Development and validation of a multimodal neuroimaging biomarker for electroconvulsive therapy outcome in depression: a multicenter machine learning analysis.

BACKGROUND: Electroconvulsive therapy (ECT) is the most effective intervention for patients with treatment resistant depression. A clinical decision support tool could guide patient selection to improve the overall response rate and avoid ineffective treatments with adverse effects. Initial small-scale, monocenter studies indicate that both structural magnetic resonance imaging (sMRI) and functional MRI (fMRI) biomarkers may predict ECT outcome, but it is not known whether those results can generalize to data from other centers. The objective of this study was to develop and validate neuroimaging biomarkers for ECT outcome in a multicenter setting. METHODS: Multimodal data (i.e. clinical, sMRI and resting-state fMRI) were collected from seven centers of the Global ECT-MRI Research Collaboration (GEMRIC). We used data from 189 depressed patients to evaluate which data modalities or combinations thereof could provide the best predictions for treatment remission (HAM-D score ⩽7) using a support vector machine classifier. RESULTS: Remission classification using a combination of gray matter volume and functional connectivity led to good performing models with average 0.82-0.83 area under the curve (AUC) when trained and tested on samples coming from the three largest centers (N = 109), and remained acceptable when validated using leave-one-site-out cross-validation (0.70-0.73 AUC). CONCLUSIONS: These results show that multimodal neuroimaging data can be used to predict remission with ECT for individual patients across different treatment centers, despite significant variability in clinical characteristics across centers. Future development of a clinical decision support tool applying these biomarkers may be feasible.

An Improved Average Atomic Number Calculation for Estimating Backscatter and Continuum Production in Compounds.

It is often assumed that electron backscatter and continuum (bremsstrahlung) productions emitted from electron-solid interactions during X-ray microanalysis in compounds can be extrapolated from pure element observations by means of the assumption of average atomic number, or Z-bar (Z¯). For pure elements the average Z is equal to the atomic number, but this direct approach fails for compounds. The use of simple atomic fractions yields completely spurious results, and while the commonly used mass fraction Z averaging produces fairly reasonable results, we know from physical considerations that the mass of the neutron plays only a negligible role in such interactions below ∼1 MeV. Therefore, including the mass or atomic weight in such calculations can only introduce further errors in these models. We present an expression utilizing atomic fractions of the atomic numbers of the elements in the compound (Z fraction), with an exponent to account for the variation in nuclear screening as a function of the element Z value.

Anterior default mode network and posterior insular connectivity is predictive of depressive symptom reduction following serial ketamine infusion.

BACKGROUND: Ketamine is a rapidly-acting antidepressant treatment with robust response rates. Previous studies have reported that serial ketamine therapy modulates resting state functional connectivity in several large-scale networks, though it remains unknown whether variations in brain structure, function, and connectivity impact subsequent treatment success. We used a data-driven approach to determine whether pretreatment multimodal neuroimaging measures predict changes along symptom dimensions of depression following serial ketamine infusion. METHODS: Patients with depression (n = 60) received structural, resting state functional, and diffusion MRI scans before treatment. Depressive symptoms were assessed using the 17-item Hamilton Depression Rating Scale (HDRS-17), the Inventory of Depressive Symptomatology (IDS-C), and the Rumination Response Scale (RRS) before and 24 h after patients received four (0.5 mg/kg) infusions of racemic ketamine over 2 weeks. Nineteen unaffected controls were assessed at similar timepoints. Random forest regression models predicted symptom changes using pretreatment multimodal neuroimaging and demographic measures. RESULTS: Two HDRS-17 subscales, the HDRS-6 and core mood and anhedonia (CMA) symptoms, and the RRS: reflection (RRSR) scale were predicted significantly with 19, 27, and 1% variance explained, respectively. Increased right medial prefrontal cortex/anterior cingulate and posterior insula (PoI) and lower kurtosis of the superior longitudinal fasciculus predicted reduced HDRS-6 and CMA symptoms following treatment. RRSR change was predicted by global connectivity of the left posterior cingulate, left insula, and right superior parietal lobule. CONCLUSIONS: Our findings support that connectivity of the anterior default mode network and PoI may serve as potential biomarkers of antidepressant outcomes for core depressive symptoms.

Modulation of the functional connectome in major depressive disorder by ketamine therapy.

BACKGROUND: Subanesthetic ketamine infusion therapy can produce fast-acting antidepressant effects in patients with major depression. How single and repeated ketamine treatment modulates the whole-brain functional connectome to affect clinical outcomes remains uncharacterized. METHODS: Data-driven whole brain functional connectivity (FC) analysis was used to identify the functional connections modified by ketamine treatment in patients with major depressive disorder (MDD). MDD patients (N = 61, mean age = 38, 19 women) completed baseline resting-state (RS) functional magnetic resonance imaging and depression symptom scales. Of these patients, n = 48 and n = 51, completed the same assessments 24 h after receiving one and four 0.5 mg/kg intravenous ketamine infusions. Healthy controls (HC) (n = 40, 24 women) completed baseline assessments with no intervention. Analysis of RS FC addressed effects of diagnosis, time, and remitter status. RESULTS: Significant differences (p < 0.05, corrected) in RS FC were observed between HC and MDD at baseline in the somatomotor network and between association and default mode networks. These disruptions in FC in MDD patients trended toward control patterns with ketamine treatment. Furthermore, following serial ketamine infusions, significant decreases in FC were observed between the cerebellum and salience network (SN) (p < 0.05, corrected). Patient remitters showed increased FC between the cerebellum and the striatum prior to treatment that decreased following treatment, whereas non-remitters showed the opposite pattern. CONCLUSION: Results support that ketamine treatment leads to neurofunctional plasticity between distinct neural networks that are shown as disrupted in MDD patients. Cortico-striatal-cerebellar loops that encompass the SN could be a potential biomarker for ketamine treatment.

Hippocampal subregions and networks linked with antidepressant response to electroconvulsive therapy.

Electroconvulsive therapy (ECT) has been repeatedly linked to hippocampal plasticity. However, it remains unclear what role hippocampal plasticity plays in the antidepressant response to ECT. This magnetic resonance imaging (MRI) study tracks changes in separate hippocampal subregions and hippocampal networks in patients with depression (n = 44, 23 female) to determine their relationship, if any, with improvement after ECT. Voxelwise analyses were restricted to the hippocampus, amygdala, and parahippocampal cortex, and applied separately for responders and nonresponders to ECT. In analyses of arterial spin-labeled (ASL) MRI, nonresponders exhibited increased cerebral blood flow (CBF) in bilateral anterior hippocampus, while responders showed CBF increases in right middle and left posterior hippocampus. In analyses of gray matter volume (GMV) using T1-weighted MRI, GMV increased throughout bilateral hippocampus and surrounding tissue in nonresponders, while responders showed increased GMV in right anterior hippocampus only. Using CBF loci as seed regions, BOLD-fMRI data from healthy controls (n = 36, 19 female) identified spatially separable neurofunctional networks comprised of different brain regions. In graph theory analyses of these networks, functional connectivity within a hippocampus-thalamus-striatum network decreased only in responders after two treatments and after index. In sum, our results suggest that the location of ECT-related plasticity within the hippocampus may differ according to antidepressant outcome, and that larger amounts of hippocampal plasticity may not be conducive to positive antidepressant response. More focused targeting of hippocampal subregions and/or circuits may be a way to improve ECT outcome.

Accounting for symptom heterogeneity can improve neuroimaging models of antidepressant response after electroconvulsive therapy.

Depression symptom heterogeneity limits the identifiability of treatment-response biomarkers. Whether improvement along dimensions of depressive symptoms relates to separable neural networks remains poorly understood. We build on work describing three latent symptom dimensions within the 17-item Hamilton Depression Rating Scale (HDRS) and use data-driven methods to relate multivariate patterns of patient clinical, demographic, and brain structural changes over electroconvulsive therapy (ECT) to dimensional changes in depressive symptoms. We included 110 ECT patients from Global ECT-MRI Research Collaboration (GEMRIC) sites who underwent structural MRI and HDRS assessments before and after treatment. Cross validated random forest regression models predicted change along symptom dimensions. HDRS symptoms clustered into dimensions of somatic disturbances (SoD), core mood and anhedonia (CMA), and insomnia. The coefficient of determination between predicted and actual changes were 22%, 39%, and 39% (all p < .01) for SoD, CMA, and insomnia, respectively. CMA and insomnia change were predicted more accurately than HDRS-6 and HDRS-17 changes (p < .05). Pretreatment symptoms, body-mass index, and age were important predictors. Important imaging predictors included the right transverse temporal gyrus and left frontal pole for the SoD dimension; right transverse temporal gyrus and right rostral middle frontal gyrus for the CMA dimension; and right superior parietal lobule and left accumbens for the insomnia dimension. Our findings support that recovery along depressive symptom dimensions is predicted more accurately than HDRS total scores and are related to unique and overlapping patterns of clinical and demographic data and volumetric changes in brain regions related to depression and near ECT electrodes.

Modulation of amygdala reactivity following rapidly acting interventions for major depression.

Electroconvulsive therapy (ECT) and ketamine treatment both induce rapidly acting antidepressant effects in patients with major depressive disorder unresponsive to standard treatments, yet their specific impact on emotion processing is unknown. Here, we examined the neural underpinnings of emotion processing within and across patients (N = 44) receiving either ECT (N = 17, mean age: 36.8, 11.0 SD) or repeated subanesthetic (0.5 mg/kg) intravenous ketamine therapy (N = 27, mean age: 37.3, 10.8 SD) using a naturalistic study design. MRI and clinical data were collected before (TP1) and after treatment (TP2); healthy controls (N = 31, mean age: 34.5, 13.5 SD) completed one MRI session (TP1). An fMRI face-matching task probed negative- and positive-valence systems. Whole-brain analysis, comparing neurofunctional changes within and across treatment groups, targeted brain regions involved in emotional facial processing, and included regions-of-interest analysis of amygdala responsivity. Main findings revealed a decrease in amygdalar reactivity after both ECT and ketamine for positive and negative emotional face processing (p < .05 family wise-error (FWE) corrected). Subthreshold changes were observed between treatments within the dorsolateral prefrontal cortex and insula (p < .005, uncorrected). BOLD change for positive faces in the inferior parietal cortex significantly correlated with overall symptom improvement, and BOLD change in frontal regions correlated with anxiety for negative faces, and anhedonia for positive faces (p < .05 FWE corrected). Both serial ketamine and ECT treatment modulate amygdala response, while more subtle treatment-specific changes occur in the larger functional network. Findings point to both common and differential mechanistic upstream systems-level effects relating to fast-acting antidepressant response, and symptoms of anxiety and anhedonia, for the processing of emotionally valenced stimuli.

Inflammation and depression treatment response to electroconvulsive therapy: Sex-specific role of interleukin-8.

Females suffer from depression at twice the rate of males and have differential neural and emotional responses to inflammation. However, sex-specific evaluation of relationships between inflammation and response to depression treatments are lacking. Some data suggest that interleukin(IL)-8 predicts treatment response to antidepressants and has a relationship with depressive symptom severity. This study examines whether IL-8 predicts treatment response to electroconvulsive therapy (ECT), and whether there are sex specific effects. In 40 depressed patients (22 female), plasma levels of IL-8, as well as other markers of inflammation including IL-6, IL-10, tumor necrosis factor (TNF)-α, and C-reactive protein (CRP) were obtained prior to administration of ECT and after completion of the index treatment series. Depression treatment response was defined as ≥ 50% reduction in Hamilton Depression Rating Scale (HAM-D) Score. Baseline levels of IL-8 differed by responder status, depending on sex (group × sex interaction: ß = -0.571, p = 0.04), with female responders having lower levels of IL-8 at baseline as compared to female non-responders [t(20) = 2.37, p = 0.03]. Further, IL-8 levels from baseline to end of treatment differed by responder status, depending on sex (group × sex × time interaction: [F(1,36) = 9.48, p = 0.004]), and change in IL-8 from baseline to end of treatment was negatively correlated with percentage change in HAM-D score in females (ß = -0.458, p = 0.03), but not in males (ß = 0.315, p = 0.20). Other inflammatory markers did not differ in relation to responder status and sex. Further evaluation of sex differences in the relationship between IL-8, depression, and treatment response, across disparate treatment modalities, may inform mechanisms of response and aid in development of personalized medicine strategies.

Depressive Symptom Dimensions in Treatment-Resistant Major Depression and Their Modulation With Electroconvulsive Therapy.

OBJECTIVE: Symptom heterogeneity in major depressive disorder obscures diagnostic and treatment-responsive biomarker identification. Whether symptom constellations are differentially changed by electroconvulsive therapy (ECT) remains unknown. We investigate the clustering of depressive symptoms over the ECT index and whether ECT differentially influences symptom clusters. METHODS: The 17-item Hamilton Depression Rating Scale (HDRS-17) was collected from 111 patients with current depressive episode before and after ECT from 4 independent participating sites of the Global ECT-MRI Research Collaboration. Exploratory factor analysis of HDRS-17 items pre- and post-ECT treatment identified depressive symptom dimensions before and after ECT. A 2-way analysis of covariance was used to determine whether baseline symptom clusters were differentially changed by ECT between treatment remitters (defined as patients with posttreatment HDRS-17 total score ≤8) and nonremitters while controlling for pulse width, titration method, concurrent antidepressant treatment, use of benzodiazepine, and demographic variables. RESULTS: A 3-factor solution grouped pretreatment HDRS-17 items into core mood/anhedonia, somatic, and insomnia dimensions. A 2-factor solution best described the symptoms at posttreatment despite poorer separation of items. Among remitters, core mood/anhedonia symptoms were significantly more reduced than somatic and insomnia dimensions. No differences in symptom dimension trajectories were observed among nonremitting patients. CONCLUSIONS: Electroconvulsive therapy targets the underlying source of depressive symptomatology and may confer differential degrees of improvement in certain core depressive symptoms. Our findings of differential trajectories of symptom clusters over the ECT index might help related predictive biomarker studies to refine their approaches by identifying predictors of change along each latent symptom dimension.

Structural studies of RFCCtf18 reveal a novel chromatin recruitment role for Dcc1.

Replication factor C complexes load and unload processivity clamps from DNA and are involved in multiple DNA replication and repair pathways. The RFCCtf18 variant complex is required for activation of the intra-S-phase checkpoint at stalled replication forks and aids the establishment of sister chromatid cohesion. Unlike other RFC complexes, RFCCtf18 contains two non-Rfc subunits, Dcc1 and Ctf8. Here, we present the crystal structure of the Dcc1-Ctf8 heterodimer bound to the C-terminus of Ctf18. We find that the C-terminus of Dcc1 contains three-winged helix domains, which bind to both ssDNA and dsDNA We further show that these domains are required for full recruitment of the complex to chromatin, and correct activation of the replication checkpoint. These findings provide the first structural data on a eukaryotic seven-subunit clamp loader and define a new biochemical activity for Dcc1.

Relationships Between Subcortical Shape Measures and Subjective Symptom Reporting in US Service Members With Mild Traumatic Brain Injury.

OBJECTIVE: To assess interactions of subcortical structure with subjective symptom reporting associated with mild traumatic brain injury (mTBI), using advanced shape analysis derived from volumetric MRI. PARTICIPANTS: Seventy-six cognitively symptomatic individuals with mTBI and 59 service members sustaining only orthopedic injury. DESIGN: Descriptive cross-sectional study. MAIN MEASURES: Self-report symptom measures included the PTSD Checklist-Military, Neurobehavioral Symptom Inventory, and Symptom Checklist-90-Revised. High-dimensional measures of shape characteristics were generated from volumetric MRI for 7 subcortical structures in addition to standard volume measures. RESULTS: Several significant interactions between group status and symptom measures were observed across the various shape measures. These interactions were revealed in the right thalamus and globus pallidus for each of the shape measures, indicating differences in structure thickness and expansion/contraction for these regions. No relationships with volume were observed. CONCLUSION: Results provide evidence for the sensitivity of shape measures in differentiating symptomatic mTBI individuals from controls, while volumetric measures did not exhibit this same sensitivity. Disruptions to thalamic nuclei identified here highlight the role of the thalamus in the spectrum of symptoms associated with mTBI. Additional work is needed to prospectively, and longitudinally, assess these measures along with cognitive performance and advanced multimodal imaging methods to extend the utility of shape analysis in relation to functional outcomes in this population.

Diffusion Imaging Findings in US Service Members With Mild Traumatic Brain Injury and Posttraumatic Stress Disorder.

OBJECTIVE: Use diffusion tensor imaging to investigate white matter microstructure attributable to mild TBI (mTBI) and/or posttraumatic stress disorder (PTSD). PARTICIPANTS: Twenty-seven individuals with mTBI only, 16 with PTSD only, 42 with mTBI + PTSD, and 43 service members who sustained orthopedic injury. DESIGN: Descriptive cross-sectional study. MAIN MEASURES: Clinical diffusion tensor imaging sequence to assess fractional anisotropy, mean, axial, and radial diffusivity within selected regions of interest. RESULTS: Corrected analyses revealed a pattern of lower white matter integrity in the PTSD group for several scalar metrics. Regions affected included primarily right hemisphere areas of the internal capsule. These differences associated with the PTSD only cohort were observed in relation to all 3 comparison groups, while the mTBI + PTSD group did not exhibit any notable pattern of white matter abnormalities. CONCLUSION: Results suggest that lower resolution scan sequences are sensitive to post-acute abnormalities associated with PTSD, particularly in the right hemisphere. In addition, these findings suggest that ongoing PTSD symptoms are associated with differences in white matter diffusion that are more readily detected in a clinical scan sequence than mTBI abnormalities. Future studies are needed to prospectively assess service members prior to onset of injury to verify this pattern of results.

Evaluating the impact a proposed family planning model would have on maternal and infant mortality in Afghanistan.

OBJECTIVE: This study aimed to assess the potential impact a proposed family planning model would have on reducing maternal and infant mortality in Afghanistan. BACKGROUND: Afghanistan has a high total fertility rate, high infant mortality rate, and high maternal mortality rate. Afghanistan also has tremendous socio-cultural barriers to and misconceptions about family planning services. METHODS: We applied predictive statistical models to a proposed family planning model for Afghanistan to better understand the impact increased family planning can have on Afghanistan's maternal mortality rate and infant mortality rate. We further developed a sensitivity analysis that illustrates the number of maternal and infant deaths that can be averted over 5 years according to different increases in contraceptive prevalence rates. RESULTS: Incrementally increasing contraceptive prevalence rates in Afghanistan from 10% to 60% over the course of 5 years could prevent 11,653 maternal deaths and 317,084 infant deaths, a total of 328,737 maternal and infant deaths averted. CONCLUSION: Achieving goals in reducing maternal and infant mortality rates in Afghanistan requires a culturally relevant approach to family planning that will be supported by the population. The family planning model for Afghanistan presents such a solution and holds the potential to prevent hundreds of thousands of deaths.

Novel lactoferrin-conjugated gallium complex to treat Pseudomonas aeruginosa wound infection.

Pseudomonas aeruginosa is one of the leading causes of opportunistic infections such as chronic wound infection that could lead to multiple organ failure and death. Gallium (Ga3+) ions are known to inhibit P. aeruginosa growth and biofilm formation but require carrier for localized controlled delivery. Lactoferrin (LTf), a two-lobed protein, can deliver Ga3+ at sites of infection. This study aimed to develop a Ga-LTf complex for the treatment of wound infection. The characterisation of the Ga-LTf complex was conducted using differential scanning calorimetry (DSC), Infra-Red (FTIR) and Inductive Coupled Plasma Optical Emission Spectrometry (ICP-OES). The antibacterial activity was assessed by agar disc diffusion, liquid broth and biofilm inhibition assays using the colony forming units (CFUs). The healing capacity and biocompatibility were evaluated using a P.aeruginosa infected wound in a rat model. DSC analyses showed thermal transition consistent with apo-lactoferrin; FTIR confirmed the complexation of gallium to lactoferrin. ICP-OES confirmed the controlled local delivery of Ga3+. Ga-LTf showed a 0.57 log10 CFUs reduction at 24 h compared with untreated control in planktonic liquid broth assay. Ga-LTf showed the highest antibiofilm activity with a 2.24 log10 CFUs reduction at 24 h. Furthermore, Ga-LTf complex is biocompatible without any adverse effect on brain, kidney, liver and spleen of rats tested in this study. Ga-LTf can be potentially promising novel therapeutic agent to treat pathogenic bacterial infections.

Modulation of habenular and nucleus accumbens functional connectivity by ketamine in major depression.

INTRODUCTION: Major depressive disorder (MDD) is associated with dysfunctional reward processing, which involves functional circuitry of the habenula (Hb) and nucleus accumbens (NAc). Since ketamine elicits rapid antidepressant and antianhedonic effects in MDD, this study sought to investigate how serial ketamine infusion (SKI) treatment modulates static and dynamic functional connectivity (FC) in Hb and NAc functional networks. METHODS: MDD participants (n = 58, mean age = 40.7 years, female = 28) received four ketamine infusions (0.5 mg/kg) 2-3 times weekly. Resting-state functional magnetic resonance imaging (fMRI) scans and clinical assessments were collected at baseline and 24 h post-SKI. Static FC (sFC) and dynamic FC variability (dFCv) were calculated from left and right Hb and NAc seeds to all other brain regions. Changes in FC pre-to-post SKI, and correlations with changes with mood and anhedonia were examined. Comparisons of FC between patients and healthy controls (HC) at baseline (n = 55, mean age = 32.6, female = 31), and between HC assessed twice (n = 16) were conducted as follow-up analyses. RESULTS: Following SKI, significant increases in left Hb-bilateral visual cortex FC, decreases in left Hb-left inferior parietal cortex FC, and decreases in left NAc-right cerebellum FC occurred. Decreased dFCv between left Hb and right precuneus and visual cortex, and decreased dFCv between right NAc and right visual cortex both significantly correlated with improvements in mood ratings. Decreased FC between left Hb and bilateral visual/parietal cortices as well as increased FC between left NAc and right visual/parietal cortices both significantly correlated with improvements in anhedonia. No differences were observed between HC at baseline or over time. CONCLUSION: Subanesthetic ketamine modulates functional pathways linking the Hb and NAc with visual, parietal, and cerebellar regions in MDD. Overlapping effects between Hb and NAc functional systems were associated with ketamine's therapeutic response.

Microstructural Organization of Distributed White Matter Associated With Fine Motor Control in US Service Members With Mild Traumatic Brain Injury.

Mild traumatic brain injury (mTBI) is the most common form of brain injury. While most individuals recover from mTBI, roughly 20% experience persistent symptoms, potentially including reduced fine motor control. We investigate relationships between regional white matter organization and subcortical volumes associated with performance on the Grooved Pegboard (GPB) test in a large cohort of military Service Members and Veterans (SM&Vs) with and without a history of mTBI(s). Participants were enrolled in the Long-term Impact of Military-relevant Brain Injury Consortium-Chronic Effects of Neurotrauma Consortium. SM&Vs with a history of mTBI(s) (n = 847) and without mTBI (n = 190) underwent magnetic resonance imaging and the GPB test. We first examined between-group differences in GPB completion time. We then investigated associations between GPB performance and regional structural imaging measures (tractwise diffusivity, subcortical volumes, and cortical thickness) in SM&Vs with a history of mTBI(s). Lastly, we explored whether mTBI history moderated associations between imaging measures and GPB performance. SM&Vs with mTBI(s) performed worse than those without mTBI(s) on the non-dominant hand GPB test at a trend level (p < 0.1). Higher fractional anisotropy (FA) of tracts including the posterior corona radiata, superior longitudinal fasciculus, and uncinate fasciculus were associated with better GPB performance in the dominant hand in SM&Vs with mTBI(s). These findings support that the organization of several white matter bundles are associated with fine motor performance in SM&Vs. We did not observe that mTBI history moderated associations between regional FA and GPB test completion time, suggesting that chronic mTBI may not significantly influence fine motor control.

Persistent MRI Findings Unique to Blast and Repetitive Mild TBI: Analysis of the CENC/LIMBIC Cohort Injury Characteristics.

INTRODUCTION: MRI represents one of the clinical tools at the forefront of research efforts aimed at identifying diagnostic and prognostic biomarkers following traumatic brain injury (TBI). Both volumetric and diffusion MRI findings in mild TBI (mTBI) are mixed, making the findings difficult to interpret. As such, additional research is needed to continue to elucidate the relationship between the clinical features of mTBI and quantitative MRI measurements. MATERIAL AND METHODS: Volumetric and diffusion imaging data in a sample of 976 veterans and service members from the Chronic Effects of Neurotrauma Consortium and now the Long-Term Impact of Military-Relevant Brain Injury Consortium observational study of the late effects of mTBI in combat with and without a history of mTBI were examined. A series of regression models with link functions appropriate for the model outcome were used to evaluate the relationships among imaging measures and clinical features of mTBI. Each model included acquisition site, participant sex, and age as covariates. Separate regression models were fit for each region of interest where said region was a predictor. RESULTS: After controlling for multiple comparisons, no significant main effect was noted for comparisons between veterans and service members with and without a history of mTBI. However, blast-related mTBI were associated with volumetric reductions of several subregions of the corpus callosum compared to non-blast-related mTBI. Several volumetric (i.e., hippocampal subfields, etc.) and diffusion (i.e., corona radiata, superior longitudinal fasciculus, etc.) MRI findings were noted to be associated with an increased number of repetitive mTBIs versus. CONCLUSIONS: In deployment-related mTBI, significant findings in this cohort were only observed when considering mTBI sub-groups (blast mechanism and total number/dose). Simply comparing healthy controls and those with a positive mTBI history is likely an oversimplification that may lead to non-significant findings, even in consortium analyses.

Predicting Dimensional Antidepressant Response to Repetitive Transcranial Magnetic Stimulation using Pretreatment Resting-state Functional Connectivity.

Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for depression and has been shown to modulate resting-state functional connectivity (RSFC) of depression-relevant neural circuits. To date, however, few studies have investigated whether individual treatment-related symptom changes are predictable from pretreatment RSFC. We use machine learning to predict dimensional changes in depressive symptoms using pretreatment patterns of RSFC. We hypothesized that changes in dimensional depressive symptoms would be predicted more accurately than scale total scores. Patients with depression (n=26) underwent pretreatment RSFC MRI. Depressive symptoms were assessed with the 17-item Hamilton Depression Rating Scale (HDRS-17). Random forest regression (RFR) models were trained and tested to predict treatment-related symptom changes captured by the HDRS-17, HDRS-6 and three previously identified HDRS subscales: core mood/anhedonia (CMA), somatic disturbances, and insomnia. Changes along the CMA, HDRS-17, and HDRS-6 were predicted significantly above chance, with 9%, 2%, and 2% of out-of-sample outcome variance explained, respectively (all p<0.01). CMA changes were predicted more accurately than the HDRS-17 (p<0.05). Higher baseline global connectivity (GC) of default mode network (DMN) subregions and the somatomotor network (SMN) predicted poorer symptom reduction, while higher GC of the right dorsal attention (DAN) frontoparietal control (FPCN), and visual networks (VN) predicted reduced CMA symptoms. HDRS-17 and HDRS-6 changes were predicted with similar GC patterns. These results suggest that RSFC spanning the DMN, SMN, DAN, FPCN, and VN subregions predict dimensional changes with greater accuracy than syndromal changes following rTMS. These findings highlight the need to assess more granular clinical dimensions in therapeutic studies, particularly device neuromodulation studies, and echo earlier studies supporting that dimensional outcomes improve model accuracy.

Individual Prediction of Optimal Treatment Allocation Between Electroconvulsive Therapy or Ketamine using the Personalized Advantage Index.

Introduction: Electroconvulsive therapy (ECT) and ketamine are two effective treatments for depression with similar efficacy; however, individual patient outcomes may be improved by models that predict optimal treatment assignment. Here, we adapt the Personalized Advantage Index (PAI) algorithm using machine learning to predict optimal treatment assignment between ECT and ketamine using medical record data from a large, naturalistic patient cohort. We hypothesized that patients who received a treatment predicted to be optimal would have significantly better outcomes following treatment compared to those who received a non-optimal treatment. Methods: Data on 2526 ECT and 235 mixed IV ketamine and esketamine patients from McLean Hospital was aggregated. Depressive symptoms were measured using the Quick Inventory of Depressive Symptomatology (QIDS) before and during acute treatment. Patients were matched between treatments on pretreatment QIDS, age, inpatient status, and psychotic symptoms using a 1:1 ratio yielding a sample of 470 patients (n=235 per treatment). Random forest models were trained and predicted differential patientwise minimum QIDS scores achieved during acute treatment (min-QIDS) scores for ECT and ketamine using pretreatment patient measures. Analysis of Shapley Additive exPlanations (SHAP) values identified predictors of differential outcomes between treatments. Results: Twenty-seven percent of patients with the largest PAI scores who received a treatment predicted optimal had significantly lower min-QIDS scores compared to those who received a non-optimal treatment (mean difference=1.6, t=2.38, q<0.05, Cohen's D=0.36). Analysis of SHAP values identified prescriptive pretreatment measures. Conclusions: Patients assigned to a treatment predicted to be optimal had significantly better treatment outcomes. Our model identified pretreatment patient factors captured in medical records that can provide interpretable and actionable guidelines treatment selection.