Machine learning and phone data can improve targeting of humanitarian aid.

The COVID-19 pandemic has devastated many low- and middle-income countries, causing widespread food insecurity and a sharp decline in living standards1. In response to this crisis, governments and humanitarian organizations worldwide have distributed social assistance to more than 1.5 billion people2. Targeting is a central challenge in administering these programmes: it remains a difficult task to rapidly identify those with the greatest need given available data3,4. Here we show that data from mobile phone networks can improve the targeting of humanitarian assistance. Our approach uses traditional survey data to train machine-learning algorithms to recognize patterns of poverty in mobile phone data; the trained algorithms can then prioritize aid to the poorest mobile subscribers. We evaluate this approach by studying a flagship emergency cash transfer program in Togo, which used these algorithms to disburse millions of US dollars worth of COVID-19 relief aid. Our analysis compares outcomes-including exclusion errors, total social welfare and measures of fairness-under different targeting regimes. Relative to the geographic targeting options considered by the Government of Togo, the machine-learning approach reduces errors of exclusion by 4-21%. Relative to methods requiring a comprehensive social registry (a hypothetical exercise; no such registry exists in Togo), the machine-learning approach increases exclusion errors by 9-35%. These results highlight the potential for new data sources to complement traditional methods for targeting humanitarian assistance, particularly in crisis settings in which traditional data are missing or out of date.

Transcranial Magnetic Stimulation for Posttraumatic Stress Disorder and Major Depression: Comparing Commonly Used Clinical Protocols.

Transcranial magnetic stimulation (TMS) is increasingly being used to treat posttraumatic stress disorder (PTSD) comorbid with major depressive disorder (MDD). Yet, identifying the most effective stimulation parameters remains an active area of research. We recently reported on the use of 5 Hz TMS to reduce PTSD and MDD symptoms. A recently developed form of TMS, intermittent theta burst stimulation (iTBS), appears noninferior for treating MDD. Because iTBS can be delivered in a fraction of the time, it provides significant logistical advantages; however, evaluations of whether iTBS provides comparable PTSD and MDD symptom reductions are lacking. We performed a retrospective chart review comparing clinical outcomes in veterans with PTSD and MDD who received iTBS (n = 10) with a matched cohort that received 5-Hz TMS (n = 10). Symptoms were evaluated using self-reported rating scales at baseline and every five treatments for up to 30 sessions. Both protocols were safe and reduced symptoms, ps < .001, but veterans who received iTBS reported poorer outcomes. These results were observed using mixed-model analyses, Group x Time interaction: p = .011, and effect sizes, where 5 Hz TMS demonstrated superior PTSD and MDD symptom improvement, ds = 1.81 and 1.51, respectively, versus iTBS, ds = 0.63 and 0.88, respectively. Data from prior controlled trials of iTBS, with increased stimulation exposure, have appeared to provide comparable clinical outcomes compared with 5 Hz TMS. Prospective and controlled comparisons are required; however, the present findings provide important information for clinicians using TMS to treat these commonly comorbid disorders.

Heart Rate Variability Features as Predictors of Intermittent Theta-Burst Stimulation Response in Posttraumatic Stress Disorder.

BACKGROUND: Posttraumatic stress disorder (PTSD) is associated with autonomic dysfunction as indicated by deficits in the sympathetic and parasympathetic nervous systems. These abnormalities are expressed as elevated heart rate and reduced heart rate variability (HRV), respectively. Intermittent theta-burst stimulation (iTBS), a form of transcranial magnetic stimulation, has demonstrated effectiveness in PTSD. Nevertheless, it remains unclear whether HRV may be an iTBS biomarker for PTSD and whether iTBS impacts autonomic activity. MATERIALS AND METHODS: Fifty veterans with PTSD participated in a randomized controlled trial, receiving ten daily sessions of sham-controlled iTBS (right dorsolateral prefrontal cortex, 1800 pulses/day, 80% active motor threshold, 9.5 min). With a usable dataset (N = 47), HRV parameters were assessed as predictors of clinical response immediately after stimulation. iTBS effects on autonomic response (mean RR interval, root mean square of successive differences [RMSSD], total power [TP], and low-frequency/high-frequency [LF/HF] ratio) were evaluated using an ultra-short approach. RESULTS: TP and RMSSD were significant predictors of acute clinical response to iTBS. Individuals with higher TP had better response to iTBS with improved symptoms on the Clinician-Administered PTSD Scale (rs = -0.58, p = 0.004), and higher functionality on the Social and Occupational Function Scale (rs = 0.43, p = 0.04). Similarly, higher RMSSD was associated with superior outcomes (rs = -0.44, p = 0.04). No other significant changes in HRV metrics were observed (p ≥ 0.05). CONCLUSIONS: Our findings indicate that autonomic activity is a potential low-cost and technically simple predictive biomarker of iTBS response in PTSD. Less autonomic dysfunction was associated with superior clinical improvements with iTBS. Future studies might consider HRV acquisition during iTBS, as well as prospective testing of these findings in patients with elevated hyperarousal.

Real-time estimation of disease activity in emerging outbreaks using internet search information.

Understanding the behavior of emerging disease outbreaks in, or ahead of, real-time could help healthcare officials better design interventions to mitigate impacts on affected populations. Most healthcare-based disease surveillance systems, however, have significant inherent reporting delays due to data collection, aggregation, and distribution processes. Recent work has shown that machine learning methods leveraging a combination of traditionally collected epidemiological information and novel Internet-based data sources, such as disease-related Internet search activity, can produce meaningful "nowcasts" of disease incidence ahead of healthcare-based estimates, with most successful case studies focusing on endemic and seasonal diseases such as influenza and dengue. Here, we apply similar computational methods to emerging outbreaks in geographic regions where no historical presence of the disease of interest has been observed. By combining limited available historical epidemiological data available with disease-related Internet search activity, we retrospectively estimate disease activity in five recent outbreaks weeks ahead of traditional surveillance methods. We find that the proposed computational methods frequently provide useful real-time incidence estimates that can help fill temporal data gaps resulting from surveillance reporting delays. However, the proposed methods are limited by issues of sample bias and skew in search query volumes, perhaps as a result of media coverage.

Virtual Reality and Transcranial Direct Current Stimulation for Posttraumatic Stress Disorder: A Randomized Clinical Trial.

Importance: Posttraumatic stress disorder (PTSD) is a common psychiatric disorder that is particularly difficult to treat in military veterans. Noninvasive brain stimulation has significant potential as a novel treatment to reduce PTSD symptoms. Objective: To test whether active transcranial direct current stimulation (tDCS) plus virtual reality (VR) is superior to sham tDCS plus VR for warzone-related PTSD. Design, Setting, and Participants: This double-blind randomized clinical trial was conducted among US military veterans enrolled from April 2018 to May 2023 at a secondary care Department of Veterans Affairs hospital and included 1- and 3-month follow-up visits. Participants included US military veterans with chronic PTSD and warzone-related exposure, recruited via referral and advertisement. Patients in psychiatric treatment had to be on a stable regimen for at least 6 weeks to be eligible for enrollment. Data were analyzed from May to September 2023. Intervention: Participants were randomly assigned to receive 2-mA anodal tDCS or sham tDCS targeted to the ventromedial prefrontal cortex, during six 25-minute sessions of standardized warzone VR exposure, delivered over 2 to 3 weeks. Main Outcomes and Measures: The co-primary outcomes were self-reported PTSD symptoms, measured via the PTSD checklist for DSM-5 (PCL-5), alongside quality of life. Other outcomes included psychophysiological arousal, clinician-assessed PTSD, depression, and social/occupational function. Results: A total of 54 participants (mean [SD] age, 45.7 [10.5] years; 51 [94%] males) were assessed, including 26 in the active tDCS group and 28 in the sham tDCS group. Participants in the active tDCS group reported a superior reduction in self-reported PTSD symptom severity at 1 month (t = -2.27, P = .02; Cohen d = -0.82). There were no significant differences in quality of life between active and sham tDCS groups. Active tDCS significantly accelerated psychophysiological habituation to VR events between sessions compared with sham tDCS (F5,7689.8 = 4.65; P < .001). Adverse effects were consistent with the known safety profile of the corresponding interventions. Conclusions and Relevance: These findings suggest that combined tDCS plus VR may be a promising strategy for PTSD reduction and underscore the innovative potential of these combined technologies. Trial Registration: ClinicalTrials.gov Identifier: NCT03372460.

Toward the use of neural networks for influenza prediction at multiple spatial resolutions.

Mitigating the effects of disease outbreaks with timely and effective interventions requires accurate real-time surveillance and forecasting of disease activity, but traditional health care-based surveillance systems are limited by inherent reporting delays. Machine learning methods have the potential to fill this temporal "data gap," but work to date in this area has focused on relatively simple methods and coarse geographic resolutions (state level and above). We evaluate the predictive performance of a gated recurrent unit neural network approach in comparison with baseline machine learning methods for estimating influenza activity in the United States at the state and city levels and experiment with the inclusion of real-time Internet search data. We find that the neural network approach improves upon baseline models for long time horizons of prediction but is not improved by real-time internet search data. We conduct a thorough analysis of feature importances in all considered models for interpretability purposes.

Intermittent Theta Burst Stimulation in Veterans with Mild Alcohol Use Disorder.

BACKGROUND: Alcohol use disorder (AUD) is highly comorbid with depression and posttraumatic stress disorder (PTSD) and can complicate their treatment. Transcranial magnetic stimulation is a promising treatment for these disorders, yet prior research often excluded AUD patients out of concern for safety or poorer outcomes. To this end, we revisited a prior study of intermittent theta burst stimulation (iTBS) for PTSD, to evaluate whether mild AUD impacted safety and clinical outcomes. METHODS: Fifty veterans with PTSD (n=17, with comorbid AUD) received 10 days of sham-controlled iTBS, followed by 10 unblinded sessions. Stimulation was delivered at 80% of the motor threshold for 1800 pulses to the right dorsolateral prefrontal cortex. Safety, PTSD and depressive outcomes were evaluated with repeated measures analysis of variance, to examine the effects of time, treatment group and comorbid AUD. RESULTS: iTBS was safe, although AUD patients reported more adverse events, regardless of whether they received active or sham stimulation. Regarding clinical outcomes, patients with AUD who received active stimulation demonstrated a greater rate of improvement in depression symptoms than those without comorbid AUD. The presence of AUD did not impact PTSD symptom change. LIMITATIONS: Limitations include a modest sample size and use of a categorical, rather than continuous, index of AUD diagnosis. CONCLUSION: While these results require replication, they indicate that iTBS is likely safe in patients with mild comorbid AUD. We propose that comorbid AUD should not preclude clinical use of iTBS, and that iTBS should be further investigated as a novel treatment option for AUD.

One-year clinical outcomes following theta burst stimulation for post-traumatic stress disorder.

Theta burst transcranial magnetic stimulation (TBS) is a potential new treatment for post-traumatic stress disorder (PTSD). We previously reported active intermittent TBS (iTBS) was associated with superior clinical outcomes for up to 1-month, in a sample of fifty veterans with PTSD, using a crossover design. In that study, participants randomized to the active group received a total of 4-weeks of active iTBS, or 2-weeks if randomized to sham. Results were superior with greater exposure to active iTBS, which raised the question of whether observed effects persisted over the longer-term. This study reviewed naturalistic outcomes up to 1-year from study endpoint, to test the hypothesis that greater exposure to active iTBS would be associated with superior outcomes. The primary outcome measure was clinical relapse, defined as any serious adverse event (e.g., suicide, psychiatric hospitalization, etc.,) or need for retreatment with repetitive transcranial magnetic stimulation (rTMS). Forty-six (92%) of the initial study's intent-to-treat participants were included. Mean age was 51.0 ± 12.3 years and seven (15.2%) were female. The group originally randomized to active iTBS (4-weeks active iTBS) demonstrated superior outcomes at one year compared to those originally randomized to sham (2-weeks active iTBS); log-rank ChiSq = 5.871, df = 1, p = 0.015; OR = 3.50, 95% CI = 1.04-11.79. Mean days to relapse were 296.0 ± 22.1 in the 4-week group, and 182.0 ± 31.9 in the 2-week group. When used, rTMS retreatment was generally effective. Exploratory neuroimaging revealed default mode network connectivity was predictive of 1-year outcomes (corrected p < 0.05). In summary, greater accumulated exposure to active iTBS demonstrated clinically meaningful improvements in the year following stimulation, and default mode connectivity could be used to predict longer-term outcomes.

Mapping PTSD symptoms to brain networks: a machine learning study.

Posttraumatic Stress Disorder (PTSD) is a prevalent and debilitating condition with complex and variable presentation. While PTSD symptom domains (intrusion, avoidance, cognition/mood, and arousal/reactivity) correlate highly, the relative importance of these symptom subsets often differs across patients. In this study, we used machine learning to derive how PTSD symptom subsets differ based upon brain functional connectivity. We acquired resting-state magnetic resonance imaging in a sample (N = 50) of PTSD patients and characterized clinical features using the PTSD Checklist for DSM-5 (PCL-5). We compared connectivity among 100 cortical and subcortical regions within the default mode, salience, executive, and affective networks. We then used principal component analysis and least-angle regression (LARS) to identify relationships between symptom domain severity and brain networks. We found connectivity predicted PTSD symptom profiles. The goodness of fit (R2) for total PCL-5 score was 0.29 and the R2 for intrusion, avoidance, cognition/mood, and arousal/reactivity symptoms was 0.33, 0.23, -0.01, and 0.06, respectively. The model performed significantly better than chance in predicting total PCL-5 score (p = 0.030) as well as intrusion and avoidance scores (p = 0.002 and p = 0.034). It was not able to predict cognition and arousal scores (p = 0.412 and p = 0.164). While this work requires replication, these findings demonstrate that this computational approach can directly link PTSD symptom domains with neural network connectivity patterns. This line of research provides an important step toward data-driven diagnostic assessments in PTSD, and the use of computational methods to identify individual patterns of network pathology that can be leveraged toward individualized treatment.

Neuroimaging Correlates of Suicidality in Decision-Making Circuits in Posttraumatic Stress Disorder.

In depression, brain and behavioral correlates of decision-making differ between individuals with and without suicidal thoughts and behaviors. Though promising, it remains unknown if these potential biomarkers of suicidality will generalize to other high-risk clinical populations. To preliminarily assess whether brain structure or function tracked suicidality in individuals with posttraumatic stress disorder (PTSD), we measured resting-state functional connectivity and cortical thickness in two functional networks involved in decision-making, a ventral fronto-striatal reward network and a lateral frontal cognitive control network. Neuroimaging data and self-reported suicidality ratings, and suicide-related hospitalization data were obtained from 50 outpatients with PTSD and also from 15 healthy controls, and all were subjected to seed-based resting-state functional connectivity and cortical thickness analyses using a priori seeds from reward and cognitive control networks. First, general linear models (GLM) were used to evaluate whether ROI-to-ROI functional connectivity was predictive of self-reported suicidality after false discovery rate (FDR)-correction for multiple comparisons and covariance of age and depression symptoms. Next, regional cortical thickness statistics were included as predictors of ROI-to-ROI functional connectivity in follow-up GLMs evaluating structure-function relationships. Functional connectivity between reward regions was positively correlated with suicidality (p-FDR ≤ 0.05). Functional connectivity of the lateral pars orbitalis to anterior cingulate/paracingulate control regions also tracked suicidality (p-FDR ≤ 0.05). Furthermore, cortical thickness in anterior cingulate/paracingulate was associated with functional correlates of suicidality in the control network (p-FDR < 0.05). These results provide a preliminary demonstration that biomarkers of suicidality in decision-making networks observed in depression may generalize to PTSD and highlight the promise of these circuits as transdiagnostic biomarkers of suicidality.

Theta-Burst Transcranial Magnetic Stimulation for Posttraumatic Stress Disorder.

OBJECTIVE: Posttraumatic stress disorder (PTSD) is a highly prevalent psychiatric disorder associated with disruption in social and occupational function. Transcranial magnetic stimulation (TMS) represents a novel approach to PTSD, and intermittent theta-burst stimulation (iTBS) is a new, more rapid administration protocol with data supporting efficacy in depression. The authors conducted a sham-controlled study of iTBS for PTSD. METHODS: Fifty veterans with PTSD received 10 days of sham-controlled iTBS (1,800 pulses/day), followed by 10 unblinded sessions. Primary outcome measures included acceptability (retention rates), changes in PTSD symptoms (clinician- and self-rated), quality of life, social and occupational function, and depression, obtained at the end of 2 weeks; analysis of variance was used to compare active with sham stimulation. Secondary outcomes were evaluated 1 month after treatment, using mixed-model analyses. Resting-state functional MRI was acquired at pretreatment baseline on an eligible subset of participants (N=26) to identify response predictors. RESULTS: Retention was high, side effects were consistent with standard TMS, and blinding was successful. At 2 weeks, active iTBS was significantly associated with improved social and occupational function (Cohen's d=0.39); depression was improved with iTBS compared with the sham treatment (d=-0.45), but the difference fell short of significance, and moderate nonsignificant effect sizes were observed on self-reported PTSD symptoms (d=-0.34). One-month outcomes, which incorporated data from the unblinded phase of the study, indicated superiority of active iTBS on clinician- and self-rated PTSD symptoms (d=-0.74 and -0.63, respectively), depression (d=-0.47), and social and occupational function (d=0.93) (all significant). Neuroimaging indicated that clinical improvement was significantly predicted by stronger (greater positive) connectivity within the default mode network and by anticorrelated (greater negative) cross-network connectivity. CONCLUSIONS: iTBS appears to be a promising new treatment for PTSD. Most clinical improvements from stimulation occurred early, which suggests a need for further investigation of optimal iTBS time course and duration. Consistent with previous neuroimaging studies of TMS, default mode network connectivity played an important role in response prediction.

A review of prospective memory in individuals with acquired brain injury.

OBJECTIVE: Prospective memory (PM) deficits have emerged as an important predictor of difficulty in daily life for individuals with acquired brain injury (BI). This review examines the variables that have been found to influence PM performance in this population. In addition, current methods of assessment are reviewed with a focus on clinical measures. Finally, cognitive rehabilitation therapies are reviewed, including compensatory, restorative and metacognitive approaches. METHOD: Preferred reporting items for systematic reviews and meta-analyses guidelines were used to identify studies. Studies were added that were identified from the reference lists of these. RESULTS: Research has begun to elucidate the contributing variables to PM deficits after BI, such as attention, executive function and retrospective memory components. Imaging studies have identified prefrontal deficits, especially in the region of BA10 as contributing to these deficits. There are now several clinical measures available with good psychometric properties. Rehabilitation techniques have mostly focused on compensatory strategies, but, in addition, some restorative and metacognitive approaches have shown preliminary promise. CONCLUSIONS: PM deficits are a common and important deficit after BI. Clinical evaluation is recommended and further understanding of rehabilitation techniques is needed.

Neuroimaging Mechanisms of Therapeutic Transcranial Magnetic Stimulation for Major Depressive Disorder.

Research into therapeutic transcranial magnetic stimulation (TMS) for major depression has dramatically increased in the last decade. Understanding the mechanism of action of TMS is crucial to improve efficacy and develop the next generation of therapeutic stimulation. Early imaging research provided initial data supportive of widely held assumptions about hypothesized inhibitory or excitatory consequences of stimulation. Early work also indicated that while TMS modulated brain activity under the stimulation site, effects at deeper regions, in particular, the subgenual anterior cingulate cortex, were associated with clinical improvement. Concordant with earlier findings, functional connectivity studies also demonstrated that clinical improvements were related to changes distal, rather than proximal, to the site of stimulation. Moreover, recent work suggests that TMS modulates and potentially normalizes functional relationships between neural networks. An important observation that emerged from this review is that similar patterns of connectivity changes are observed across studies regardless of TMS parameters. Though promising, we stress that these imaging findings must be evaluated cautiously given the widespread reliance on modest sample sizes and little implementation of statistical validation. Additional limitations included use of imaging before and after a course of TMS, which provided little insight into changes that might occur during the weeks of stimulation. Furthermore, as studies to date have focused on depression, it is unclear whether our observations were related to mechanisms of action of TMS for depression or represented broader patterns of functional brain changes associated with clinical improvement.

Network Functional Architecture and Aberrant Functional Connectivity in Post-Traumatic Stress Disorder: A Clinical Application of Network Convergence.

Posttraumatic stress disorder (PTSD) is associated with disrupted functional connectivity in multiple neural networks. Multinetwork models of PTSD hypothesize that aberrant regional connectivity emerges from broad network-level disruptions. However, few studies have tested how characteristics of network-level organization influence regional functional connectivity in PTSD. This gap in knowledge impacts both our understanding of the pathophysiology of the disorder and the development of network-targeted PTSD treatments. We acquired resting-state imaging from a naturalistic sample of patients with PTSD (n = 42) and healthy controls (n = 42). Group differences in functional connectivity were identified using region of interest analyses and estimations of within- and between neural network activity; PTSD patients demonstrated reduced amygdala-orbitofrontal connectivity and increased default mode network (DMN) connectivity compared with controls. We then used convergence-a novel measure representing the capacity for functional integration-to test whether differences in functional architecture underlie connectivity signatures of PTSD. This approach found that reduced frontoparietal network (FPN) convergence was associated with reduced amygdala-orbitofrontal connectivity. Furthermore, in controls only, increased DMN convergence was associated with reduced DMN-to-salience network connectivity, and increased FPN convergence was associated with reduced FPN-to-ventral attention network connectivity. These results suggest that FPN functional architecture may underlie insufficiencies in top-down control in PTSD, with results broadly supporting the notion that networks' functional architecture influences the breakdown of normative functional relationships in PTSD. This work also indicates the potential of convergence to be applied to clinical populations in future research studies.