Interpretable deep learning model for major depressive disorder assessment based on functional near-infrared spectroscopy.

BACKGROUND: Major depressive disorder (MDD) affects a substantial number of individuals worldwide. New approaches are required to improve the diagnosis of MDD, which relies heavily on subjective reports of depression-related symptoms. AIM: Establish an objective measurement and evaluation of MDD. METHODS: Functional near-infrared spectroscopy (fNIRS) was used to investigate the brain activity of MDD patients and healthy controls (HCs). Leveraging a sizeable fNIRS dataset of 263 HCs and 251 patients with MDD, including mild to moderate MDD (mMDD; n = 139) and severe MDD (sMDD; n = 77), we developed an interpretable deep learning model for screening MDD and staging its severity. RESULTS: The proposed deep learning model achieved an accuracy of 80.9% in diagnostic classification and 78.6% in severity staging for MDD. We discerned five channels with the most significant contribution to MDD identification through Shapley additive explanations (SHAP), located in the right medial prefrontal cortex, right dorsolateral prefrontal cortex, right superior temporal gyrus, and left posterior superior frontal cortex. The findings corresponded closely to the features of haemoglobin responses between HCs and individuals with MDD, as we obtained a good discriminative ability for MDD using cortical channels that are related to the disorder, namely the frontal and temporal cortical channels with areas under the curve of 0.78 and 0.81, respectively. CONCLUSION: Our study demonstrated the potential of integrating the fNIRS system with artificial intelligence algorithms to classify and stage MDD in clinical settings using a large dataset. This approach can potentially enhance MDD assessment and provide insights for clinical diagnosis and intervention.

In vivo serotonin transporter and 1A receptor binding potential and ecological momentary assessment (EMA) of stress in major depression and suicidal behavior.

We examined relationships between the serotonin system and stress in major depression and suicidal behavior. Twenty-five medication-free depressed participants (13 suicide attempters) underwent same-day [11C]DASB and [11C]CUMI-101 positron emission tomography (PET) imaging. Binding potential (BPND) to the serotonin transporter (5-HTT) and serotonin 1A (5-HT1A) receptor, respectively, was quantified using the NRU 5-HT atlas, reflecting distinct spatial distributions of multiple serotonin targets. Ecological momentary assessment (EMA) measured current stress over one week proximal to imaging. EMA stress did not differ between attempters and non-attempters. In all depressed participants, 5-HTT and 5-HT1A BPND were unrelated to EMA stress. There were region-specific effects of 5-HTT (p=0.002) and 5-HT1A BPND (p=0.03) in attempters vs. nonattempters. In attempters, region-specific associations between 5-HTT (p=0.03) and 5-HT1A (p=0.005) BPND and EMA stress emerged. While no post-hoc 5-HTT BPND correlations were significant, 5-HT1A BPND correlated positively with EMA stress in attempters in 9/10 regions (p-values<0.007), including the entire cortex except the largely occipital region 5. Brodmann-based regional analyses found diminished effects for 5-HTT and subcortically localized positive corrrelations between 5-HT1A and EMA stress, in attempters only. Given comparable depression severity and childhood and current stress between attempters and nonattempters, lower 5-HTT binding in attempters vs. nonattempters may suggest a biological risk marker. Localized lower 5-HTT and widespread higher 5-HT1A binding with stress among attempters specifically may suggest that a serotonergic phenotype might be a key determinant of risk or resiliency for suicidal behavior.

Co-alteration Network Architecture of Major Depressive Disorder: A Multi-modal Neuroimaging Assessment of Large-scale Disease Effects.

Major depressive disorder (MDD) exhibits diverse symptomology and neuroimaging studies report widespread disruption of key brain areas. Numerous theories underpinning the network degeneration hypothesis (NDH) posit that neuropsychiatric diseases selectively target brain areas via meaningful network mechanisms rather than as indistinct disease effects. The present study tests the hypothesis that MDD is a network-based disorder, both structurally and functionally. Coordinate-based meta-analysis and Activation Likelihood Estimation (CBMA-ALE) were used to assess the convergence of findings from 92 previously published studies in depression. An extension of CBMA-ALE was then used to generate a node-and-edge network model representing the co-alteration of brain areas impacted by MDD. Standardized measures of graph theoretical network architecture were assessed. Co-alteration patterns among the meta-analytic MDD nodes were then tested in independent, clinical T1-weighted structural magnetic resonance imaging (MRI) and resting-state functional (rs-fMRI) data. Differences in co-alteration profiles between MDD patients and healthy controls, as well as between controls and clinical subgroups of MDD patients, were assessed. A 65-node 144-edge co-alteration network model was derived for MDD. Testing of co-alteration profiles in replication data using the MDD nodes provided distinction between MDD and healthy controls in structural data. However, co-alteration profiles were not distinguished between patients and controls in rs-fMRI data. Improved distinction between patients and healthy controls was observed in clinically homogenous MDD subgroups in T1 data. MDD abnormalities demonstrated both structural and functional network architecture, though only structural networks exhibited between-groups differences. Our findings suggest improved utility of structural co-alteration networks for ongoing biomarker development.

Assessment of Relationship of Ketamine Dose With Magnetic Resonance Spectroscopy of Glx and GABA Responses in Adults With Major Depression: A Randomized Clinical Trial.

Importance: A single subanesthetic dose of ketamine produces an antidepressant response in patients with major depressive disorder (MDD) within hours, but the mechanism of antidepressant effect is uncertain. Objective: To evaluate whether ketamine dose and brain glutamate and glutamine (Glx) and γ-aminobutyric acid (GABA) level responses to ketamine are related to antidepressant benefit and adverse effects. Design, Setting, and Participants: This randomized, parallel-group, triple-masked clinical trial included 38 physically healthy, psychotropic medication-free adult outpatients who were in a major depressive episode of MDD but not actively suicidal. The trial was conducted at Columbia University Medical Center. Data were collected from February 2012 to May 2015. Data analysis was conducted from January to March 2020. Intervention: Participants received 1 dose of placebo or ketamine (0.1, 0.2, 0.3, 0.4, or 0.5 mg/kg) intravenously during 40 minutes of a proton magnetic resonance spectroscopy scan that measured ventro-medial prefrontal cortex Glx and GABA levels in 13-minute data frames. Main Outcomes and Measures: Clinical improvement was measured using a 22-item version of the Hamilton Depression Rating Scale (HDRS-22) 24 hours after ketamine was administered. Ketamine and metabolite blood levels were measured after the scan. Results: A total of 38 individuals participated in the study, with a mean (SD) age of 38.6 (11.2) years, 23 (60.5%) women, and 25 (65.8%) White patients. Improvement in HDRS-22 score at 24 hours correlated positively with ketamine dose (t36 = 2.81; P = .008; slope estimate, 19.80 [95% CI, 5.49 to 34.11]) and blood level (t36 = 2.25; P = .03; slope estimate, 0.070 [95% CI, 0.007 to 0.133]). The lower the Glx response, the better the antidepressant response (t33 = -2.400; P = .02; slope estimate, -9.85 [95% CI, -18.2 to -1.50]). Although GABA levels correlated with Glx (t33 = 8.117; P < .001; slope estimate, 0.510 [95% CI, 0.382 to 0.638]), GABA response did not correlate with antidepressant effect. When both ketamine dose and Glx response were included in a mediation analysis model, ketamine dose was no longer associated with antidepressant effect, indicating that Glx response mediated the relationship. Adverse effects were related to blood levels in men only (t5 = 2.606; P = .048; estimated slope, 0.093 [95% CI, 0.001 to 0.186]), but Glx and GABA response were not related to adverse effects. Conclusions and Relevance: In this study, intravenous ketamine dose and blood levels correlated positively with antidepressant response. The Glx response correlated inversely with ketamine dose and with antidepressant effect. Future studies are needed to determine whether the relationship between Glx level and antidepressant effect is due to glutamate or glutamine. Trial Registration: ClinicalTrials.gov Identifier: NCT01558063.

Major depressive disorder assessment via enhanced k-nearest neighbor method and EEG signals.

The aim of this paper is to introduce a novel method using short-term EEG signals to separate depressed patients from healthy controls. Five common frequency bands (delta, theta, alpha, beta and gamma) were extracted from the signals as linear features, as well as, wavelet packet decomposition to break down signals into certain frequency bands. Afterwards, two entropy measures, namely sample entropy and approximate entropy were applied on the wavelet packet coefficients as nonlinear features, and significant features were selected via genetic algorithm (GA). Three machine-learning algorithms were used for classification; including support vector machine (SVM), multilayer perceptron (MLP) a novel enhanced K-nearest neighbors (E-KNN), which uses GA to optimize the feature-space distances and provides a feature importance index. The highest accuracy obtained by using frequency-based features was from gamma oscillations which resulted in 91.38%. Performance of nonlinear features were better compared to the frequency-based features and the results showed 94.28% accuracy. The combination of the features showed 98.44% accuracy with the new proposed E-KNN classifier.

Simultaneous confidence corridors for mean functions in functional data analysis of imaging data.

Motivated by recent work involving the analysis of biomedical imaging data, we present a novel procedure for constructing simultaneous confidence corridors for the mean of imaging data. We propose to use flexible bivariate splines over triangulations to handle an irregular domain of the images that is common in brain imaging studies and in other biomedical imaging applications. The proposed spline estimators of the mean functions are shown to be consistent and asymptotically normal under some regularity conditions. We also provide a computationally efficient estimator of the covariance function and derive its uniform consistency. The procedure is also extended to the two-sample case in which we focus on comparing the mean functions from two populations of imaging data. Through Monte Carlo simulation studies, we examine the finite sample performance of the proposed method. Finally, the proposed method is applied to analyze brain positron emission tomography data in two different studies. One data set used in preparation of this article was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database.

Assessment of Striatal Dopamine Transporter Binding in Individuals With Major Depressive Disorder: In Vivo Positron Emission Tomography and Postmortem Evidence.

Importance: Major depressive disorder (MDD) might involve dopamine (DA) reductions. The DA transporter (DAT) regulates DA clearance and neurotransmission and is sensitive to DA levels, with preclinical studies (including those involving inescapable stressors) showing that DAT density decreases when DA signaling is reduced. Despite preclinical data, evidence of reduced DAT in MDD is inconclusive. Objective: Using a highly selective DAT positron emission tomography (PET) tracer ([11C] altropane), DAT availability was probed in individuals with MDD who were not taking medication. Levels of DAT expression were also evaluated in postmortem tissues from donors with MDD who died by suicide. Design, Setting, and Participants: This cross-sectional PET study was conducted at McLean Hospital (Belmont, Massachusetts) and Massachusetts General Hospital (Boston) and enrolled consecutive individuals with MDD who were not taking medication and demographically matched healthy controls between January 2012 and March 2014. Brain tissues were obtained from the Douglas-Bell Canada Brain Bank. For the PET component, 25 individuals with current MDD who were not taking medication and 23 healthy controls recruited from McLean Hospital were included (all provided usable data). For the postmortem component, 15 individuals with depression and 14 healthy controls were considered. Intervention: PET scan. Main Outcomes and Measures: Striatal and midbrain DAT binding potential was assessed. For the postmortem component, tyrosine hydroxylase and DAT levels were evaluated using Western blots. Results: Compared with 23 healthy controls (13 women [56.5%]; mean [SD] age, 26.49 [7.26] years), 25 individuals with MDD (19 women [76.0%]; mean [SD] age, 26.52 [5.92] years) showed significantly lower in vivo DAT availability in the bilateral putamen and ventral tegmental area (Cohen d range, -0.62 to -0.71), and both reductions were exacerbated with increasing numbers of depressive episodes. Unlike healthy controls, the MDD group failed to show an age-associated reduction in striatal DAT availability, with young individuals with MDD being indistinguishable from older healthy controls. Moreover, DAT availability in the ventral tegmental area was lowest in individuals with MDD who reported feeling trapped in stressful circumstances. Lower DAT levels (and tyrosine hydroxylase) in the putamen of MDD compared with healthy controls were replicated in postmortem analyses (Cohen d range, -0.92 to -1.15). Conclusions and Relevance: Major depressive disorder, particularly with recurring episodes, is characterized by decreased striatal DAT expression, which might reflect a compensatory downregulation due to low DA signaling within mesolimbic pathways.

Blunted amygdala activity is associated with depression severity in treatment-resistant depression.

Up to 50% of individuals with major depressive disorder (MDD) do not recover after two antidepressant medication trials, and therefore meet the criteria for treatment-resistant depression (TRD). Mindfulness-based cognitive therapy (MBCT) is one promising treatment; however, the extent to which MBCT influences clinical outcomes relative to baseline neural activation remains unknown. In the present study we investigated baseline differences in amygdala activation between TRD patients and healthy controls (HCs), related amygdala activation to depression symptoms, and examined the impacts of MBCT and amygdala activation on longitudinal depression outcomes. At baseline, TRD patients (n = 80) and HCs (n = 37) participated in a functional magnetic resonance imaging task in which they identified either the emotion (affect labeling) or the gender (gender labeling) of faces, or passively viewed faces (observing). The TRD participants then completed eight weeks of MBCT or a health enhancement program (HEP). Relative to HCs, the TRD patients demonstrated less amygdala activation during affect labeling, and marginally less during gender labeling. Blunted amygdala activation in TRD patients during affect labeling was associated with greater depression severity. MBCT was associated with greater depression reductions than was HEP directly following treatment; however, at 52 weeks the treatment effect was not significant, and baseline amygdala activation across the task conditions predicted depression severity in both groups. TRD patients have blunted amygdala responses during affect labeling that are associated with greater concurrent depression. Furthermore, although MBCT produced greater short-term improvements in depression than did HEP, overall baseline amygdala reactivity was predictive of long-term clinical outcomes in both groups.

Effects of Antenatal Maternal Depressive Symptoms and Socio-Economic Status on Neonatal Brain Development are Modulated by Genetic Risk.

This study included 168 and 85 mother-infant dyads from Asian and United States of America cohorts to examine whether a genomic profile risk score for major depressive disorder (GPRSMDD) moderates the association between antenatal maternal depressive symptoms (or socio-economic status, SES) and fetal neurodevelopment, and to identify candidate biological processes underlying such association. Both cohorts showed a significant interaction between antenatal maternal depressive symptoms and infant GPRSMDD on the right amygdala volume. The Asian cohort also showed such interaction on the right hippocampal volume and shape, thickness of the orbitofrontal and ventromedial prefrontal cortex. Likewise, a significant interaction between SES and infant GPRSMDD was on the right amygdala and hippocampal volumes and shapes. After controlling for each other, the interaction effect of antenatal maternal depressive symptoms and GPRSMDD was mainly shown on the right amygdala, while the interaction effect of SES and GPRSMDD was mainly shown on the right hippocampus. Bioinformatic analyses suggested neurotransmitter/neurotrophic signaling, SNAp REceptor complex, and glutamate receptor activity as common biological processes underlying the influence of antenatal maternal depressive symptoms on fetal cortico-limbic development. These findings suggest gene-environment interdependence in the fetal development of brain regions implicated in cognitive-emotional function. Candidate biological mechanisms involve a range of brain region-specific signaling pathways that converge on common processes of synaptic development.

Improvement of health-related quality of life in depression after transcranial magnetic stimulation in a naturalistic trial is associated with decreased perfusion in precuneus.

BACKGROUND: Assessing Health-related Quality of life (HRQoL) is necessary to evaluate care and treatments provided to patients with major depressive disorder (MDD), in addition to the traditional assessment of clinical outcomes. However, HRQoL remains under-utilized to assess the effectiveness of repetitive transcranial magnetic stimulation (rTMS) in research or in a routine clinical setting. The primary objective of this exploratory study on MDD was to investigate the impact of low-frequency rTMS on HRQoL using the SF-36 questionnaire. A secondary objective was to study the functional neural substrate underlying HRQoL changes using neuroimaging. METHODS: Fifteen right-handed patients who met DSM-IV criteria for MDD participated in the study. HRQoL was assessed using the SF-36, and regional cerebral blood (rCBF) flow using 99mTc-ECD-SPECT. Voxel based correlation was searched between concomitant changes in rCBF and in HRQoL after rTMS. RESULTS: Role-Physical Problems dimension showed a statistical significant improvement of 73.2% (p = 0.001) and an effect size (Cohen's d) of 0.43, indicating moderate effect. Five SF-36 dimension scores and the two composite scores showed effect sizes ranged from 0.28 to 0.43. Improvement of Mental Composite Score (MCS)-SF-36 after rTMS was correlated with a concomitant decrease of precuneus perfusion (p < 0.001). Post-hoc analyses confirmed that decreased perfusion in precuneus was correlated with improvement of HRQoL, especially for MCS (r = -0.71; p < 0.001), Mental Health (r = -0.81; p < 0.001) and Social Functioning (r = -0.57; p = 0.026) dimensions. CONCLUSIONS: This study suggests low-frequency rTMS can improve HRQoL, through its role-physical problems dimension, in patients with MDD. This improvement is associated with a decreased perfusion of the precuneus, a brain area involved in self-focus and self-processing, arguing for a neural substrate to the impact of rTMS on HRQoL.

Assessment of cerebral blood flow findings using 99mTc-ECD single-photon emission computed tomography in patients diagnosed with major depressive disorder.

BACKGROUND: Single-photon emission computed tomography (SPECT) is used as an ancillary diagnostic tool in clinical psychiatry. A variety of SPECT studies has been conducted on the findings and the factors that affect the findings, and there is a possibility that age has an effect on cerebral blood flow. We used SPECT to verify the cerebral blood flow findings of patients with major depressive disorder (MDD) considering the effect of age on the findings. METHODS: We conducted a retrospective survey of inpatients who fulfilled the DSM-IV diagnostic criteria for MDD and who had undergone imaging by technetium-99m ethyl cysteinate dimer ((99m)Tc-ECD) SPECT (N=98, 37 males). After excluding organic factors and comorbidities, we established a depression group (N=61, 24 males) and conducted an inter-group comparison with a normal control group by using SPM software considering the effect of age. RESULTS: The depression group showed the reduction of cerebral blood flow in the prefrontal area bilaterally, predominantly on the left, including the orbitofrontal cortex, anterior portion of the gyrus cinguli, and dorsolateral prefrontal area, in the left temporal lobe, and in the occipital lobe bilaterally, predominantly on the left. The findings were common to all age groups and that age-specific pattern was not detected. LIMITATIONS: The facts that this was a retrospective study and small sample size in each age group were limitations of this research. Although it also seems important to evaluate the impact of medication on cerebral blood flow and conduct an evaluation according to the subtype of depression, but we couldn't in this study. In the future it will be necessary to accumulate additional cases and conduct additional studies, including a prospective survey. CONCLUSION: The results of this study may suggest the existence of a common biological background in patients with MDD that is unaffected by age.

Regional cerebral blood flow in the assessment of major depression and Alzheimer's disease in the early elderly.

BACKGROUND: Alzheimer's disease and major depression are representative diseases that present forgetfulness and a depressive mood. It is often difficult to make a differential diagnosis between the two in the initial phase. AIM: To evaluate the differential diagnosis method using regional cerebral blood flow patterns with a three-dimensional stereotactic surface projection technique. METHODS: Twenty early-elderly patients with mild and moderate forgetfulness were studied. Among them, 10 were diagnosed as having major depression (the MD group) and the other 10 as having Alzheimer's disease (the AD group). All patients underwent cerebral perfusion single photon emission computed tomography (SPECT) with [(123)I]iodoamphetamine. A z-score was calculated for each pixel of the cerebral surface. Twenty-one circular regions of interest (ROIs) were placed on the z-score map. The significance of the statistical difference in ROI values between the two groups was determined by using the two-sided Mann-Whitney U-test. RESULTS: The z-scores for the lateral parietal, lateral temporal, bilateral precuneus and bilateral posterior cingulate gyrus were significantly reduced in the AD group compared with those in the MD group. The z-scores for the lateral frontal, left thalamus and bilateral medial frontal regions were significantly lower in the MD group than in the AD group. CONCLUSION: Our study demonstrated a difference in regional cerebral blood flow patterns between the early elderly with Alzheimer's disease and those with major depression. All patients were classified into the appropriate categories using discriminant analysis and z-scores of frontal and parietal regions. Brain perfusion SPECT was a useful tool for the differential diagnosis between Alzheimer's disease and major depression.