Sleep disturbance-related neuroimaging features as potential biomarkers for the diagnosis of major depressive disorder: A multicenter study based on machine learning

Identification of multimodal brain imaging biomarkers in first-episode drugs-naive major depressive disorder through a multi-site large-scale MRI consortium data

BACKGROUND

Major depressive disorder (MDD) is a severe and common mental illness. The first-episode drugs-naive MDD (FEDN-MDD) patients, who have not undergone medication intervention, contribute to understanding the biological basis of MDD. Multimodal Magnetic Resonance Imaging can provide a comprehensive understanding of brain functional and structural abnormalities in MDD. However, most MDD studies use single-modal, small-scale MRI data. And several multimodal studies of MDD are limited to simple linear combinations of functional and structural features.

METHODS

We screened a large sample of FEDN-MDD patients and healthy controlsmultimodal MRI data. Extracting the fractional amplitude of low-frequency fluctuations (fALFF) feature from functional magnetic resonance imaging and the gray matter volume (GMV) feature from structural magnetic resonance imaging. The mCCA-jICA method was used to integrate these two modal features to investigate the functional-structural co-variation abnormalities in MDD. To validate the stability of the extracted functional-structural covariant abnormalities features, we apply them to identify FEDN-MDD patients.

RESULTS

The results show that compared to healthy controls, FEDN-MDD patients exhibit joint group-discriminative independent component and modality-specific group-discriminative independent component, suggesting functional-structural covariant abnormalities in MDD patients. Using lightGBM classifier, we achieve a classification accuracy of 99.84 %.

LIMITATION

We use GMV and fALFF for multimodal fusion shows promise, but requires further validation with other datasets and exploration of additional multimodal features.

CONCLUSIONS

This may indicate that multimodal fusion features can effectively explore information between different modalities and can accurately identify FEDN-MDD patients, suggesting their potential as multimodal brain imaging biomarkers for MDD.

Effectiveness of artificial intelligence in detecting and managing depressive disorders: Systematic review

OBJECTIVES

This study underscores the importance of exploring AI's creative applications in treating depressive disorders to revolutionize mental health care. Through innovative integration of AI technologies, the research confirms their positive effects on preventing, diagnosing, and treating depression. The systematic review establishes an evidence base for AI in depression management, offering directions for effective interventions.

METHODS

This systematic literature review investigates the effectiveness of AI in depression management by analyzing studies from January 1, 2017, to May 31, 2022. Utilizing search engines like IEEE Xplore, PubMed, and Web of Science, the review focused on keywords such as Depression/Mental Health, Machine Learning/Artificial Intelligence, and Prediction/Diagnosis. The analysis of 95 documents involved classification based on use, data type, and algorithm type.

RESULTS

The study revealed that AI in depression management excelled in accuracy, particularly in monitoring and prediction. Biomarker-derived data demonstrated the highest accuracy, with the CNN algorithm proving most effective. The findings affirm the therapeutic benefits of AI, including treatment, detection, and disease prediction, highlighting its potential in analyzing monitored data for depression management.

LIMITATIONS

This study exclusively examined the application of AI in individuals with depressive disorders. Interpretation should be cautious due to the limited scope of subjects to this specific population.

CONCLUSIONS

To introduce digital healthcare and therapies for ongoing depression management, it's crucial to present empirical evidence on the medical fee payment system, safety, and efficacy. These findings support enhanced medical accessibility through digital healthcare, offering personalized disease management for patients seeking non-face-to-face treatment.

Plasma alpha-trypsin inhibitor heavy chain 4 as an age-specific biomarker in the diagnosis and treatment of major depressive disorder

Background

The diagnosis of major depressive disorder (MDD) mainly depends on subjective clinical symptoms, without an acceptable objective biomarker for the clinical application of MDD. Inter-alpha-trypsin inhibitor heavy chain 4 (ITIH4) showed a high specificity as biomarker for the diagnosis and treatment of MDD. The present study aimed to investigate differences in plasma ITIH4 in two different aged MDD patients and underlying pathological mechanisms of plasma ITIH4 in the occurrence and development of MDD.

Methods

Sixty-five adult MDD patients, 51 adolescent MDD patients, and 64 healthy controls (HCs) were included in the present study. A 14-days' antidepressive treatment was conducted in all MDD patients. Psychological assessments were performed and plasma ITIH4 and astrocyte-related markers were detected for all participants.

Results

(1) Plasma levels of ITIH4 in adult MDD patients were significantly higher than adolescent MDD patients and HCs, and significantly increased plasma ITIH4 levels was observed in adolescent MDD patients compared with HCs (2). There were positive correlations between plasma ITIH4 levels and 24-item Hamilton Depression Scale (HAMD-24) scores and plasma glial fibrillary acidic protein (GFAP) levels in MDD patients, however, plasma ITIH4 levels were significantly correlated with age just in adult MDD patients (3). Plasma ITIH4 showed area under the curve values of 0.824 and 0.729 to differentiate adult MDD patients and adolescent MDD patients from HCs, respectively (4). There was significant decrease in plasma levels of ITIH4 between before and after antidepressive treatment in adult MDD patients, but not in adolescent MDD patients (5). Changed value of ITIH4 levels were correlated with the changed value of GFAP levels and changed rate of HAMD-24 scores in adult MDD patients following antidepressive treatment.

Conclusion

Plasma ITIH4 may be potential plasma biomarkers of MDD with age-related specificity, which was associated with depressive symptoms astrocyte-related pathologic changes, and antidepressive treatment efficacy.

Insomnia-related brain functional correlates in first-episode drug-naïve major depressive disorder revealed by resting-state fMRI

Introduction

Insomnia is a common comorbidity symptom in major depressive disorder (MDD) patients. Abnormal brain activities have been observed in both MDD and insomnia patients, however, the central pathological mechanisms underlying the co-occurrence of insomnia in MDD patients are still unclear. This study aimed to explore the differences of spontaneous brain activity between MDD patients with and without insomnia, as well as patients with different level of insomnia.

Methods

A total of 88 first-episode drug-naïve MDD patients including 44 with insomnia (22 with high insomnia and 22 with low insomnia) and 44 without insomnia, as well as 44 healthy controls (HC), were enrolled in this study. The level of depression and insomnia were evaluated by HAMD-17, adjusted HAMD-17 and its sleep disturbance subscale in all subjects. Resting-state functional and structural magnetic resonance imaging data were acquired from all participants and then were preprocessed by the software of DPASF. Regional homogeneity (ReHo) values of brain regions were calculated by the software of REST and were compared. Finally, receiver operating characteristic (ROC) curves were conducted to determine the values of abnormal brain regions for identifying MDD patients with insomnia and evaluating the severity of insomnia.

Results

Analysis of variance showed that there were significant differences in ReHo values in the left middle frontal gyrus, left pallidum, right superior frontal gyrus, right medial superior frontal gyrus and right rectus gyrus among three groups. Compared with HC, MDD patients with insomnia showed increased ReHo values in the medial superior frontal gyrus, middle frontal gyrus, triangular inferior frontal gyrus, calcarine fissure and right medial superior frontal gyrus, medial orbital superior frontal gyrus, as well as decreased ReHo values in the left middle occipital gyrus, pallidum and right superior temporal gyrus, inferior temporal gyrus, middle cingulate gyrus, hippocampus, putamen. MDD patients without insomnia demonstrated increased ReHo values in the left middle frontal gyrus, orbital middle frontal gyrus, anterior cingulate gyrus and right triangular inferior frontal gyrus, as well as decreased ReHo values in the left rectus gyrus, postcentral gyrus and right rectus gyrus, fusiform gyrus, pallidum. In addition, MDD patients with insomnia had decreased ReHo values in the left insula when compared to those without insomnia. Moreover, MDD patients with high insomnia exhibited increased ReHo values in the right middle temporal gyrus, and decreased ReHo values in the left orbital superior frontal gyrus, lingual gyrus, right inferior parietal gyrus and postcentral gyrus compared to those with low insomnia. ROC analysis demonstrated that impaired brain region might be helpful for identifying MDD patients with insomnia and evaluating the severity of insomnia.

Conclusion

These findings suggested that MDD patients with insomnia had wider abnormalities of brain activities in the prefrontal-limbic circuits including increased activities in the prefrontal cortex, which might be the compensatory mechanism underlying insomnia in MDD. In addition, decreased activity of left insula might be associated with the occurrence of insomnia in MDD patients and decreased activities of the frontal-parietal network might cause more serious insomnia related to MDD.

Macrophage Migration Inhibitory Factor as a Potential Plasma Biomarker of Cognitive Impairment in Cerebral Small Vessel Disease

As the pathogenesis of cerebral small vessel disease with cognitive impairment (CSVD-CI) remains unclear, identifying effective biomarkers can contribute to the clinical management of CSVD-CI. This study recruited 54 healthy controls (HCs), 60 CSVD-CI patients, and 57 CSVD cognitively normal (CSVD-CN) patients. All participants underwent neuropsychological assessments and multimodal magnetic resonance imaging. Macrophage migration inhibitory factors (MIFs) were assessed in plasma. The least absolute shrinkage and selection operator model was used to determine a composite marker. Compared with HCs or CSVD-CN patients, CSVD-CI patients had significantly increased plasma MIF levels. In CSVD-CI patients, plasma MIF levels were significantly correlated with multiple cognitive assessment scores, plasma levels of blood-brain barrier (BBB)-related indices, white matter hyperintensity Fazekas scores, and the mean amplitude of low-frequency fluctuation in the right superior temporal gyrus. Higher plasma MIF levels were significantly associated with worse global cognition and information processing speed in CSVD-CI patients. The composite marker (including plasma MIF) distinguished CSVD-CI patients from CSVD-CN and HCs with >80% accuracy. Meta-analysis indicated that blood MIF levels were significantly increased in CSVD-CI patients. In conclusion, plasma MIF is a potential biomarker for early identification of CSVD-CI. Plasma MIF may play a role in cognitive decline in CSVD through BBB dysfunction and changes in white matter hyperintensity and brain activity.

Identifying major depressive disorder with associated sleep disturbances through fMRI regional homogeneity at rest

BACKGROUND

Anomalies in regional homogeneity (ReHo) have been documented in patients with major depressive disorder (MDD) and sleep disturbances (SDs). This investigation aimed to scrutinize changes in ReHo in MDD patients with comorbid SD, and to devise potential diagnostic biomarkers for detecting sleep-related conditions in patients with MDD.

METHODS

Patients with MDD and healthy controls underwent resting-state functional magnetic resonance imaging scans. SD severity was quantified using the 17-item Hamilton Rating Scale for Depression. Subsequent to the acquisition of imaging data, ReHo analysis was performed, and a support vector machine (SVM) method was employed to assess the utility of ReHo in discriminating MDD patients with SD.

RESULTS

Compared with MDD patients without SD, MDD patients with SD exhibited increased ReHo values in the right posterior cingulate cortex (PCC)/precuneus, right median cingulate cortex, left postcentral gyrus (postCG), and right inferior temporal gyrus (ITG). Furthermore, the ReHo values in the right PCC/precuneus and ITG displayed a positive correlation with clinical symptoms across all patients. SVM classification results showed that a combination of abnormal ReHo in the left postCG and right ITG achieved an overall accuracy of 84.21%, a sensitivity of 81.82%, and a specificity of 87.50% in identifying MDD patients with SD from those without SD.

CONCLUSION

We identified disrupted ReHo patterns in MDD patients with SD, and presented a prospective neuroimaging-based diagnostic biomarker for these patients.

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.

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.

Shared and differential fractional amplitude of low-frequency fluctuation patterns at rest in major depressive disorders with or without sleep disturbance

Objective

Sleep disturbances (SD) are commonly found in patients with major depressive disorder (MDD). This study aims to explore the influence of SD symptoms on clinical characteristics in patients with MDD and to investigate the shared and distinct fractional amplitude of low-frequency fluctuation (fALFF) patterns in these patients with or without SD symptoms.

Methods

Twenty-four MDD patients with SD symptoms (Pa_s), 33 MDD patients without SD symptoms (Pa_ns) and 32 healthy controls (HCs) were included in this study. The fALFF and correlation analyses were applied to analyze the features of imaging and clinical data.

Results

Pa_s showed more severe anxiety and depression than Pa_ns. Compared with Pa_ns, Pa_s exhibited increased fALFF value in the left precuneus. Patients shared abnormal fALFF in the frontal-occipital brain regions. There was a positive correlation between fALFF values of the left precuneus and sleep disturbance scores (r = 0.607, p = 0.0000056734) in all patients in addition to a negative correlation between fALFF values of the left MOG/cuneus and HAMD-17 total scores (r = -0.595, p = 0.002141) in Pa_s. The receiver operating characteristic (ROC) results of the fALFF could be used to discriminate Pa_s from Pa_ns with a specificity of 72.73% and a sensitivity of 70.83%.

Conclusion

Pa_s displayed more serious anxiety and depression symptoms. Patients shared abnormal fALFF in the frontal-occipital brain regions, which may be a common characteristic for MDD. And increased fALFF value in the left precuneus might be a specific neuroimaging feature of MDD patients with SD symptoms.

Potential of brain age in identifying early cognitive impairment in subcortical small-vessel disease patients

Background

Reliable and individualized biomarkers are crucial for identifying early cognitive impairment in subcortical small-vessel disease (SSVD) patients. Personalized brain age prediction can effectively reflect cognitive impairment. Thus, the present study aimed to investigate the association of brain age with cognitive function in SSVD patients and assess the potential value of brain age in clinical assessment of SSVD.

Materials and methods

A prediction model for brain age using the relevance vector regression algorithm was developed using 35 healthy controls. Subsequently, the prediction model was tested using 51 SSVD patients [24 subjective cognitive impairment (SCI) patients and 27 mild cognitive impairment (MCI) patients] to identify brain age-related imaging features. A support vector machine (SVM)-based classification model was constructed to differentiate MCI from SCI patients. The neurobiological basis of brain age-related imaging features was also investigated based on cognitive assessments and oxidative stress biomarkers.

Results

The gray matter volume (GMV) imaging features accurately predicted brain age in individual patients with SSVD (R² = 0.535, p < 0.001). The GMV features were primarily distributed across the subcortical system (e.g., thalamus) and dorsal attention network. SSVD patients with age acceleration showed significantly poorer Mini-Mental State Examination and Montreal Cognitive Assessment (MoCA) scores. The classification model based on GMV features could accurately distinguish MCI patients from SCI patients (area under the curve = 0.883). The classification outputs of the classification model exhibited significant associations with MoCA scores, Trail Making Tests A and B scores, Stroop Color and Word Test C scores, information processing speed total scores, and plasma levels of total antioxidant capacity in SSVD patients.

Conclusion

Brain age can be accurately quantified using GMV imaging data and shows potential clinical value for identifying early cognitive impairment in SSVD patients.

Personalized Diagnosis and Treatment for Neuroimaging in Depressive Disorders

Depressive disorders are highly heterogeneous in nature. Previous studies have not been useful for the clinical diagnosis and prediction of outcomes of major depressive disorder (MDD) at the individual level, although they provide many meaningful insights. To make inferences beyond group-level analyses, machine learning (ML) techniques can be used for the diagnosis of subtypes of MDD and the prediction of treatment responses. We searched PubMed for relevant studies published until December 2021 that included depressive disorders and applied ML algorithms in neuroimaging fields for depressive disorders. We divided these studies into two sections, namely diagnosis and treatment outcomes, for the application of prediction using ML. Structural and functional magnetic resonance imaging studies using ML algorithms were included. Thirty studies were summarized for the prediction of an MDD diagnosis. In addition, 19 studies on the prediction of treatment outcomes for MDD were reviewed. We summarized and discussed the results of previous studies. For future research results to be useful in clinical practice, ML enabling individual inferences is important. At the same time, there are important challenges to be addressed in the future.

The aberrant dynamic amplitude of low-frequency fluctuations in melancholic major depressive disorder with insomnia

BACKGROUND

Insomnia is considered one of the manifestations of sleep disorders, and its intensity is linked to the treatment effect or suicidal thoughts. Major depressive disorder (MDD) is classified into various subtypes due to heterogeneous symptoms. Melancholic MDD has been considered one of the most common subtypes with special sleep features. However, the brain functional mechanisms in melancholic MDD with insomnia remain unclear.

MATERIALS AND METHODS

Melancholic MDD and healthy controls (HCs, n = 46) were recruited for the study. Patients were divided into patients with melancholic MDD with low insomnia (mMDD-LI, n = 23) and patients with melancholic MDD with high insomnia (mMDD-HI, n = 30), according to the sleep disturbance subscale of the 17-item Hamilton Depression Rating Scale. The dynamic amplitude of low-frequency fluctuation was employed to investigate the alterations of brain activity among the three groups. Then, the correlations between abnormal dALFF values of brain regions and the severity of symptoms were investigated.

RESULTS

Lower dALFF values were found in the mMDD-HI group in the right middle temporal gyrus (MTG)/superior temporal gyrus (STG) than in the mMDD-LI (p = 0.014) and HC groups (p < 0.001). Melancholic MDD groups showed decreased dALFF values than HC in the right middle occipital gyri (MOG)/superior occipital gyri (SOG), the right cuneus, the bilateral lingual gyrus, and the bilateral calcarine (p < 0.05). Lower dALFF values than HC in the left MOG/SOG and the left cuneus in melancholic MDD groups were found, but no significant difference was found between the mMDD-LI group and HC group (p = 0.079). Positive correlations between the dALFF values in the right MTG/STG and HAMD-SD scores (the sleep disturbance subscale of the HAMD-17) in the mMDD-HI group (r = 0.41, p = 0.042) were found. In the pooled melancholic MDD, the dALFF values in the right MOG/SOG and the right cuneus (r = 0.338, p = 0.019), the left MOG/SOG and the left cuneus (r = 0.299, p = 0.039), and the bilateral lingual gyrus and the bilateral calcarine (r = 0.288, p = 0.047) were positively correlated with adjusted HAMD scores.

CONCLUSION

The occipital cortex may be related to depressive symptoms in melancholic MDD. Importantly, the right MTG/STG may play a critical role in patients with melancholic MDD with more severe insomnia.