promising tool for the future of clinical care for patients with psychiatric disorders.

Multimodal Integration of Brain Images for MRI-Based Diagnosis in Schizophrenia

Magnetic resonance imaging (MRI) has been proposed as a source of information for automatic prediction of individual diagnosis in schizophrenia. Optimal integration of data from different MRI modalities is an active area of research aimed at increasing diagnostic accuracy. Based on a sample of 96 patients with schizophrenia and a matched sample of 115 healthy controls that had undergone a single multimodal MRI session, we generated individual brain maps of gray matter vbm, 1back, and 2back levels of activation (nback fMRI), maps of amplitude of low-frequency fluctuations (resting-state fMRI), and maps of weighted global brain connectivity (resting-state fMRI). Four unimodal classifiers (Ridge, Lasso, Random Forests, and Gradient boosting) were applied to these maps to evaluate their classification accuracies. Based on the assignments made by the algorithms on test individuals, we quantified the amount of predictive information shared between maps (what we call redundancy analysis). Finally, we explored the added accuracy provided by a set of multimodal strategies that included post-classification integration based on probabilities, two-step sequential integration, and voxel-level multimodal integration through one-dimensional-convolutional neural networks (1D-CNNs). All four unimodal classifiers showed the highest test accuracies with the 2back maps (80% on average) achieving a maximum of 84% with the Lasso. Redundancy levels between brain maps were generally low (overall mean redundancy score of 0.14 in a 0–1 range), indicating that each brain map contained differential predictive information. The highest multimodal accuracy was delivered by the two-step Ridge classifier (87%) followed by the Ridge maximum and mean probability classifiers (both with 85% accuracy) and by the 1D-CNN, which achieved the same accuracy as the best unimodal classifier (84%). From these results, we conclude that from all MRI modalities evaluated task-based fMRI may be the best unimodal diagnostic option in schizophrenia. Low redundancy values point to ample potential for accuracy improvements through multimodal integration, with the two-step Ridge emerging as a suitable strategy.

Clinical-learning versus machine-learning for transdiagnostic prediction of psychosis onset in individuals at-risk

Predicting the onset of psychosis in individuals at-risk is based on robust prognostic model building methods including a priori clinical knowledge (also termed clinical-learning) to preselect predictors or machine-learning methods to select predictors automatically. To date, there is no empirical research comparing the prognostic accuracy of these two methods for the prediction of psychosis onset. In a first experiment, no improved performance was observed when machine-learning methods (LASSO and RIDGE) were applied-using the same predictors-to an individualised, transdiagnostic, clinically based, risk calculator previously developed on the basis of clinical-learning (predictors: age, gender, age by gender, ethnicity, ICD-10 diagnostic spectrum), and externally validated twice. In a second experiment, two refined versions of the published model which expanded the granularity of the ICD-10 diagnosis were introduced: ICD-10 diagnostic categories and ICD-10 diagnostic subdivisions. Although these refined versions showed an increase in apparent performance, their external performance was similar to the original model. In a third experiment, the three refined models were analysed under machine-learning and clinical-learning with a variable event per variable ratio (EPV). The best performing model under low EPVs was obtained through machine-learning approaches. The development of prognostic models on the basis of a priori clinical knowledge, large samples and adequate events per variable is a robust clinical prediction method to forecast psychosis onset in patients at-risk, and is comparable to machine-learning methods, which are more difficult to interpret and implement. Machine-learning methods should be preferred for high dimensional data when no a priori knowledge is available.

Reproducible grey matter patterns index a multivariate, global alteration of brain structure in schizophrenia and bipolar disorder

Schizophrenia is a severe mental disorder characterized by numerous subtle changes in brain structure and function. Machine learning allows exploring the utility of combining structural and functional brain magnetic resonance imaging (MRI) measures for diagnostic application, but this approach has been hampered by sample size limitations and lack of differential diagnostic data. Here, we performed a multi-site machine learning analysis to explore brain structural patterns of T1 MRI data in 2668 individuals with schizophrenia, bipolar disorder or attention-deficit/ hyperactivity disorder, and healthy controls. We found reproducible changes of structural parameters in schizophrenia that yielded a classification accuracy of up to 76% and provided discrimination from ADHD, through it lacked specificity against bipolar disorder. The observed changes largely indexed distributed grey matter alterations that could be represented through a combination of several global brain-structural parameters. This multi-site machine learning study identified a brain-structural signature that could reproducibly differentiate schizophrenia patients from controls, but lacked specificity against bipolar disorder. While this currently limits the clinical utility of the identified signature, the present study highlights that the underlying alterations index substantial global grey matter changes in psychotic disorders, reflecting the biological similarity of these conditions, and provide a roadmap for future exploration of brain structural alterations in psychiatric patients.

Machine learning techniques in a structural and functional MRI diagnostic approach in schizophrenia: a systematic review

BACKGROUND

Diagnosis of schizophrenia (SCZ) is made exclusively clinically, since specific biomarkers that can predict the disease accurately remain unknown. Machine learning (ML) represents a promising approach that could support clinicians in the diagnosis of mental disorders.

OBJECTIVES

A systematic review, according to the PRISMA statement, was conducted to evaluate its accuracy to distinguish SCZ patients from healthy controls.

METHODS

We systematically searched PubMed, Embase, MEDLINE, PsychINFO and the Cochrane Library through December 2018 using generic terms for ML techniques and SCZ without language or time restriction. Thirty-five studies were included in this review: eight of them used structural neuroimaging, twenty-six used functional neuroimaging and one both, with a minimum accuracy >60% (most of them 75-90%). Sensitivity, Specificity and accuracy were extracted from each publication or obtained directly from authors.

RESULTS

Support vector machine, the most frequent technique, if associated with other ML techniques achieved accuracy close to 100%. The prefrontal and temporal cortices appeared to be the most useful brain regions for the diagnosis of SCZ. ML analysis can efficiently detect significantly altered brain connectivity in patients with SCZ (eg, default mode network, visual network, sensorimotor network, frontoparietal network and salience network).

CONCLUSION

The greater accuracy demonstrated by these predictive models and the new models resulting from the integration of multiple ML techniques will be increasingly decisive for early diagnosis and evaluation of the treatment response and to establish the prognosis of patients with SCZ. To achieve a real benefit for patients, the future challenge is to reach an accurate diagnosis not only through clinical evaluation but also with the aid of ML algorithms.

Multimodal Neuroimaging: Basic Concepts and Classification of Neuropsychiatric Diseases

Neuroimaging techniques are widely used in neuroscience to visualize neural activity, to improve our understanding of brain mechanisms, and to identify biomarkers-especially for psychiatric diseases; however, each neuroimaging technique has several limitations. These limitations led to the development of multimodal neuroimaging (MN), which combines data obtained from multiple neuroimaging techniques, such as electroencephalography, functional magnetic resonance imaging, and yields more detailed information about brain dynamics. There are several types of MN, including visual inspection, data integration, and data fusion. This literature review aimed to provide a brief summary and basic information about MN techniques (data fusion approaches in particular) and classification approaches. Data fusion approaches are generally categorized as asymmetric and symmetric. The present review focused exclusively on studies based on symmetric data fusion methods (data-driven methods), such as independent component analysis and principal component analysis. Machine learning techniques have recently been introduced for use in identifying diseases and biomarkers of disease. The machine learning technique most widely used by neuroscientists is classification-especially support vector machine classification. Several studies differentiated patients with psychiatric diseases and healthy controls with using combined datasets. The common conclusion among these studies is that the prediction of diseases increases when combining data via MN techniques; however, there remain a few challenges associated with MN, such as sample size. Perhaps in the future N-way fusion can be used to combine multiple neuroimaging techniques or nonimaging predictors (eg, cognitive ability) to overcome the limitations of MN.

Gender, Personality Traits and Experience With Psychiatric Patients as Predictors of Stigma in Italian Psychology Students

A sample of undergraduate Psychology students (n = 1005), prevalently females (82.4%), mean age 20.5 (sd 2.5), was examined regarding their attitudes toward people suffering from mental illness. The survey instrument included a brief form for demographic variables, the Attribution Questionnaire-9 (AQ-9), the Ten Items Personality Inventory (TIPI), and two questions exploring attitudes toward open-door and restraint-free policies in Psychiatry. Higher levels of stigmatizing attitudes were found in males (Pity, Blame, Help, and Avoidance) and in those (76.5%) who had never had any experience with psychiatric patients (Danger, Fear, Blame, Segregation, Help, Avoidance and Coercion). A similar trend was also found in those who don't share the policy of no seclusion/restraint, while subjects who are favorable to open-door policies reported higher Coercion scores. No correlations were found between dimensions of stigma and personality traits. A machine learning approach was then used to explore the role of demographic, academic and personality variables as predictors of stigmatizing attitudes. Agreeableness and Extraversion emerged as the most relevant predictors for blaming attitudes, while Emotional Stability and Openness appeared to be the most effective contributors to Anger. Our results confirmed that a training experience in Psychiatry might successfully reduce stigma in Psychology students. Further research, with increased generalizability of samples and more reliable instruments, should address the role of personality traits and gender on attitudes toward people suffering from mental illness.

Identifying a neuroanatomical signature of schizophrenia, reproducible across sites and stages, using machine learning with structured sparsity

OBJECTIVE

Structural MRI (sMRI) increasingly offers insight into abnormalities inherent to schizophrenia. Previous machine learning applications suggest that individual classification is feasible and reliable and, however, is focused on the predictive performance of the clinical status in cross-sectional designs, which has limited biological perspectives. Moreover, most studies depend on relatively small cohorts or single recruiting site. Finally, no study controlled for disease stage or medication's effect. These elements cast doubt on previous findings' reproducibility.

METHOD

We propose a machine learning algorithm that provides an interpretable brain signature. Using large datasets collected from 4 sites (276 schizophrenia patients, 330 controls), we assessed cross-site prediction reproducibility and associated predictive signature. For the first time, we evaluated the predictive signature regarding medication and illness duration using an independent dataset of first-episode patients.

RESULTS

Machine learning classifiers based on neuroanatomical features yield significant intersite prediction accuracies (72%) together with an excellent predictive signature stability. This signature provides a neural score significantly correlated with symptom severity and the extent of cognitive impairments. Moreover, this signature demonstrates its efficiency on first-episode psychosis patients (73% accuracy).

CONCLUSION

These results highlight the existence of a common neuroanatomical signature for schizophrenia, shared by a majority of patients even from an early stage of the disorder.