Multi-Site Diagnostic Classification of Schizophrenia Using Discriminant Deep Learning with Functional Connectivity MRI

Measuring functional connectivity in frequency-domain helps to better characterize brain function

Resting-state functional connectivity (FC) is widely used in multivariate pattern analysis of functional magnetic resonance imaging (fMRI), including identifying the locations of putative brain functional borders, predicting individual phenotypes, and diagnosing clinical mental diseases. However, limited attention has been paid to the analysis of functional interactions from a frequency perspective. In this study, by contrasting coherence-based and correlation-based FC with two machine learning tasks, we observed that measuring FC in the frequency domain helped to identify finer functional subregions and achieve better pattern discrimination capability relative to the temporal correlation. This study has proven the feasibility of coherence in the analysis of fMRI, and the results indicate that modeling functional interactions in the frequency domain may provide richer information than that in the time domain, which may provide a new perspective on the analysis of functional neuroimaging.

Connectional-style-guided contextual representation learning for brain disease diagnosis

Structural magnetic resonance imaging (sMRI) has shown great clinical value and has been widely used in deep learning (DL) based computer-aided brain disease diagnosis. Previous DL-based approaches focused on local shapes and textures in brain sMRI that may be significant only within a particular domain. The learned representations are likely to contain spurious information and have poor generalization ability in other diseases and datasets. To facilitate capturing meaningful and robust features, it is necessary to first comprehensively understand the intrinsic pattern of the brain that is not restricted within a single data/task domain. Considering that the brain is a complex connectome of interlinked neurons, the connectional properties in the brain have strong biological significance, which is shared across multiple domains and covers most pathological information. In this work, we propose a connectional style contextual representation learning model (CS-CRL) to capture the intrinsic pattern of the brain, used for multiple brain disease diagnosis. Specifically, it has a vision transformer (ViT) encoder and leverages mask reconstruction as the proxy task and Gram matrices to guide the representation of connectional information. It facilitates the capture of global context and the aggregation of features with biological plausibility. The results indicate that CS-CRL achieves superior accuracy in multiple brain disease diagnosis tasks across six datasets and three diseases and outperforms state-of-the-art models. Furthermore, we demonstrate that CS-CRL captures more brain-network-like properties, and better aggregates features, is easier to optimize, and is more robust to noise, which explains its superiority in theory.

Diagnosing schizophrenia using deep learning: Novel interpretation approaches and multi-site validation

Schizophrenia is a profound and enduring mental disorder that imposes significant negative impacts on individuals, their families, and society at large. The development of more accurate and objective diagnostic tools for schizophrenia can be expedited through the employment of deep learning (DL), that excels at deciphering complex hierarchical non-linear patterns. However, the limited interpretability of deep learning has eroded confidence in the model and restricted its clinical utility. At the same time, if the data source is only derived from a single center, the model's generalizability is difficult to test. To enhance the model's reliability and applicability, leave-one-center-out validation with a large and diverse sample from multiple centers is crucial. In this study, we utilized Nine different global centers to train and test the 3D Resnet model's generalizability, resulting in an 82% classification performance (area under the curve) on all datasets sourced from different countries, employing a leave-one-center-out-validation approach. Per our approximation of the feature significance of each region on the atlas, we identified marked differences in the thalamus, pallidum, and inferior frontal gyrus between individuals with schizophrenia and healthy controls, lending credence to prior research findings. At the same time, in order to translate the model's output into clinically applicable insights, the SHapley Additive exPlanations (SHAP) permutation explainer method with an anatomical atlas have been refined, thereby offering precise neuroanatomical and functional interpretations of different brain regions.

Gradient Matching Federated Domain Adaptation for Brain Image Classification

Federated learning has shown its unique advantages in many different tasks, including brain image analysis. It provides a new way to train deep learning models while protecting the privacy of medical image data from multiple sites. However, previous studies suggest that domain shift across different sites may influence the performance of federated models. As a solution, we propose a gradient matching federated domain adaptation (GM-FedDA) method for brain image classification, aiming to reduce domain discrepancy with the assistance of a public image dataset and train robust local federated models for target sites. It mainly includes two stages: 1) pretraining stage; we propose a one-common-source adversarial domain adaptation (OCS-ADA) strategy, i.e., adopting ADA with gradient matching loss to pretrain encoders for reducing domain shift at each target site (private data) with the assistance of a common source domain (public data) and 2) fine-tuning stage; we develop a gradient matching federated (GM-Fed) fine-tuning method for updating local federated models pretrained with the OCS-ADA strategy, i.e., pushing the optimization direction of a local federated model toward its specific local minimum by minimizing gradient matching loss between sites. Using fully connected networks as local models, we validate our method with the diagnostic classification tasks of schizophrenia and major depressive disorder based on multisite resting-state functional MRI (fMRI), respectively. Results show that the proposed GM-FedDA method outperforms other commonly used methods, suggesting the potential of our method in brain imaging analysis and other fields, which need to utilize multisite data while preserving data privacy.

Effects of long-term closed and socially isolating spaceflight analog environment on default mode network connectivity as indicated by fMRI

Long-term manned spaceflight and extraterrestrial planet settlement become the focus of space powers. However, the potential influence of closed and socially isolating spaceflight on the brain function remains unclear. A 180-day controlled ecological life support system integrated experiment was conducted, establishing a spaceflight analog environment to explore the effect of long-term socially isolating living. Three crewmembers were enrolled and underwent resting-state fMRI scanning before and after the experiment. We performed both seed-based and network-based analyses to investigate the functional connectivity (FC) changes of the default mode network (DMN), considering its key role in multiple higher-order cognitive functions. Compared with normal controls, the leader of crewmembers exhibited significantly reduced within-DMN and between-DMN FC after the experiment, while two others exhibited opposite trends. Moreover, individual differences of FC changes were further supported by evidence from behavioral analyses. The findings may shed new light on the development of psychological protection for space exploration.

Sleep deprivation changes frequency-specific functional organization of the resting human brain

Previous resting-state functional magnetic resonance imaging (rs-fMRI) studies have widely explored the temporal connection changes in the human brain following long-term sleep deprivation (SD). However, the frequency-specific topological properties of sleep-deprived functional networks remain virtually unclear. In this study, thirty-seven healthy male subjects underwent resting-state fMRI during rested wakefulness (RW) and after 36 hours of SD, and we examined frequency-specific spectral connection changes (0.01-0.08 Hz, interval = 0.01 Hz) caused by SD. First, we conducted a multivariate pattern analysis combining linear SVM classifiers with a robust feature selection algorithm, and the results revealed that accuracies of 74.29%-84.29% could be achieved in the classification between RW and SD states in leave-one-out cross-validation at different frequency bands, moreover, the spectral connection at the lowest and highest frequency bands exhibited higher discriminative power. Connection involving the cingulo-opercular network increased most, while connection involving the default-mode network decreased most following SD. Then we performed a graph-theoretic analysis and observed reduced low-frequency modularity and high-frequency global efficiency in the SD state. Moreover, hub regions, which were primarily situated in the cerebellum and the cingulo-opercular network after SD, exhibited high discriminative power in the aforementioned classification consistently. The findings may indicate the frequency-dependent effects of SD on the functional network topology and its efficiency of information exchange, providing new insights into the impact of SD on the human brain.

Automated linguistic analysis in speech samples of Turkish-speaking patients with schizophrenia-spectrum disorders

Modern natural language processing (NLP) methods provide ways to objectively quantify language disturbances for potential use in diagnostic classification. We performed computerized language analysis in speech samples of 82 Turkish-speaking subjects, including 44 patients with schizophrenia spectrum disorders (SSD) and 38 healthy controls (HC). Exploratory analysis of speech samples involved 16 sentence-level semantic similarity features using SBERT (Sentence Bidirectional Encoder Representation from Text) as well as 8 generic and 8 part-of-speech (POS) features. The random forest classifier using SBERT-derived semantic similarity features achieved a mean accuracy of 85.6 % for the classification of SSD and HC. When semantic similarity features were combined with generic and POS features, the classifier's mean accuracy reached to 86.8 %. Our analysis reflected increased sentence-level semantic similarity scores in SSD. Generic and POS analyses revealed an increase in the use of verbs, proper nouns and pronouns in SSD while our results showed a decrease in the utilization of conjunctions, determiners, and both average and maximum sentence length in SSD compared to HC. Quantitative language features were correlated with the expressive deficit domain of BNSS (Brief Negative Symptom Scale) as well as with the duration of illness. These findings from Turkish-speaking interviews contribute to the growing evidence-based NLP-derived assessments in non-English-speaking patients.

The Use of Generative Adversarial Network and Graph Convolution Network for Neuroimaging-Based Diagnostic Classification

Functional connectivity (FC) obtained from resting-state functional magnetic resonance imaging has been integrated with machine learning algorithms to deliver consistent and reliable brain disease classification outcomes. However, in classical learning procedures, custom-built specialized feature selection techniques are typically used to filter out uninformative features from FC patterns to generalize efficiently on the datasets. The ability of convolutional neural networks (CNN) and other deep learning models to extract informative features from data with grid structure (such as images) has led to the surge in popularity of these techniques. However, the designs of many existing CNN models still fail to exploit the relationships between entities of graph-structure data (such as networks). Therefore, graph convolution network (GCN) has been suggested as a means for uncovering the intricate structure of brain network data, which has the potential to substantially improve classification accuracy. Furthermore, overfitting in classifiers can be largely attributed to the limited number of available training samples. Recently, the generative adversarial network (GAN) has been widely used in the medical field for its generative aspect that can generate synthesis images to cope with the problems of data scarcity and patient privacy. In our previous work, GCN and GAN have been designed to investigate FC patterns to perform diagnosis tasks, and their effectiveness has been tested on the ABIDE-I dataset. In this paper, the models will be further applied to FC data derived from more public datasets (ADHD, ABIDE-II, and ADNI) and our in-house dataset (PTSD) to justify their generalization on all types of data. The results of a number of experiments show the powerful characteristic of GAN to mimic FC data to achieve high performance in disease prediction. When employing GAN for data augmentation, the diagnostic accuracy across ADHD-200, ABIDE-II, and ADNI datasets surpasses that of other machine learning models, including results achieved with BrainNetCNN. Specifically, in ADHD, the accuracy increased from 67.74% to 73.96% with GAN, in ABIDE-II from 70.36% to 77.40%, and in ADNI, reaching 52.84% and 88.56% for multiclass and binary classification, respectively. GCN also obtains decent results, with the best accuracy in ADHD datasets at 71.38% for multinomial and 75% for binary classification, respectively, and the second-best accuracy in the ABIDE-II dataset (72.28% and 75.16%, respectively). Both GAN and GCN achieved the highest accuracy for the PTSD dataset, reaching 97.76%. However, there are still some limitations that can be improved. Both methods have many opportunities for the prediction and diagnosis of diseases.

Few-Shot Domain-Adaptive Anomaly Detection for Cross-Site Brain Images

Early screening is essential for effective intervention and treatment of individuals with mental disorders. Functional magnetic resonance imaging (fMRI) is a noninvasive tool for depicting neural activity and has demonstrated strong potential as a technique for identifying mental disorders. Due to the difficulty in data collection and diagnosis, imaging data from patients are rare at a single site, whereas abundant healthy control data are available from public datasets. However, joint use of these data from multiple sites for classification model training is hindered by cross-domain distribution discrepancy and diverse label spaces. Herein, we propose few-shot domain-adaptive anomaly detection (FAAD) to achieve cross-site anomaly detection of brain images based on only a few labeled samples. We introduce domain adaptation to mitigate cross-domain distribution discrepancy and jointly align the general and conditional feature distributions of imaging data across multiple sites. We utilize fMRI data of healthy subjects in the Human Connectome Project (HCP) as the source domain and fMRI images from six independent sites, including patients with mental disorders and demographically matched healthy controls, as target domains. Experiments showed the superiority of the proposed method compared with binary classification, traditional anomaly detection methods, and several recognized domain adaptation methods.

Structural connectome alterations between individuals with autism and neurotypical controls using feature representation learning

Autism spectrum disorder is one of the most common neurodevelopmental conditions associated with sensory and social communication impairments. Previous neuroimaging studies reported that atypical nodal- or network-level functional brain organization in individuals with autism was associated with autistic behaviors. Although dimensionality reduction techniques have the potential to uncover new biomarkers, the analysis of whole-brain structural connectome abnormalities in a low-dimensional latent space is underinvestigated. In this study, we utilized autoencoder-based feature representation learning for diffusion magnetic resonance imaging-based structural connectivity in 80 individuals with autism and 61 neurotypical controls that passed strict quality controls. We generated low-dimensional latent features using the autoencoder model for each group and adopted an integrated gradient approach to assess the contribution of the input data for predicting latent features during the encoding process. Subsequently, we compared the integrated gradient values between individuals with autism and neurotypical controls and observed differences within the transmodal regions and between the sensory and limbic systems. Finally, we identified significant associations between integrated gradient values and communication abilities in individuals with autism. Our findings provide insights into the whole-brain structural connectome in autism and may help identify potential biomarkers for autistic connectopathy.

Brain connectivity for subtypes of parkinson's disease using structural MRI

Objective. Delineating Parkinson's disease (PD) into distinct subtypes is a major challenge. Most studies use clinical symptoms to label PD subtypes while our work uses an imaging-based data-mining approach to subtype PD. Our study comprises two major objectives - firstly, subtyping Parkinson's patients based on grey matter information from structural magnetic resonance imaging scans of human brains; secondly, comparative structural brain connectivity analysis of PD subtypes derived from the former step.Approach. Source-based-morphometry decomposition was performed on 131 Parkinson's patients and 78 healthy controls from PPMI dataset, to derive at components (regions) with significance in disease and high effect size. The loading coefficients of significant components were thresholded for arriving at subtypes. Further, regional grey matter maps of subtype-specific subjects were separately parcellated and employed for construction of subtype-specific association matrices using Pearson correlation. These association matrices were binarized using sparsity threshold and leveraged for structural brain connectivity analysis using network metrics.Main results. Two distinct Parkinson's subtypes (namely A and B) were detected employing loadings of two components satisfying the selection criteria, and a third subtype (AB) was detected, common to these two components. Subtype A subjects were highly weighted in inferior, middle and superior frontal gyri while subtype B subjects in inferior, middle and superior temporal gyri. Network metrics analyses through permutation test revealed significant inter-subtype differences (p < 0.05) in clustering coefficient, local efficiency, participation coefficient and betweenness centrality. Moreover, hubs were obtained using betweenness centrality and mean network degree.Significance. MRI-based data-driven subtypes show frontal and temporal lobes playing a key role in PD. Graph theory-driven brain network analyses could untangle subtype-specific differences in structural brain connections showing differential network architecture. Replication of these initial results in other Parkinson's datasets may be explored in future. Clinical Relevance- Investigating structural brain connections in Parkinson's disease may provide subtype-specific treatment.

Brain network classification based on dynamic graph attention information bottleneck

BACKGROUND AND OBJECTIVES

Graph neural networks (GNN) have demonstrated remarkable encoding capabilities in the context of brain network classification tasks. They excel at uncovering hidden static connections between brain states. However, brain network signals can be influenced by physiological traits and external variables during clinical detection, resulting in noisy brain graphs. Additionally, many existing algorithms for brain networks primarily focus on static topologies determined by threshold-based criteria, thereby overlooking the real-time variability in brain channel connectivity. These sources of noise and the persistence of static structures inevitably hinder the effective exchange of information during brain network computations.

METHODS

To address these challenges, we propose a novel framework called the dynamic graph attention information bottleneck (DGAIB). This framework is designed to dynamically enhance the input raw brain graph structure from the perspective of information theory and graph theory. First, we employ the Spearman function to construct a raw graph. Then, we use a graph information bottleneck (GIB) to optimize the internal graph connections by selectively masking redundant feature embeddings. Finally, we enhance the feature aggregation of each brain state by utilizing a graph attention network (GAT), which promotes improved information exchange among distinct brain regions within the model. These processed representations serve as input for subsequent classification tasks.

EXPERIMENT AND RESULTS

We systematically evaluated the robustness and generalizability of our proposed framework through a series of experiments. This evaluation included patient-specific experiments using the electroencephalography (EEG)-based CHB-MIT dataset and cross-patient experiments leveraging the functional magnetic resonance imaging (fMRI)-based ABIDE-I dataset from multiple perspectives.

Multi feature fusion network for schizophrenia classification and abnormal brain network recognition

Schizophrenia classification and abnormal brain network recognition have an important research significance. Researchers have proposed many classification methods based on machine learning and deep learning. However, fewer studies utilized the advantages of complementary information from multi feature to learn the best representation of schizophrenia. In this study, we proposed a multi-feature fusion network (MFFN) using functional network connectivity (FNC) and time courses (TC) to distinguish schizophrenia patients from healthy controls. DNN backbone was adopted to learn the feature map of functional network connectivity, C-RNNAM backbone was designed to learn the feature map of time courses, and Deep SHAP was applied to obtain the most discriminative brain networks. We proved the effectiveness of this proposed model using the combining two public datasets and evaluated this model quantitatively using the evaluation indexes. The results showed that the functional network connectivity generated by independent component analysis has advantage in schizophrenia classification by comparing static and dynamic functional connections. This method obtained the best classification accuracy (ACC=87.30%, SPE=89.28%, SEN=85.71%, F1 =88.23%, and AUC=0.9081), and it demonstrated the superiority of this proposed model by comparing state-of-the-art methods. Ablation experiment also demonstrated that multi feature fusion and attention module can improve classification accuracy. The most discriminative brain networks showed that default mode network and visual network of schizophrenia patients have aberrant connections in brain networks. In conclusion, this method can identify schizophrenia effectively and visualize the abnormal brain network, and it has important clinical application value.

Identifying subgroups of eating behavior traits unrelated to obesity using functional connectivity and feature representation learning

Eating behavior is highly heterogeneous across individuals and cannot be fully explained using only the degree of obesity. We utilized unsupervised machine learning and functional connectivity measures to explore the heterogeneity of eating behaviors measured by a self-assessment instrument using 424 healthy adults (mean ± standard deviation [SD] age = 47.07 ± 18.89 years; 67% female). We generated low-dimensional representations of functional connectivity using resting-state functional magnetic resonance imaging and estimated latent features using the feature representation capabilities of an autoencoder by nonlinearly compressing the functional connectivity information. The clustering approaches applied to latent features identified three distinct subgroups. The subgroups exhibited different levels of hunger traits, while their body mass indices were comparable. The results were replicated in an independent dataset consisting of 212 participants (mean ± SD age = 38.97 ± 19.80 years; 35% female). The model interpretation technique of integrated gradients revealed that the between-group differences in the integrated gradient maps were associated with functional reorganization in heteromodal association and limbic cortices and reward-related subcortical structures such as the accumbens, amygdala, and caudate. The cognitive decoding analysis revealed that these systems are associated with reward- and emotion-related systems. Our findings provide insights into the macroscopic brain organization of eating behavior-related subgroups independent of obesity.

Accurate measurement of magnetic resonance parkinsonism index by a fully automatic and deep learning quantification pipeline

OBJECTIVES

This study aims at a fully automatic pipeline for measuring the magnetic resonance parkinsonism index (MRPI) using deep learning methods.

METHODS

MRPI is defined as the product of the pons area to the midbrain area ratio and the middle cerebellar peduncle (MCP) width to the superior cerebellar peduncle (SCP) width ratio. In our proposed pipeline, we first used nnUNet to segment the brainstem and then employed HRNet to identify two key boundary points so as to sub-divide the whole brainstem into midbrain and pons. HRNet was also employed to predict the MCP endpoints for measuring the MCP width. Finally, we segmented the SCP on an oblique coronal plane and calculated its width. A total of 400 T1-weighted magnetic resonance images (MRIs) were used to train the nnUNet and HRNet models. Five-fold cross-validation was conducted to evaluate our proposed pipeline's performance on the training dataset. We also evaluated the performance of our proposed pipeline on three external datasets. Two of them had two raters manually measuring the MRPI values, providing insights into automatic accuracy versus inter-rater variability.

RESULTS

We obtained average absolute percentage errors (APEs) of 17.21%, 18.17%, 20.83%, and 22.83% on the training dataset and the three external validation datasets, while the inter-rater average APE measured on the first two external validation datasets was 11.31%. Our proposed pipeline significantly improved the MRPI quantification accuracy over a representative state-of-the-art traditional approach (p < 0.001).

CONCLUSION

The proposed automatic pipeline can accurately predict MRPI that is comparable with manual measurement.

CLINICAL RELEVANCE STATEMENT

This study presents an automated magnetic resonance parkinsonism index measurement tool that can analyze large amounts of magnetic resonance images, enhance the efficiency of Parkinsonism-Plus syndrome diagnosis, reduce the workload of clinicians, and minimize the impact of human factors on diagnosis.

KEY POINTS

• We propose an automatic pipeline for measuring the magnetic resonance parkinsonism index from magnetic resonance images. • The effectiveness of the proposed pipeline is successfully established on multiple datasets and comparisons with inter-rater measurements. • The proposed pipeline significantly outperforms a state-of-the-art quantification approach, being much closer to ground truth.

A neuromarker for deficit syndrome in schizophrenia from a combination of structural and functional magnetic resonance imaging

AIM

Deficit schizophrenia (DS), defined by primary and enduring negative symptoms, has been proposed as a promising homogeneous subtype of schizophrenia. It has been demonstrated that unimodal neuroimaging characteristics of DS were different from non-deficit schizophrenia (NDS), however, whether multimodal-based neuroimaging features could identify deficit syndrome remains to be determined.

METHODS

Functional and structural multimodal magnetic resonance imaging of DS, NDS and healthy controls were scanned. Voxel-based features of gray matter volume, fractional amplitude of low-frequency fluctuations, and regional homogeneity were extracted. The support vector machine classification models were constructed using these features separately and jointly. The most discriminative features were defined as the first 10% of features with the greatest weights. Moreover, relevance vector regression was applied to explore the predictive values of these top-weighted features in predicting negative symptoms.

RESULTS

The multimodal classifier achieved a higher accuracy (75.48%) compared with the single modal model in distinguishing DS from NDS. The most predictive brain regions were mainly located in the default mode and visual networks, exhibiting differences between functional and structural features. Further, the identified discriminative features significantly predicted scores of diminished expressivity factor in DS but not NDS.

CONCLUSIONS

The present study demonstrated that local properties of brain regions extracted from multimodal imaging data could distinguish DS from NDS with a machine learning-based approach and confirmed the relationship between distinctive features and the negative symptoms subdomain. These findings may improve the identification of potential neuroimaging signatures and improve the clinical assessment of the deficit syndrome.

Elucidating salient site-specific functional connectivity features and site-invariant biomarkers in schizophrenia via deep neural networks

Schizophrenia is a highly heterogeneous disorder and salient functional connectivity (FC) features have been observed to vary across study sites, warranting the need for methods that can differentiate between site-invariant FC biomarkers and site-specific salient FC features. We propose a technique named Semi-supervised learning with data HaRmonisation via Encoder-Decoder-classifier (SHRED) to examine these features from resting state functional magnetic resonance imaging scans gathered from four sites. Our approach involves an encoder-decoder-classifier architecture that simultaneously performs data harmonisation and semi-supervised learning (SSL) to deal with site differences and labelling inconsistencies across sites respectively. The minimisation of reconstruction loss from SSL was shown to improve model performance even within small datasets whilst data harmonisation often led to lower model generalisability, which was unaffected using the SHRED technique. We show that our proposed model produces site-invariant biomarkers, most notably the connection between transverse temporal gyrus and paracentral lobule. Site-specific salient FC features were also elucidated, especially implicating the paracentral lobule for our local dataset. Our examination of these salient FC features demonstrates how site-specific features and site-invariant biomarkers can be differentiated, which can deepen our understanding of the neurobiology of schizophrenia.

Analysis of functional connectivity using machine learning and deep learning in different data modalities from individuals with schizophrenia

Objective. Schizophrenia(SCZ) is a severe mental disorder associated with persistent or recurrent psychosis, hallucinations, delusions, and thought disorders that affect approximately 26 million people worldwide, according to the World Health Organization. Several studies encompass machine learning (ML) and deep learning algorithms to automate the diagnosis of this mental disorder. Others study SCZ brain networks to get new insights into the dynamics of information processing in individuals suffering from the condition. In this paper, we offer a rigorous approach with ML and deep learning techniques for evaluating connectivity matrices and measures of complex networks to establish an automated diagnosis and comprehend the topology and dynamics of brain networks in SCZ individuals.Approach.For this purpose, we employed an functional magnetic resonance imaging (fMRI) and electroencephalogram (EEG) dataset. In addition, we combined EEG measures, i.e. Hjorth mobility and complexity, with complex network measurements to be analyzed in our model for the first time in the literature.Main results.When comparing the SCZ group to the control group, we found a high positive correlation between the left superior parietal lobe and the left motor cortex and a positive correlation between the left dorsal posterior cingulate cortex and the left primary motor. Regarding complex network measures, the diameter, which corresponds to the longest shortest path length in a network, may be regarded as a biomarker because it is the most crucial measure in different data modalities. Furthermore, the SCZ brain networks exhibit less segregation and a lower distribution of information. As a result, EEG measures outperformed complex networks in capturing the brain alterations associated with SCZ.Significance. Our model achieved an area under receiver operating characteristic curve (AUC) of 100% and an accuracy of 98.5% for the fMRI, an AUC of 95%, and an accuracy of 95.4% for the EEG data set. These are excellent classification results. Furthermore, we investigated the impact of specific brain connections and network measures on these results, which helped us better describe changes in the diseased brain.

Evaluating the clinical utility of speech analysis and machine learning in schizophrenia: A pilot study

BACKGROUND

Schizophrenia is a serious mental disorder that significantly impacts social functioning and quality of life. However, current diagnostic methods lack objective biomarker support. While some studies have indicated differences in audio features between patients with schizophrenia and healthy controls, these findings are influenced by demographic information and variations in experimental paradigms. Therefore, it is crucial to explore stable and reliable audio biomarkers for an auxiliary diagnosis and disease severity prediction of schizophrenia.

METHOD

A total of 130 individuals (65 patients with schizophrenia and 65 healthy controls) read three fixed texts containing positive, neutral, and negative emotions, and recorded them. All audio signals were preprocessed and acoustic features were extracted by a librosa-0.9.2 toolkit. Independent sample t-tests were performed on two sets of acoustic features, and Pearson correlation on the acoustic features and Positive and Negative Syndrome Scale (PANSS) scores of the schizophrenia group. Classification algorithms in scikit-learn were used to diagnose schizophrenia and predict the level of negative symptoms.

RESULTS

Significant differences were observed between the two groups in the mfcc_8, mfcc_11, and mfcc_33 of mel-frequency cepstral coefficient (MFCC). Furthermore, a significant correlation was found between mfcc_7 and the negative PANSS scores. Through acoustic features, we could not only differentiate patients with schizophrenia from healthy controls with an accuracy of 0.815 but also predict the grade of the negative symptoms in schizophrenia with an average accuracy of 0.691.

CONCLUSIONS

The results demonstrated the considerable potential of acoustic characteristics as reliable biomarkers for diagnosing schizophrenia and predicting clinical symptoms.

Dissecting Psychiatric Heterogeneity and Comorbidity with Core Region-Based Machine Learning

Machine learning approaches are increasingly being applied to neuroimaging data from patients with psychiatric disorders to extract brain-based features for diagnosis and prognosis. The goal of this review is to discuss recent practices for evaluating machine learning applications to obsessive-compulsive and related disorders and to advance a novel strategy of building machine learning models based on a set of core brain regions for better performance, interpretability, and generalizability. Specifically, we argue that a core set of co-altered brain regions (namely 'core regions') comprising areas central to the underlying psychopathology enables the efficient construction of a predictive model to identify distinct symptom dimensions/clusters in individual patients. Hypothesis-driven and data-driven approaches are further introduced showing how core regions are identified from the entire brain. We demonstrate a broadly applicable roadmap for leveraging this core set-based strategy to accelerate the pursuit of neuroimaging-based markers for diagnosis and prognosis in a variety of psychiatric disorders.

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.

Adaptive Feature Selection With Augmented Attributes

In many dynamic environment applications, with the evolution of data collection ways, the data attributes are incremental and the samples are stored with accumulated feature spaces gradually. For instance, in the neuroimaging-based diagnosis of neuropsychiatric disorders, with emerging of diverse testing ways, we get more brain image features over time. The accumulation of different types of features will unavoidably bring difficulties in manipulating the high-dimensional data. It is challenging to design an algorithm to select valuable features in this feature incremental scenario. To address this important but rarely studied problem, we propose a novel Adaptive Feature Selection method (AFS). It enables the reusability of the feature selection model trained on previous features and adapts it to fit the feature selection requirements on all features automatically. Besides, an ideal l0-norm sparse constraint for feature selection is imposed with a proposed effective solving strategy. We present the theoretical analyses about the generalization bound and convergence behavior. After tackling this problem in a one-shot case, we extend it to the multi-shot scenario. Plenty of experimental results demonstrate the effectiveness of reusing previous features and the superior of l0-norm constraint in various aspects, together with its effectiveness in discriminating schizophrenic patients from healthy controls.

Incomplete Multi-View Learning Under Label Shift

In image processing, images are usually composed of partial views due to the uncertainty of collection and how to efficiently process these images, which is called incomplete multi-view learning, has attracted widespread attention. The incompleteness and diversity of multi-view data enlarges the difficulty of annotation, resulting in the divergence of label distribution between the training and testing data, named as label shift. However, existing incomplete multi-view methods generally assume that the label distribution is consistent and rarely consider the label shift scenario. To address this new but important challenge, we propose a novel framework termed as Incomplete Multi-view Learning under Label Shift (IMLLS). In this framework, we first give the formal definitions of IMLLS and the bidirectional complete representation which describes the intrinsic and common structure. Then, a multilayer perceptron which combines the reconstruction and classification loss is employed to learn the latent representation, whose existence, consistency and universality are proved with the theoretical satisfaction of label shift assumption. After that, to align the label distribution, the learned representation and trained source classifier are used to estimate the importance weight by designing a new estimation scheme which balances the error generated by finite samples in theory. Finally, the trained classifier reweighted by the estimated weight is fine-tuned to reduce the gap between the source and target representations. Extensive experimental results validate the effectiveness of our algorithm over existing state-of-the-arts methods in various aspects, together with its effectiveness in discriminating schizophrenic patients from healthy controls.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

Neuroimaging mechanisms of acupuncture on functional reorganization for post-stroke motor improvement: a machine learning-based functional magnetic resonance imaging study

Objective

Motor recovery is crucial in stroke rehabilitation, and acupuncture can influence recovery. Neuroimaging and machine learning approaches provide new research directions to explore the brain functional reorganization and acupuncture mechanisms after stroke. We applied machine learning to predict the classification of the minimal clinically important differences (MCID) for motor improvement and identify the neuroimaging features, in order to explore brain functional reorganization and acupuncture mechanisms for motor recovery after stroke.

Methods

In this study, 49 patients with unilateral motor pathway injury (basal ganglia and/or corona radiata) after ischemic stroke were included and evaluated the motor function by Fugl-Meyer Assessment scores (FMA) at baseline and at 2-week follow-up sessions. Patients were divided by the difference between the twice FMA scores into one group showing minimal clinically important difference (MCID group, n = 28) and the other group with no minimal clinically important difference (N-MCID, n = 21). Machine learning was performed by PRoNTo software to predict the classification of the patients and identify the feature brain regions of interest (ROIs). In addition, a matched group of healthy controls (HC, n = 26) was enrolled. Patients and HC underwent magnetic resonance imaging examination in the resting state and in the acupuncture state (acupuncture at the Yanglingquan point on one side) to compare the differences in brain functional connectivity (FC) and acupuncture effects.

Results

Through machine learning, we obtained a balance accuracy rate of 75.51% and eight feature ROIs. Compared to HC, we found that the stroke patients with lower FC between these feature ROIs with other brain regions, while patients in the MCID group exhibited a wider range of lower FC. When acupuncture was applied to Yanglingquan (GB 34), the abnormal FC of patients was decreased, with different targets of effects in different groups.

Conclusion

Feature ROIs identified by machine learning can predict the classification of stroke patients with different motor improvements, and the FC between these ROIs with other brain regions is decreased. Acupuncture can modulate the bilateral cerebral hemispheres to restore abnormal FC via different targets, thereby promoting motor recovery after stroke.

Clinical trial registration

https://www.chictr.org.cn/showproj.html?proj=37359, ChiCTR1900022220.

Hippocampus Parcellation via Discriminative Embedded Clustering of fMRI Functional Connectivity

Dividing a pre-defined brain region into several heterogenous subregions is crucial for understanding its functional segregation and integration. Due to the high dimensionality of brain functional features, clustering is often postponed until dimensionality reduction in traditional parcellation frameworks occurs. However, under such stepwise parcellation, it is very easy to fall into the dilemma of local optimum since dimensionality reduction could not take into account the requirement of clustering. In this study, we developed a new parcellation framework based on the discriminative embedded clustering (DEC), combining subspace learning and clustering in a common procedure with alternative minimization adopted to approach global optimum. We tested the proposed framework in functional connectivity-based parcellation of the hippocampus. The hippocampus was parcellated into three spatial coherent subregions along the anteroventral-posterodorsal axis; the three subregions exhibited distinct functional connectivity changes in taxi drivers relative to non-driver controls. Moreover, compared with traditional stepwise methods, the proposed DEC-based framework demonstrated higher parcellation consistency across different scans within individuals. The study proposed a new brain parcellation framework with joint dimensionality reduction and clustering; the findings might shed new light on the functional plasticity of hippocampal subregions related to long-term navigation experience.

Automated detection of schizophrenia using deep learning: a review for the last decade

Schizophrenia (SZ) is a devastating mental disorder that disrupts higher brain functions like thought, perception, etc., with a profound impact on the individual's life. Deep learning (DL) can detect SZ automatically by learning signal data characteristics hierarchically without the need for feature engineering associated with traditional machine learning. We performed a systematic review of DL models for SZ detection. Various deep models like long short-term memory, convolution neural networks, AlexNet, etc., and composite methods have been published based on electroencephalographic signals, and structural and/or functional magnetic resonance imaging acquired from SZ patients and healthy patients control subjects in diverse public and private datasets. The studies, the study datasets, and model methodologies are reported in detail. In addition, the challenges of DL models for SZ diagnosis and future works are discussed.

Sparse non-convex regularization based explainable DBN in the analysis of brain abnormalities in schizophrenia

Deep belief networks have been widely used in medical image analysis. However, the high-dimensional but small-sample-size characteristic of medical image data makes the model prone to dimensional disaster and overfitting. Meanwhile, the traditional DBN is driven by performance and ignores the explainability which is important for medical image analysis. In this paper, a sparse non-convex based explainable deep belief network is proposed by combining DBN with non-convex sparsity learning. For sparsity, the non-convex regularization and Kullback-Leibler divergence penalty are embedded into DBN to obtain the sparse connection and sparse response representation of the network. It effectively reduces the complexity of the model and improves the generalization ability of the model. Considering explainability, the crucial features for decision-making are selected through the feature back-selection based on the row norm of each layer's weight after network training. We apply the model to schizophrenia data and demonstrate it achieves the best performance among several typical feature selection models. It reveals 28 functional connections highly correlated with schizophrenia, which provides an effective foundation for the treatment and prevention of schizophrenia and methodological assurance for similar brain disorders.

Evaluation of Risk of Bias in Neuroimaging-Based Artificial Intelligence Models for Psychiatric Diagnosis: A Systematic Review

IMPORTANCE

Neuroimaging-based artificial intelligence (AI) diagnostic models have proliferated in psychiatry. However, their clinical applicability and reporting quality (ie, feasibility) for clinical practice have not been systematically evaluated.

OBJECTIVE

To systematically assess the risk of bias (ROB) and reporting quality of neuroimaging-based AI models for psychiatric diagnosis.

EVIDENCE REVIEW

PubMed was searched for peer-reviewed, full-length articles published between January 1, 1990, and March 16, 2022. Studies aimed at developing or validating neuroimaging-based AI models for clinical diagnosis of psychiatric disorders were included. Reference lists were further searched for suitable original studies. Data extraction followed the CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines. A closed-loop cross-sequential design was used for quality control. The PROBAST (Prediction Model Risk of Bias Assessment Tool) and modified CLEAR (Checklist for Evaluation of Image-Based Artificial Intelligence Reports) benchmarks were used to systematically evaluate ROB and reporting quality.

FINDINGS

A total of 517 studies presenting 555 AI models were included and evaluated. Of these models, 461 (83.1%; 95% CI, 80.0%-86.2%) were rated as having a high overall ROB based on the PROBAST. The ROB was particular high in the analysis domain, including inadequate sample size (398 of 555 models [71.7%; 95% CI, 68.0%-75.6%]), poor model performance examination (with 100% of models lacking calibration examination), and lack of handling data complexity (550 of 555 models [99.1%; 95% CI, 98.3%-99.9%]). None of the AI models was perceived to be applicable to clinical practices. Overall reporting completeness (ie, number of reported items/number of total items) for the AI models was 61.2% (95% CI, 60.6%-61.8%), and the completeness was poorest for the technical assessment domain with 39.9% (95% CI, 38.8%-41.1%).

CONCLUSIONS AND RELEVANCE

This systematic review found that the clinical applicability and feasibility of neuroimaging-based AI models for psychiatric diagnosis were challenged by a high ROB and poor reporting quality. Particularly in the analysis domain, ROB in AI diagnostic models should be addressed before clinical application.

Computer-aided diagnosis of schizophrenia based on node2vec and Transformer

OBJECTIVE

Compared with the healthy control (HC) group, the brain structure and function of schizophrenia (SZ) patients are significantly abnormal, so brain imaging methods can be used to achieve the aided diagnosis of SZ. However, a brain network based on brain imaging data is non-Euclidean, and its intrinsic features cannot be learned effectively by general deep learning models. Furthermore, in the majority of existing studies, brain network features were manually specified as the input of machine learning models.

METHODS

In this study, brain functional network constructed from the subject's fMRI data is analyzed, and its small-world value is calculated and t-tested; the node2vec algorithm in graph embedding is introduced to transform the constructed brain network into low-dimensional dense vectors, and the brain network's non-Euclidean spatial structure characteristics are retained to the greatest extent, so that its intrinsic features can be extracted by deep learning models; GridMask is used to randomly mask part of the information in the vectors to enhance the data; and then features can be extracted using the Transformer model to identify SZ.

RESULTS

It is again shown that the small-world value of the brain network in SZ is significantly lower than that in HC by t-test (p=0.014¡0.05). 97.78% classification accuracy is achieved by the proposed methods (node2vec + GridMask + Transformer) in 30 SZ patients and 30 healthy people.

CONCLUSION

The experiment shows that the node2vec used in this paper can effectively solve the problem of brain network features being difficult to learn by general deep learning models. The high-precision computer-aided diagnosis of SZ can be obtained by combining node2vec with Transformer and GridMask.

SIGNIFICANCE

The proposed methods in the paper are expected to be used for aided diagnosis of SZ.

Optimized adaptive neuro-fuzzy inference system based on hybrid grey wolf-bat algorithm for schizophrenia recognition from EEG signals

Schizophrenia is a chronic mental disorder that impairs a person's thinking capacity, feelings and emotions, behavioural traits, etc., Emotional distortions, delusions, hallucinations, and incoherent speech are all some of the symptoms of schizophrenia, and cause disruption of routine activities. Computer-assisted diagnosis of schizophrenia is significantly needed to give its patients a higher quality of life. Hence, an improved adaptive neuro-fuzzy inference system based on the Hybrid Grey Wolf-Bat Algorithm for accurate prediction of schizophrenia from multi-channel EEG signals is presented in this study. The EEG signals are pre-processed using a Butterworth band pass filter and wICA initially, from which statistical, time-domain, frequency-domain, and spectral features are extracted. Discriminating features are selected using the ReliefF algorithm and are then forwarded to ANFIS for classification into either schizophrenic or normal. ANFIS is optimized by the Hybrid Grey Wolf-Bat Algorithm (HWBO) for better efficiency. The method is experimented on two separate EEG datasets-1 and 2, demonstrating an accuracy of 99.54% and 99.35%, respectively, with appreciable F1-score and MCC. Further experiments reveal the efficiency of the Hybrid Wolf-Bat algorithm in optimizing the ANFIS parameters when compared with traditional ANFIS model and other proven algorithms like genetic algorithm-ANFIS, particle optimization-ANFIS, crow search optimization algorithm-ANFIS and ant colony optimization algorithm-ANFIS, showing high R2 value and low RSME value. To provide a bias free classification, tenfold cross validation is performed which produced an accuracy of 97.8% and 98.5% on the two datasets respectively. Experimental outcomes demonstrate the superiority of the Hybrid Grey Wolf-Bat Algorithm over the similar techniques in predicting schizophrenia.

Graph-Based Analysis of Brain Connectivity in Multiple Sclerosis Using Functional MRI: A Systematic Review

(1) Background: Multiple sclerosis (MS) is an immune system disease in which myelin in the nervous system is affected. This abnormal immune system mechanism causes physical disabilities and cognitive impairment. Functional magnetic resonance imaging (fMRI) is a common neuroimaging technique used in studying MS. Computational methods have recently been applied for disease detection, notably graph theory, which helps researchers understand the entire brain network and functional connectivity. (2) Methods: Relevant databases were searched to identify articles published since 2000 that applied graph theory to study functional brain connectivity in patients with MS based on fMRI. (3) Results: A total of 24 articles were included in the review. In recent years, the application of graph theory in the MS field received increased attention from computational scientists. The graph-theoretical approach was frequently combined with fMRI in studies of functional brain connectivity in MS. Lower EDSSs of MS stage were the criteria for most of the studies (4) Conclusions: This review provides insights into the role of graph theory as a computational method for studying functional brain connectivity in MS. Graph theory is useful in the detection and prediction of MS and can play a significant role in identifying cognitive impairment associated with MS.

A Review of Machine Learning and Deep Learning Approaches on Mental Health Diagnosis

Combating mental illnesses such as depression and anxiety has become a global concern. As a result of the necessity for finding effective ways to battle these problems, machine learning approaches have been included in healthcare systems for the diagnosis and probable prediction of the treatment outcomes of mental health conditions. With the growing interest in machine and deep learning methods, analysis of existing work to guide future research directions is necessary. In this study, 33 articles on the diagnosis of schizophrenia, depression, anxiety, bipolar disorder, post-traumatic stress disorder (PTSD), anorexia nervosa, and attention deficit hyperactivity disorder (ADHD) were retrieved from various search databases using the preferred reporting items for systematic reviews and meta-analysis (PRISMA) review methodology. These publications were chosen based on their use of machine learning and deep learning technologies, individually assessed, and their recommended methodologies were then classified into the various disorders included in this study. In addition, the difficulties encountered by the researchers are discussed, and a list of some public datasets is provided.

Psychotic Symptom, Mood, and Cognition-associated Multimodal MRI Reveal Shared Links to the Salience Network Within the Psychosis Spectrum Disorders

Schizophrenia (SZ), schizoaffective disorder (SAD), and psychotic bipolar disorder share substantial overlap in clinical phenotypes, associated brain abnormalities and risk genes, making reliable diagnosis among the three illness challenging, especially in the absence of distinguishing biomarkers. This investigation aims to identify multimodal brain networks related to psychotic symptom, mood, and cognition through reference-guided fusion to discriminate among SZ, SAD, and BP. Psychotic symptom, mood, and cognition were used as references to supervise functional and structural magnetic resonance imaging (MRI) fusion to identify multimodal brain networks for SZ, SAD, and BP individually. These features were then used to assess the ability in discriminating among SZ, SAD, and BP. We observed shared links to functional and structural covariation in prefrontal, medial temporal, anterior cingulate, and insular cortices among SZ, SAD, and BP, although they were linked with different clinical domains. The salience (SAN), default mode (DMN), and fronto-limbic (FLN) networks were the three identified multimodal MRI features within the psychosis spectrum disorders from psychotic symptom, mood, and cognition associations. In addition, using these networks, we can classify patients and controls and distinguish among SZ, SAD, and BP, including their first-degree relatives. The identified multimodal SAN may be informative regarding neural mechanisms of comorbidity for psychosis spectrum disorders, along with DMN and FLN may serve as potential biomarkers in discriminating among SZ, SAD, and BP, which may help investigators better understand the underlying mechanisms of psychotic comorbidity from three different disorders via a multimodal neuroimaging perspective.

Confound-leakage: confound removal in machine learning leads to leakage

BACKGROUND

Machine learning (ML) approaches are a crucial component of modern data analysis in many fields, including epidemiology and medicine. Nonlinear ML methods often achieve accurate predictions, for instance, in personalized medicine, as they are capable of modeling complex relationships between features and the target. Problematically, ML models and their predictions can be biased by confounding information present in the features. To remove this spurious signal, researchers often employ featurewise linear confound regression (CR). While this is considered a standard approach for dealing with confounding, possible pitfalls of using CR in ML pipelines are not fully understood.

RESULTS

We provide new evidence that, contrary to general expectations, linear confound regression can increase the risk of confounding when combined with nonlinear ML approaches. Using a simple framework that uses the target as a confound, we show that information leaked via CR can increase null or moderate effects to near-perfect prediction. By shuffling the features, we provide evidence that this increase is indeed due to confound-leakage and not due to revealing of information. We then demonstrate the danger of confound-leakage in a real-world clinical application where the accuracy of predicting attention-deficit/hyperactivity disorder is overestimated using speech-derived features when using depression as a confound.

CONCLUSIONS

Mishandling or even amplifying confounding effects when building ML models due to confound-leakage, as shown, can lead to untrustworthy, biased, and unfair predictions. Our expose of the confound-leakage pitfall and provided guidelines for dealing with it can help create more robust and trustworthy ML models.

Neuroimaging biomarkers for detecting schizophrenia: A resting-state functional MRI-based radiomics analysis

Schizophrenia (SZ) is a common psychiatric disorder that is difficult to accurately diagnose in clinical practice. Quantifiable biomarkers are urgently required to explore the potential physiological mechanism of SZ and improve its diagnostic accuracy. Thus, this study aimed to identify biomarkers that classify SZ patients and healthy control subjects and investigate the potential neural mechanisms of SZ using degree centrality (DC)- and voxel-mirrored homotopic connectivity (VMHC)-based radiomics. Radiomics features were extracted from DC and VMHC metrics generated via resting-state functional magnetic resonance imaging, and significant features were selected and dimensionality was reduced using t-tests and least absolute shrinkage and selection operator. Subsequently, we built our model using a support vector machine classifier. We observed that our method obtained great classification performance (area under the curve, 0.808; accuracy, 74.02%), and it could be generalized to different brain atlases. The regions that we identified as discriminative features mainly included bilateral dorsal caudate and front-parietal, somatomotor, limbic, and default mode networks. Our findings showed that the radiomics-based machine learning method could facilitate us to understand the potential pathological mechanism of SZ more comprehensively and contribute to the accurate diagnosis of patients with SZ.

Harnessing the Power of Advanced Fetal Neuroimaging to Understand In Utero Footprints for Later Neuropsychiatric Disorders

Adverse intrauterine events may profoundly impact fetal risk for future adult diseases. The mechanisms underlying this increased vulnerability are complex and remain poorly understood. Contemporary advances in fetal magnetic resonance imaging (MRI) have provided clinicians and scientists with unprecedented access to in vivo human fetal brain development to begin to identify emerging endophenotypes of neuropsychiatric disorders such as autism spectrum disorder, attention-deficit/hyperactivity disorder, and schizophrenia. In this review, we discuss salient findings of normal fetal neurodevelopment from studies using advanced, multimodal MRI that have provided unparalleled characterization of in utero prenatal brain morphology, metabolism, microstructure, and functional connectivity. We appraise the clinical utility of these normative data in identifying high-risk fetuses before birth. We highlight available studies that have investigated the predictive validity of advanced prenatal brain MRI findings and long-term neurodevelopmental outcomes. We then discuss how ex utero quantitative MRI findings can inform in utero investigations toward the pursuit of early biomarkers of risk. Lastly, we explore future opportunities to advance our understanding of the prenatal origins of neuropsychiatric disorders using precision fetal imaging.

Machine Learning and Non-Affective Psychosis: Identification, Differential Diagnosis, and Treatment

PURPOSE OF REVIEW

This review will cover the most relevant findings on the use of machine learning (ML) techniques in the field of non-affective psychosis, by summarizing the studies published in the last three years focusing on illness detection and treatment.

RECENT FINDINGS

Multiple ML tools that include mostly supervised approaches such as support vector machine, gradient boosting, and random forest showed promising results by applying these algorithms to various sources of data: socio-demographic information, EEG, language, digital content, blood biomarkers, neuroimaging, and electronic health records. However, the overall performance, in the binary classification case, varied from 0.49, which is to be considered very low (i.e., noise), to over 0.90. These results are fully justified by different factors, some of which may be attributable to the preprocessing of the data, the wide variety of the data, and the a-priori setting of hyperparameters. One of the main limitations of the field is the lack of stratification of results based on biological sex, given that psychosis presents differently in men and women; hence, the necessity to tailor identification tools and data analytic strategies. Timely identification and appropriate treatment are key factors in reducing the consequences of psychotic disorders. In recent years, the emergence of new analytical tools based on artificial intelligence such as supervised ML approaches showed promises as a potential breakthrough in this field. However, ML applications in everyday practice are still in its infancy.

Deep learning in neuroimaging data analysis: Applications, challenges, and solutions

Methods for the analysis of neuroimaging data have advanced significantly since the beginning of neuroscience as a scientific discipline. Today, sophisticated statistical procedures allow us to examine complex multivariate patterns, however most of them are still constrained by assuming inherent linearity of neural processes. Here, we discuss a group of machine learning methods, called deep learning, which have drawn much attention in and outside the field of neuroscience in recent years and hold the potential to surpass the mentioned limitations. Firstly, we describe and explain the essential concepts in deep learning: the structure and the computational operations that allow deep models to learn. After that, we move to the most common applications of deep learning in neuroimaging data analysis: prediction of outcome, interpretation of internal representations, generation of synthetic data and segmentation. In the next section we present issues that deep learning poses, which concerns multidimensionality and multimodality of data, overfitting and computational cost, and propose possible solutions. Lastly, we discuss the current reach of DL usage in all the common applications in neuroimaging data analysis, where we consider the promise of multimodality, capability of processing raw data, and advanced visualization strategies. We identify research gaps, such as focusing on a limited number of criterion variables and the lack of a well-defined strategy for choosing architecture and hyperparameters. Furthermore, we talk about the possibility of conducting research with constructs that have been ignored so far or/and moving toward frameworks, such as RDoC, the potential of transfer learning and generation of synthetic data.

Automatic autism spectrum disorder detection using artificial intelligence methods with MRI neuroimaging: A review

Autism spectrum disorder (ASD) is a brain condition characterized by diverse signs and symptoms that appear in early childhood. ASD is also associated with communication deficits and repetitive behavior in affected individuals. Various ASD detection methods have been developed, including neuroimaging modalities and psychological tests. Among these methods, magnetic resonance imaging (MRI) imaging modalities are of paramount importance to physicians. Clinicians rely on MRI modalities to diagnose ASD accurately. The MRI modalities are non-invasive methods that include functional (fMRI) and structural (sMRI) neuroimaging methods. However, diagnosing ASD with fMRI and sMRI for specialists is often laborious and time-consuming; therefore, several computer-aided design systems (CADS) based on artificial intelligence (AI) have been developed to assist specialist physicians. Conventional machine learning (ML) and deep learning (DL) are the most popular schemes of AI used for diagnosing ASD. This study aims to review the automated detection of ASD using AI. We review several CADS that have been developed using ML techniques for the automated diagnosis of ASD using MRI modalities. There has been very limited work on the use of DL techniques to develop automated diagnostic models for ASD. A summary of the studies developed using DL is provided in the Supplementary Appendix. Then, the challenges encountered during the automated diagnosis of ASD using MRI and AI techniques are described in detail. Additionally, a graphical comparison of studies using ML and DL to diagnose ASD automatically is discussed. We suggest future approaches to detecting ASDs using AI techniques and MRI neuroimaging.

Morphometric Integrated Classification Index: A Multisite Model-Based, Interpretable, Shareable and Evolvable Biomarker for Schizophrenia

BACKGROUND AND HYPOTHESIS

Multisite massive schizophrenia neuroimaging data sharing is becoming critical in understanding the pathophysiological mechanism and making an objective diagnosis of schizophrenia; it remains challenging to obtain a generalizable and interpretable, shareable, and evolvable neuroimaging biomarker for schizophrenia diagnosis.

STUDY DESIGN

A Morphometric Integrated Classification Index (MICI) was proposed as a potential biomarker for schizophrenia diagnosis based on structural magnetic resonance imaging data of 1270 subjects from 10 sites (588 schizophrenia patients and 682 normal controls). An optimal XGBoost classifier plus sample-weighted SHapley Additive explanation algorithms were used to construct the MICI measure.

STUDY RESULTS

The MICI measure achieved comparable performance with the sample-weighted ensembling model and merged model based on raw data (Delong test, P > 0.82) while outperformed the single-site models (Delong test, P < 0.05) in either the independent-sample testing datasets from the 9 sites or the independent-site dataset (generalizable). Besides, when new sites were embedded in, the performance of this measure was gradually increasing (evolvable). Finally, MICI was strongly associated with the severity of schizophrenia brain structural abnormality, with the patients' positive and negative symptoms, and with the brain expression profiles of schizophrenia risk genes (interpretable).

CONCLUSIONS

In summary, the proposed MICI biomarker may provide a simple and explainable way to support clinicians for objectively diagnosing schizophrenia. Finally, we developed an online model share platform to promote biomarker generalization and provide free individual prediction services (http://micc.tmu.edu.cn/mici/index.html).

Convolutional Recurrent Neural Network for Dynamic Functional MRI Analysis and Brain Disease Identification

Dynamic functional connectivity (dFC) networks derived from resting-state functional magnetic resonance imaging (rs-fMRI) help us understand fundamental dynamic characteristics of human brains, thereby providing an efficient solution for automated identification of brain diseases, such as Alzheimer's disease (AD) and its prodromal stage. Existing studies have applied deep learning methods to dFC network analysis and achieved good performance compared with traditional machine learning methods. However, they seldom take advantage of sequential information conveyed in dFC networks that could be informative to improve the diagnosis performance. In this paper, we propose a convolutional recurrent neural network (CRNN) for automated brain disease classification with rs-fMRI data. Specifically, we first construct dFC networks from rs-fMRI data using a sliding window strategy. Then, we employ three convolutional layers and long short-term memory (LSTM) layer to extract high-level features of dFC networks and also preserve the sequential information of extracted features, followed by three fully connected layers for brain disease classification. Experimental results on 174 subjects with 563 rs-fMRI scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI) demonstrate the effectiveness of our proposed method in binary and multi-category classification tasks.

Improved Multiclassification of Schizophrenia Based on Xgboost and Information Fusion for Small Datasets

To improve the performance in multiclass classification for small datasets, a new approach for schizophrenic classification is proposed in the present study. Firstly, the Xgboost classifier is introduced to discriminate the two subtypes of schizophrenia from health controls by analyzing the functional magnetic resonance imaging (fMRI) data, while the gray matter volume (GMV) and amplitude of low-frequency fluctuations (ALFF) are extracted as the features of classifiers. Then, the D-S combination rule of evidence is used to achieve fusion to determine the basic probability assignment based on the output of different classifiers. Finally, the algorithm is applied to classify 38 healthy controls, 16 deficit schizophrenic patients, and 31 nondeficit schizophrenic patients. 10-folds cross-validation method is used to assess classification performance. The results show the proposed algorithm with a sensitivity of 73.89%, which is higher than other classification algorithms, such as supported vector machine (SVM), logistic regression (LR), K-nearest neighbor (KNN) algorithm, random forest (RF), BP neural network (NN), classification and regression tree (CART), naive Bayes classifier (NB), extreme gradient boosting (Xgboost), and deep neural network (DNN). The accuracy of the fusion algorithm is higher than that of classifier based on the GMV or ALFF in the small datasets. The accuracy rate of the improved multiclassification method based on Xgboost and fusion algorithm is higher than that of other machine learning methods, which can further assist the diagnosis of clinical schizophrenia.

Impaired time-distance reconfiguration patterns in Alzheimer's disease: a dynamic functional connectivity study with 809 individuals from 7 sites

Background

The dynamic functional connectivity (dFC) has been used successfully to investigate the dysfunction of Alzheimer's disease (AD) patients. The reconfiguration intensity of nodal dFC, which means the degree of alteration between FCs at different time scales, could provide additional information for understanding the reconfiguration of brain connectivity.

Results

In this paper, we introduced a feature named time distance nodal connectivity diversity (tdNCD), and then evaluated the network reconfiguration intensity in every specific brain region in AD using a large multicenter dataset (N = 809 from 7 independent sites). Our results showed that the dysfunction involved in three subnetworks in AD, including the default mode network (DMN), the subcortical network (SCN), and the cerebellum network (CBN). The nodal tdNCD inside the DMN increased in AD compared to normal controls, and the nodal dynamic FC of the SCN and the CBN decreased in AD. Additionally, the classification analysis showed that the classification performance was better when combined tdNCD and FC to classify AD from normal control (ACC = 81%, SEN = 83.4%, SPE = 80.6%, and F1-score = 79.4%) than that only using FC (ACC = 78.2%, SEN = 76.2%, SPE = 76.5%, and F1-score = 77.5%) with a leave-one-site-out cross-validation. Besides, the performance of the three classes classification was improved from 50% (only using FC) to 53.3% (combined FC and tdNCD) (macro F1-score accuracy from 46.8 to 48.9%). More importantly, the classification model showed significant clinically predictive correlations (two classes classification: R = −0.38, P < 0.001; three classes classification: R = −0.404, P < 0.001). More importantly, several commonly used machine learning models confirmed that the tdNCD would provide additional information for classifying AD from normal controls.

Conclusions

The present study demonstrated dynamic reconfiguration of nodal FC abnormities in AD. The tdNCD highlights the potential for further understanding core mechanisms of brain dysfunction in AD. Evaluating the tdNCD FC provides a promising way to understand AD processes better and investigate novel diagnostic brain imaging biomarkers for AD.

Robust Brain Age Estimation based on sMRI via Nonlinear Age-Adaptive Ensemble Learning

Precise prediction on brain age is urgently needed by many biomedical areas including mental rehabilitation prognosis as well as various medicine or treatment trials. People began to realize that contrasting physical (real) age and predicted brain age can help to highlight brain issues and evaluate if patients' brains are healthy or not. Such age prediction is often challenging for single model-based prediction, while the conditions of brains vary drastically over age. In this work, we present an age-adaptive ensemble model that is based on the combination of four different machine learning algorithms, including a support vector machine (SVR), a convolutional neural network (CNN) model, and the popular GoogLeNet and ResNet deep networks. The ensemble model proposed here is nonlinearly adaptive, where age is taken as a key factor in the nonlinear combination of various single-algorithm-based independent models. In our age-adaptive ensemble method, the weights of each model are learned automatically as nonlinear functions over age instead of fixed values, while brain age estimation is based on such an age-adaptive integration of various single models. The quality of the model is quantified by the mean absolute errors (MAE) and spearman correlation between the predicted age and the actual age, with the least MAE and the highest Spearman correlation representing the highest accuracy in age prediction. By testing on the Predictive Analysis Challenge 2019 (PAC 2019) dataset, our novel ensemble model has achieved a MAE down to 3.19, which is a significantly increased accuracy in this brain age competition. If deployed in the real world, our novel ensemble model having an improved accuracy could potentially help doctors to identify the risk of brain diseases more accurately and quickly, thus helping pharmaceutical companies develop drugs or treatments precisely, and potential offer a new powerful tool for researchers in the field of brain science.