A machine learning ensemble to predict treatment outcomes following an Internet intervention for depression

Things You Do: A randomized controlled trial of an unguided ultra-brief intervention to reduce symptoms of depression and anxiety

The 'Things You Do' encompass five types of actions that are strongly associated with good mental health: Healthy Thinking, Meaningful Activities, Goals and Plans, Healthy Habits, and Social Connections. Ultra-brief interventions which increase how often people perform these actions may decrease depression and anxiety. A two-arm randomized controlled trial (N = 349) compared an unguided ultra-brief intervention based on the 'Things You Do' against a waitlist control. The intervention included one online module, two practice guides, and four weeks of daily text messages. The primary timepoint was 5-weeks post-baseline. The intervention resulted in moderate reductions in depression (d = 0.51) and anxiety (d = 0.55) alongside moderate increases in the frequency of Things You Do actions (d = 0.54), compared to controls. No significant change in number of days out of role or life satisfaction were observed. Treatment completion was high (92 %), most participants reported being satisfied with the treatment (66 %), and improvements were maintained at 3-month follow-up. This study demonstrated that an automated ultra-brief 'Things You Do' intervention resulted in clinically significant reductions in depression and anxiety. Ultra-brief interventions may provide a scalable solution to support individuals who are unlikely to engage in longer forms of psychological treatment.

Select or adjust? How information from early treatment stages boosts the prediction of non-response in internet-based depression treatment

BACKGROUND

Internet-based interventions produce comparable effectiveness rates as face-to-face therapy in treating depression. Still, more than half of patients do not respond to treatment. Machine learning (ML) methods could help to overcome these low response rates by predicting therapy outcomes on an individual level and tailoring treatment accordingly. Few studies implemented ML algorithms in internet-based depression treatment using baseline self-report data, but differing results hinder inferences on clinical practicability. This work compares algorithms using features gathered at baseline or early in treatment in their capability to predict non-response to a 6-week online program targeting depression.

METHODS

Our training and test sample encompassed 1270 and 318 individuals, respectively. We trained random forest algorithms on self-report and process features gathered at baseline and after 2 weeks of treatment. Non-responders were defined as participants not fulfilling the criteria for reliable and clinically significant change on PHQ-9 post-treatment. Our benchmark models were logistic regressions trained on baseline PHQ-9 sum or PHQ-9 early change, using 100 iterations of randomly sampled 80/20 train-test-splits.

RESULTS

Best performances were reached by our models involving early treatment characteristics (recall: 0.75-0.76; AUC: 0.71-0.77). Therapeutic alliance and early symptom change constituted the most important predictors. Models trained on baseline data were not significantly better than our benchmark.

CONCLUSIONS

Fair accuracies were only attainable by involving information from early treatment stages. In-treatment adaptation, instead of a priori selection, might constitute a more feasible approach for improving response when relying on easily accessible self-report features. Implementation trials are needed to determine clinical usefulness.

Examining the role of AI technology in online mental healthcare: opportunities, challenges, and implications, a mixed-methods review

Introduction

Online mental healthcare has gained significant attention due to its effectiveness, accessibility, and scalability in the management of mental health symptoms. Despite these advantages over traditional in-person formats, including higher availability and accessibility, issues with low treatment adherence and high dropout rates persist. Artificial intelligence (AI) technologies could help address these issues, through powerful predictive models, language analysis, and intelligent dialogue with users, however the study of these applications remains underexplored. The following mixed methods review aimed to supplement this gap by synthesizing the available evidence on the applications of AI in online mental healthcare.

Method

We searched the following databases: MEDLINE, CINAHL, PsycINFO, EMBASE, and Cochrane. This review included peer-reviewed randomized controlled trials, observational studies, non-randomized experimental studies, and case studies that were selected using the PRISMA guidelines. Data regarding pre and post-intervention outcomes and AI applications were extracted and analyzed. A mixed-methods approach encompassing meta-analysis and network meta-analysis was used to analyze pre and post-intervention outcomes, including main effects, depression, anxiety, and study dropouts. We applied the Cochrane risk of bias tool and the Grading of Recommendations Assessment, Development and Evaluation (GRADE) to assess the quality of the evidence.

Results

Twenty-nine studies were included revealing a variety of AI applications including triage, psychotherapy delivery, treatment monitoring, therapy engagement support, identification of effective therapy features, and prediction of treatment response, dropout, and adherence. AI-delivered self-guided interventions demonstrated medium to large effects on managing mental health symptoms, with dropout rates comparable to non-AI interventions. The quality of the data was low to very low.

Discussion

The review supported the use of AI in enhancing treatment response, adherence, and improvements in online mental healthcare. Nevertheless, given the low quality of the available evidence, this study highlighted the need for additional robust and high-powered studies in this emerging field.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=443575, identifier CRD42023443575.

Which client with generalized anxiety disorder benefits from a mindfulness ecological momentary intervention versus a self-monitoring app? Developing a multivariable machine learning predictive model

Precision medicine methods (machine learning; ML) can identify which clients with generalized anxiety disorder (GAD) benefit from mindfulness ecological momentary intervention (MEMI) vs. self-monitoring app (SM). We used randomized controlled trial data of MEMI vs. SM for GAD (N = 110) and tested three ML models to predict one-month follow-up reliable improvement in GAD severity, perseverative cognitions (PC), trait mindfulness (TM), and executive function (EF). Eleven baseline predictors were tested regarding differential reliable change from MEMI vs. SM (age, sex, race, EF errors, inhibitory dyscontrol, set-shifting deficits, verbal fluency, working memory, GAD severity, TM, PC). The final top five prescriptive predictor models of all outcomes performed well (AUC = .752 .886). The following variables predicted better outcome from MEMI vs. SM: Higher GAD severity predicted more GAD improvement but less EF improvement. Elevated PC, inhibitory dyscontrol, and verbal dysfluency predicted better improvement in most outcomes. Greater set-shifting and TM predicted stronger improvements in GAD symptoms and TM. Older age predicted more alleviation of GAD and PC symptoms. Women exhibited more enhancements in trait mindfulness and EF than men. White individuals benefitted more than non-White. PC, TM, EF, and sociodemographic data might help predictive models optimize intervention selection for GAD.

Does outcome expectancy predict outcomes in online depression prevention? Secondary analysis of randomised-controlled trials

BACKGROUND

Evidence shows that online interventions could prevent depression. However, to improve the effectiveness of preventive online interventions in individuals with subthreshold depression, it is worthwhile to study factors influencing intervention outcomes. Outcome expectancy has been shown to predict treatment outcomes in psychotherapy for depression. However, little is known about whether this also applies to depression prevention. The aim of this study was to investigate the role of participants' outcome expectancy in an online depression prevention intervention.

METHODS

A secondary data analysis was conducted using data from two randomised-controlled trials (N = 304). Multilevel modelling was used to explore the effect of outcome expectancy on depressive symptoms and close-to-symptom-free status postintervention (6-7 weeks) and at follow-up (3-6 months). In a subsample (n = 102), Cox regression was applied to assess the effect on depression onset within 12 months. Explorative analyses included baseline characteristics as possible moderators. Outcome expectancy did not predict posttreatment outcomes or the onset of depression.

RESULTS

Small effects were observed at follow-up for depressive symptoms (β = -.39, 95% confidence interval [CI]: [-0.75, -0.03], p = .032, padjusted = .130) and close-to-symptom-free status (relative risk = 1.06, 95% CI: [1.01, 1.11], p = .013, padjusted = 0.064), but statistical significance was not maintained when controlling for multiple testing. Moderator analyses indicated that expectancy could be more influential for females and individuals with higher initial symptom severity.

CONCLUSION

More thoroughly designed, predictive studies targeting outcome expectancy are necessary to assess the full impact of the construct for effective depression prevention.

PATIENT OR PUBLIC CONTRIBUTION

This secondary analysis did not involve patients, service users, care-givers, people with lived experience or members of the public. However, the findings incorporate the expectations of participants using the preventive online intervention, and these exploratory findings may inform the future involvement of participants in the design of indicated depression prevention interventions for adults.

CLINICAL TRIAL REGISTRATION

Original studies: DRKS00004709, DRKS00005973; secondary analysis: osf.io/9xj6a.

A prognostic index to predict symptom and functional outcomes of a coached, web-based intervention for trauma-exposed veterans

Researchers at the Department of Veterans Affairs (VA) have studied interventions for posttraumatic stress disorder and co-occurring conditions in both traditional and digital formats. One such empirically supported intervention is web skills training in affective and interpersonal regulation (webSTAIR), a coached, 10-module web program based on STAIR. To understand which patient characteristics were predictive of webSTAIR outcomes in a sample of trauma-exposed veterans (N = 189), we used machine learning (ML) to develop a prognostic index from among 18 baseline characteristics (i.e., demographic, military, trauma history, and clinical) to predict posttreatment posttraumatic stress disorder severity, depression severity, and psychosocial functioning impairment. We compared the ML models to a benchmark of linear regression models in which the only predictor was the baseline severity score of the outcome measure. The ML and "severity-only" models performed similarly, explaining 39%-45% of the variance in outcomes. This suggests that baseline symptom severity and functioning are strong indicators for webSTAIR outcomes in veterans, with higher severity indicating worse prognosis, and that the other variables examined did not contribute significant added predictive signal. Findings also highlight the importance of comparing ML models to an appropriate benchmark. Future research with larger samples could potentially detect smaller patient-level effects as well as effects driven by other types of variables (e.g., therapeutic process variables). As a transdiagnostic, digital intervention, webSTAIR can potentially serve a diverse veteran population with varying trauma histories and may be best conceptualized as a beneficial first step of a stepped care model for those with heightened symptoms or impairment. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Supervised machine learning: A new method to predict the outcomes following exercise intervention in children with autism spectrum disorder

The individual differences among children with autism spectrum disorder (ASD) may make it challenging to achieve comparable benefits from a specific exercise intervention program. A new method for predicting the possible outcomes and maximizing the benefits of exercise intervention for children with ASD needs further exploration. Using the mini-basketball training program (MBTP) studies to improve the symptom performance of children with ASD as an example, we used the supervised machine learning method to predict the possible intervention outcomes based on the individual differences of children with ASD, investigated and validated the efficacy of this method. In a long-term study, we included 41 ASD children who received the MBTP. Before the intervention, we collected their clinical information, behavioral factors, and brain structural indicators as candidate factors. To perform the regression and classification tasks, the random forest algorithm from the supervised machine learning method was selected, and the cross validation method was used to determine the reliability of the prediction results. The regression task was used to predict the social communication impairment outcome following the MBTP in children with ASD, and explainable variance was used to evaluate the predictive performance. The classification task was used to distinguish the core symptom outcome groups of ASD children, and predictive performance was assessed based on accuracy. We discovered that random forest models could predict the outcome of social communication impairment (average explained variance was 30.58%) and core symptom (average accuracy was 66.12%) following the MBTP, confirming that the supervised machine learning method can predict exercise intervention outcomes for children with ASD. Our findings provide a novel and reliable method for identifying ASD children most likely to benefit from a specific exercise intervention program in advance and a solid foundation for establishing a personalized exercise intervention program recommendation system for ASD children.

Depression prediction based on LassoNet-RNN model: A longitudinal study

Depression has become a widespread health concern today. Understanding the influencing factors can promote human mental health as well as provide a basis for exploring preventive measures. Combining LassoNet with recurrent neural network (RNN), this study constructed a screening model ,LassoNet-RNN, for identifying influencing factors of individual depression. Based on multi-wave surveys of China Health and Retirement Longitudinal Study (CHARLS) dataset (11,661 observations), we analyzed the multivariate time series data and recognized 27 characteristic variables selected from four perspectives: demographics, health-related risk factors, household economic status, and living environment. Additionally, the importance rankings of the characteristic variables were obtained. These results offered insightful recommendations for theoretical developments and practical decision making in public health.

Predicting Individual Response to a Web-Based Positive Psychology Intervention: A Machine Learning Approach

Positive psychology interventions (PPIs) are effective at increasing happiness and decreasing depressive symptoms. PPIs are often administered as self-guided web-based interventions, but not all persons benefit from web-based interventions. Therefore, it is important to identify whether someone is likely to benefit from web-based PPIs, in order to triage persons who may not benefit from other interventions. In the current study, we used machine learning to predict individual response to a web-based PPI, in order to investigate baseline prognostic indicators of likelihood of response (N = 120). Our models demonstrated moderate correlations (happiness: r Test = 0.30 ± 0.09; depressive symptoms: r Test = 0.39 ± 0.06), indicating that baseline features can predict changes in happiness and depressive symptoms at a 6-month follow-up. Thus, machine learning can be used to predict outcome changes from a web-based PPI and has important clinical implications for matching individuals to PPIs based on their individual characteristics.

Development of a model to predict psychotherapy response for depression among Veterans

BACKGROUND

Fewer than half of patients with major depressive disorder (MDD) respond to psychotherapy. Pre-emptively informing patients of their likelihood of responding could be useful as part of a patient-centered treatment decision-support plan.

METHODS

This prospective observational study examined a national sample of 807 patients beginning psychotherapy for MDD at the Veterans Health Administration. Patients completed a self-report survey at baseline and 3-months follow-up (data collected 2018-2020). We developed a machine learning (ML) model to predict psychotherapy response at 3 months using baseline survey, administrative, and geospatial variables in a 70% training sample. Model performance was then evaluated in the 30% test sample.

RESULTS

32.0% of patients responded to treatment after 3 months. The best ML model had an AUC (SE) of 0.652 (0.038) in the test sample. Among the one-third of patients ranked by the model as most likely to respond, 50.0% in the test sample responded to psychotherapy. In comparison, among the remaining two-thirds of patients, <25% responded to psychotherapy. The model selected 43 predictors, of which nearly all were self-report variables.

CONCLUSIONS

Patients with MDD could pre-emptively be informed of their likelihood of responding to psychotherapy using a prediction tool based on self-report data. This tool could meaningfully help patients and providers in shared decision-making, although parallel information about the likelihood of responding to alternative treatments would be needed to inform decision-making across multiple treatments.

Predicting dropout from psychological treatment using different machine learning algorithms, resampling methods, and sample sizes

OBJECTIVE

The occurrence of dropout from psychological interventions is associated with poor treatment outcome and high health, societal and economic costs. Recently, machine learning (ML) algorithms have been tested in psychotherapy outcome research. Dropout predictions are usually limited by imbalanced datasets and the size of the sample. This paper aims to improve dropout prediction by comparing ML algorithms, sample sizes and resampling methods.

METHOD

Twenty ML algorithms were examined in twelve subsamples (drawn from a sample of N = 49,602) using four resampling methods in comparison to the absence of resampling and to each other. Prediction accuracy was evaluated in an independent holdout dataset using the F₁-Measure.

RESULTS

Resampling methods improved the performance of ML algorithms and down-sampling can be recommended, as it was the fastest method and as accurate as the other methods. For the highest mean F₁-Score of .51 a minimum sample size of N = 300 was necessary. No specific algorithm or algorithm group can be recommended.

CONCLUSION

Resampling methods could improve the accuracy of predicting dropout in psychological interventions. Down-sampling is recommended as it is the least computationally taxing method. The training sample should contain at least 300 cases.

The development and internal evaluation of a predictive model to identify for whom Mindfulness-Based Cognitive Therapy (MBCT) offers superior relapse prevention for recurrent depression versus maintenance antidepressant medication

Depression is highly recurrent, even following successful pharmacological and/or psychological intervention. We aimed to develop clinical prediction models to inform adults with recurrent depression choosing between antidepressant medication (ADM) maintenance or switching to Mindfulness-Based Cognitive Therapy (MBCT). Using data from the PREVENT trial (N=424), we constructed prognostic models using elastic net regression that combined demographic, clinical and psychological factors to predict relapse at 24 months under ADM or MBCT. Only the ADM model (discrimination performance: AUC=.68) predicted relapse better than baseline depression severity (AUC=.54; one-tailed DeLong's test: z=2.8, p=.003). Individuals with the poorest ADM prognoses who switched to MBCT had better outcomes compared to those who maintained ADM (48% vs. 70% relapse, respectively; superior survival times [z=-2.7, p=.008]). For individuals with moderate-to-good ADM prognosis, both treatments resulted in similar likelihood of relapse. If replicated, the results suggest that predictive modeling can inform clinical decision-making around relapse prevention in recurrent depression.

CO2 reactivity as a biomarker of exposure-based therapy non-response: study protocol

BACKGROUND

Exposure-based therapy is an effective first-line treatment for anxiety-, obsessive-compulsive, and trauma- and stressor-related disorders; however, many patients do not improve, resulting in prolonged suffering and poorly used resources. Basic research on fear extinction may inform the development of a biomarker for the selection of exposure-based therapy. Growing evidence links orexin system activity to deficits in fear extinction and we have demonstrated that reactivity to an inhaled carbon dioxide (CO₂) challenge-a safe, affordable, and easy-to-implement procedure-can serve as a proxy for orexin system activity and predicts fear extinction deficits in rodents. Building upon this basic research, the goal for the proposed study is to validate CO₂ reactivity as a biomarker of exposure-based therapy non-response.

METHODS

We will assess CO₂ reactivity in 600 adults meeting criteria for one or more fear- or anxiety-related disorders prior to providing open exposure-based therapy. By incorporating CO₂ reactivity into a multivariate model predicting treatment non-response that also includes reactivity to hyperventilation as well as a number of related predictor variables, we will establish the mechanistic specificity and the additive predictive utility of the potential CO₂ reactivity biomarker. By developing models independently within two study sites (University of Texas at Austin and Boston University) and predicting the other site's data, we will validate that the results are likely to generalize to future clinical samples.

DISCUSSION

Representing a necessary stage in translating basic research, this investigation addresses an important public health issue by testing an accessible clinical assessment strategy that may lead to a more effective treatment selection (personalized medicine) for patients with anxiety- and fear-related disorders, and enhanced understanding of the mechanisms governing exposure-based therapy.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT05467683 (20/07/2022).

Gradient boosting decision-tree-based algorithm with neuroimaging for personalized treatment in depression

Introduction

Pretreatment positron emission tomography (PET) with 2-deoxy-2-[18F]fluoro-D-glucose (FDG) and magnetic resonance spectroscopy (MRS) may identify biomarkers for predicting remission (absence of depression). Yet, no such image-based biomarkers have achieved clinical validity. The purpose of this study was to identify biomarkers of remission using machine learning (ML) with pretreatment FDG-PET/MRS neuroimaging, to reduce patient suffering and economic burden from ineffective trials.

Methods

This study used simultaneous PET/MRS neuroimaging from a double-blind, placebo-controlled, randomized antidepressant trial on 60 participants with major depressive disorder (MDD) before initiating treatment. After eight weeks of treatment, those with ≤ 7 on 17-item Hamilton Depression Rating Scale were designated a priori as remitters (free of depression, 37%). Metabolic rate of glucose uptake (metabolism) from 22 brain regions were acquired from PET. Concentrations (mM) of glutamine and glutamate and gamma-aminobutyric acid (GABA) in anterior cingulate cortex were quantified from MRS. The data were randomly split into 67% train and cross-validation (n = 40), and 33% test (n = 20) sets. The imaging features, along with age, sex, handedness, and treatment assignment (selective serotonin reuptake inhibitor or SSRI vs. placebo) were entered into the eXtreme Gradient Boosting (XGBoost) classifier for training.

Results

In test data, the model showed 62% sensitivity, 92% specificity, and 77% weighted accuracy. Pretreatment metabolism of left hippocampus from PET was the most predictive of remission.

Conclusions

The pretreatment neuroimaging takes around 60 minutes but has potential to prevent weeks of failed treatment trials. This study effectively addresses common issues for neuroimaging analysis, such as small sample size, high dimensionality, and class imbalance.

Machine Learning on Early Diagnosis of Depression

To review the recent progress of machine learning for the early diagnosis of depression (major depressive disorder). The source of data was 32 original studies in the Web of Science. The search terms were "depression" (title) and "random forest" (abstract). The eligibility criteria were the dependent variable of depression, the interventions of machine learning (the decision tree, the naïve Bayesian, the random forest, the support vector machine and/or the artificial neural network), the outcomes of accuracy and/or the area under the receiver operating characteristic curve (AUC) for the early diagnosis of depression, the publication year of 2000 or later, the publication language of English and the publication journal of SCIE/SSCI. Different machine learning methods would be appropriate for different types of data for the early diagnosis of depression, e.g., logistic regression, the random forest, the support vector machine and/or the artificial neural network in the case of numeric data, the random forest in the case of genomic data. Their performance measures reported varied within 60.1-100.0 for accuracy and 64.0-96.0 for the AUC. Machine learning provides an effective, non-invasive decision support system for early diagnosis of depression.

Computerized Psychological Interventions in Veterans and Service Members: Systematic Review of Randomized Controlled Trials

Background

Computerized psychological interventions can overcome logistical and psychosocial barriers to the use of mental health care in the Veterans Affairs and Department of Defense settings.

Objective

In this systematic review, we aim to outline the existing literature, with the goal of describing: the scope and quality of the available literature, intervention characteristics, study methods, study efficacy, and study limitations and potential directions for future research.

Methods

Systematic searches of two databases (PsycINFO and PubMed) using PRISMA (Preferred Reporting Item for Systematic Reviews and Meta-Analyses) guidelines were conducted from inception until November 15, 2020. The following inclusion criteria were used: the study was published in an English language peer-reviewed journal, participants were randomly allocated to a computerized psychological intervention or a control group (non–computerized psychological intervention active treatment or nonactive control group), an intervention in at least one treatment arm was primarily delivered through the computer or internet with or without additional support, participants were veterans or service members, and the study used validated measures to examine the effect of treatment on psychological outcomes.

Results

This review included 23 studies that met the predefined inclusion criteria. Most studies were at a high risk of bias. Targeted outcomes, participant characteristics, type of support delivered, adherence, and participant satisfaction were described. Most of the examined interventions (19/24, 79%) yielded positive results. Study limitations included participant characteristics limiting study inference, high rates of attrition, and an overreliance on self-reported outcomes.

Conclusions

Relatively few high-quality studies were identified, and more rigorous investigations are needed. Several recommendations for future research are discussed, including the adoption of methods that minimize attrition, optimize use, and allow for personalization of treatment.

An exploratory application of machine learning methods to optimize prediction of responsiveness to digital interventions for eating disorder symptoms

OBJECTIVE

Digital interventions show promise to address eating disorder (ED) symptoms. However, response rates are variable, and the ability to predict responsiveness to digital interventions has been poor. We tested whether machine learning (ML) techniques can enhance outcome predictions from digital interventions for ED symptoms.

METHOD

Data were aggregated from three RCTs (n = 826) of self-guided digital interventions for EDs. Predictive models were developed for four key outcomes: uptake, adherence, drop-out, and symptom-level change. Seven ML techniques for classification were tested and compared against the generalized linear model (GLM).

RESULTS

The seven ML methods used to predict outcomes from 36 baseline variables were poor for the three engagement outcomes (AUCs = 0.48-0.52), but adequate for symptom-level change (R²  = .15-.40). ML did not offer an added benefit to the GLM. Incorporating intervention usage pattern data improved ML prediction accuracy for drop-out (AUC = 0.75-0.93) and adherence (AUC = 0.92-0.99). Age, motivation, symptom severity, and anxiety emerged as influential outcome predictors.

CONCLUSION

A limited set of routinely measured baseline variables was not sufficient to detect a performance benefit of ML over traditional approaches. The benefits of ML may emerge when numerous usage pattern variables are modeled, although this validation in larger datasets before stronger conclusions can be made.

The potential of precision psychiatry: what is in reach?

Progress in developing personalised care for mental disorders is supported by numerous proof-of-concept machine learning studies in the area of risk assessment, diagnostics and precision prescribing. Most of these studies primarily use clinical data, but models might benefit from additional neuroimaging, blood and genetic data to improve accuracy. Combined, multimodal models might offer potential for stratification of patients for treatment. Clinical implementation of machine learning is impeded by a lack of wider generalisability, with efforts primarily focused on psychosis and dementia. Studies across all diagnostic groups should work to test the robustness of machine learning models, which is an essential first step to clinical implementation, and then move to prospective clinical validation. Models need to exceed clinicians' heuristics to be useful, and safe, in routine decision-making. Engagement of clinicians, researchers and patients in digitalisation and 'big data' approaches are vital to allow the generation and accessibility of large, longitudinal, prospective data needed for precision psychiatry to be applied into real-world psychiatric care.

Using Smartphone App Use and Lagged-Ensemble Machine Learning for the Prediction of Work Fatigue and Boredom

INTRO

As smartphone usage becomes increasingly prevalent in the workplace, the physical and psychological implications of this behavior warrant consideration. Recent research has investigated associations between workplace smartphone use and fatigue and boredom, yet findings are not conclusive.

METHODS

To build off recent efforts, we applied an ensemble machine learning model on a previously published dataset of N = 83 graduate students in the Netherlands to predict work boredom and fatigue from passively collected smartphone app use information. Using time-based feature engineering and lagged variations of the data to train, validate, and test idiographic models, we evaluated the efficacy of a lagged-ensemble predictive paradigm on sparse temporal data. Moreover, we probed the relative importance of both derived app use variables and lags within this predictive framework.

RESULTS

The ability to predict fatigue and boredom trajectories from app use information was heterogeneous and highly person-specific. Idiographic modeling reflected moderate to high correlative capacity (r > 0.4) in 47% of participants for fatigue and 24% for boredom, with better overall performance in the fatigue prediction task. App use relating to duration, communication, and patterns of use frequency were among the most important features driving predictions across lags, with longer lags contributing more heavily to final ensemble predictions compared with shorter ones.

CONCLUSION

A lag- specific ensemble predictive paradigm is a promising approach to leveraging high-dimensional app use behavioral data for the prediction of work fatigue and boredom. Future research will benefit from evaluating associations on densely collected data across longer time scales.

Predictors of Treatment Response to an Internet-Delivered Intervention Targeting Residual Cognitive Symptoms After Major Depressive Disorder

OBJECTIVE

Residual cognitive symptoms after depression are common and associated with reduced daily life functioning and an increased risk of depression relapse. There is a lack of knowledge on treatments targeting residual cognitive symptoms after major depressive disorder (MDD), including the factors associated with treatment response. The aim of the current study is to explore factors of treatment response to a guided internet-delivered intervention for former depressed adults experiencing residual cognitive symptoms.

METHOD

Forty-three former depressed adults with residual cognitive symptoms were included. Linear mixed model analyses were used to investigate the impact of pre-treatment demographic-, illness, and symptom variables, and therapy process variables, such as credibility, expectancy, and user behavior, on reduction in residual cognitive symptoms from pre-treatment to 6-month follow-up.

RESULTS

Having had MDD for a year or less predicted more reductions in residual cognitive symptoms from pre- to 6-month follow-up. Higher levels of perceived treatment credibility and expectancy evaluated in the early course of treatment did also predict a positive treatment response. No demographic-, symptom-variables, previous number of episodes with MDD, and user behavior were associated with change in residual cognitive symptoms.

CONCLUSION

This study suggests that individuals with shorter duration of previous depressions might have larger reductions in residual cognitive symptoms at 6-month follow-up compared to those with a longer duration of depression. Treatment credibility and expectancy also predicted treatment response and effort should also be made to ensure interventions credibility. Results should be interpreted with caution due to the study having a low sample size. Further investigation of predictors should be conducted in a full scale randomized controlled trial.

Predictors of adherence to an internet-based cognitive behavioral therapy program for individuals with chronic pain and comorbid psychological distress

BACKGROUND

The burden caused by chronic pain is significant, affecting at least 10 percent of the world´s population. While internet-based treatments based on cognitive behavioral therapy (CBT) have been shown to be promising in this area, attrition levels vary significantly. The purpose of this study was to investigate predictor variables for participants' adherence to an internet-based CBT treatment for individuals with chronic pain as well as to investigate associations between adherence and treatment outcome.

METHODS

Data for this study was retrieved from a randomized controlled trial including 95 individuals with chronic pain who received internet-based CBT. Treatment adherence was studied through three outcome variables: treatment progress, treatment completion and exercise completion. The predictor variables were grouped into four clusters: background variables (age, gender, marital status, level of education, and typical computer usage); the second cluster included health status variables (sick leave, current psychiatric diagnosis, previous psychotherapy for pain, current pharmacological treatment, previous depression, current depression, and current depressive symptoms); the third cluster included pain-related variables (opioid medication, history of pain, and pain symptoms) and the fourth cluster included motivation variables (measured with treatment preference, treatment credibility, compliance to the treatment schedule and contact with the therapists).

RESULTS

Findings showed that treatment progress was predicted by higher treatment credibility at baseline, whereas participants who were behind schedule in the second week of the program finished fewer treatment modules. When analyzing each cluster of predictor variables separately, current depressive symptoms also predicted fewer completed treatment modules. Among the pain-related variables, higher pain acceptance was the only predictor for completing more treatment modules. Treatment completion (which in this study was defined as having completed at least 75% of treatment modules) was predicted by higher treatment credibility and fewer depressive symptoms at baseline, and was thus similar to the results regarding treatment progress. Finally, all adherence variables predicted the treatment outcome pain interference.

CONCLUSIONS

Low treatment credibility, depressive symptoms and falling behind the treatment schedule early on were the most important predictor variables for low treatment adherence, while a number of demographical and pain-related variables were not related to adherence. The results from this study may help clinicians identify patients who are less likely to complete, and thus benefit from, their pain treatment. Trial registration ClinicalTrials.gov NTC03316846.

Improving Mental Health Services: A 50-Year Journey from Randomized Experiments to Artificial Intelligence and Precision Mental Health

This conceptual paper describes the current state of mental health services, identifies critical problems, and suggests how to solve them. I focus on the potential contributions of artificial intelligence and precision mental health to improving mental health services. Toward that end, I draw upon my own research, which has changed over the last half century, to highlight the need to transform the way we conduct mental health services research. I identify exemplars from the emerging literature on artificial intelligence and precision approaches to treatment in which there is an attempt to personalize or fit the treatment to the client in order to produce more effective interventions.

The promise of machine learning in predicting treatment outcomes in psychiatry

For many years, psychiatrists have tried to understand factors involved in response to medications or psychotherapies, in order to personalize their treatment choices. There is now a broad and growing interest in the idea that we can develop models to personalize treatment decisions using new statistical approaches from the field of machine learning and applying them to larger volumes of data. In this pursuit, there has been a paradigm shift away from experimental studies to confirm or refute specific hypotheses towards a focus on the overall explanatory power of a predictive model when tested on new, unseen datasets. In this paper, we review key studies using machine learning to predict treatment outcomes in psychiatry, ranging from medications and psychotherapies to digital interventions and neurobiological treatments. Next, we focus on some new sources of data that are being used for the development of predictive models based on machine learning, such as electronic health records, smartphone and social media data, and on the potential utility of data from genetics, electrophysiology, neuroimaging and cognitive testing. Finally, we discuss how far the field has come towards implementing prediction tools in real-world clinical practice. Relatively few retrospective studies to-date include appropriate external validation procedures, and there are even fewer prospective studies testing the clinical feasibility and effectiveness of predictive models. Applications of machine learning in psychiatry face some of the same ethical challenges posed by these techniques in other areas of medicine or computer science, which we discuss here. In short, machine learning is a nascent but important approach to improve the effectiveness of mental health care, and several prospective clinical studies suggest that it may be working already.

Machine Learning Predicts Outcomes of Phase III Clinical Trials for Prostate Cancer

The ability to predict the individual outcomes of clinical trials could support the development of tools for precision medicine and improve the efficiency of clinical-stage drug development. However, there are no published attempts to predict individual outcomes of clinical trials for cancer. We used machine learning (ML) to predict individual responses to a two-year course of bicalutamide, a standard treatment for prostate cancer, based on data from three Phase III clinical trials (n = 3653). We developed models that used a merged dataset from all three studies. The best performing models using merged data from all three studies had an accuracy of 76%. The performance of these models was confirmed by further modeling using a merged dataset from two of the three studies, and a separate study for testing. Together, our results indicate the feasibility of ML-based tools for predicting cancer treatment outcomes, with implications for precision oncology and improving the efficiency of clinical-stage drug development.

A Machine Learning Approach to Predicting New-onset Depression in a Military Population

OBJECTIVE

Depression is one of the most common mental disorders in the United States in both civilian and military populations, but few prospective studies assess a wide range of predictors across multiple domains for new-onset (incident) depression in adulthood. Supervised machine learning methods can identify predictors of incident depression out of many different candidate variables, without some of the assumptions and constraints that underlie traditional regression analyses. The objectives of this study were to identify predictors of incident depression across 5 years of follow-up using machine learning, and to assess prediction accuracy of the algorithms.

METHODS

Data were from a cohort of Army National Guard members free of history of depression at baseline (n = 1951 men and 298 women), interviewed once per year for probable depression. Classification trees and random forests were constructed and cross-validated, using 84 candidate predictors from the baseline interviews.

RESULTS

Stressors and traumas such as emotional mistreatment and adverse childhood experiences, demographics such as being a parent or student, and military characteristics including paygrade and deployment location were predictive of probable depression. Cross-validated random forest algorithms were moderately accurate (68% for women and 73% for men).

CONCLUSIONS

Events and characteristics throughout the life course, both in and outside of deployment, predict incident depression in adulthood among military personnel. Although replication studies are needed, these results may help inform potential intervention targets to reduce depression incidence among military personnel. Future research should further refine and explore interactions between identified variables.

Errors-in-Variables Modeling of Personalized Treatment-Response Trajectories

Estimating the impact of a treatment on a given response is needed in many biomedical applications. However, methodology is lacking for the case when the response is a continuous temporal curve, treatment covariates suffer extensively from measurement error, and even the exact timing of the treatments is unknown. We introduce a novel method for this challenging scenario. We model personalized treatment-response curves as a combination of parametric response functions, hierarchically sharing information across individuals, and a sparse Gaussian process for the baseline trend. Importantly, our model accounts for errors not only in treatment covariates, but also in treatment timings, a problem arising in practice for example when data on treatments are based on user self-reporting. We validate our model with simulated and real patient data, and show that in a challenging application of estimating the impact of diet on continuous blood glucose measurements, accounting for measurement error significantly improves estimation and prediction accuracy.

Using Artificial Intelligence to Predict Change in Depression and Anxiety Symptoms in a Digital Intervention: Evidence from a Transdiagnostic Randomized Controlled Trial

While digital psychiatric interventions reduce treatment barriers, not all persons benefit from this type of treatment. Research is needed to preemptively identify who is likely to benefit from these digital treatments in order to redirect those people to a higher level of care. The current manuscript used an ensemble of machine learning methods to predict changes in major depressive and generalized anxiety disorder symptoms from pre to 9-month follow-up in a randomized controlled trial of a transdiagnostic digital intervention based on participants' (N=632) pre-treatment data. The results suggested that baseline characteristics could accurately predict changes in depressive symptoms in both treatment groups (r=0.482, 95% CI[0.394, 0.561]; r=0.477, 95% CI[0.385, 0.560]) and anxiety symptoms in both treatment groups (r=0.569, 95% CI[0.491, 0.638]; r=0.548, 95% CI[0.464, 0.622]). These results suggest that machine learning models are capable of preemptively predicting a person's responsiveness to digital treatments, which would enable personalized decision-making about which persons should be directed towards standalone digital interventions or towards blended stepped-care.

Predictors of remission from body dysmorphic disorder after internet-delivered cognitive behavior therapy: a machine learning approach

BACKGROUND

Previous attempts to identify predictors of treatment outcomes in body dysmorphic disorder (BDD) have yielded inconsistent findings. One way to increase precision and clinical utility could be to use machine learning methods, which can incorporate multiple non-linear associations in prediction models.

METHODS

This study used a random forests machine learning approach to test if it is possible to reliably predict remission from BDD in a sample of 88 individuals that had received internet-delivered cognitive behavioral therapy for BDD. The random forest models were compared to traditional logistic regression analyses.

RESULTS

Random forests correctly identified 78% of participants as remitters or non-remitters at post-treatment. The accuracy of prediction was lower in subsequent follow-ups (68, 66 and 61% correctly classified at 3-, 12- and 24-month follow-ups, respectively). Depressive symptoms, treatment credibility, working alliance, and initial severity of BDD were among the most important predictors at the beginning of treatment. By contrast, the logistic regression models did not identify consistent and strong predictors of remission from BDD.

CONCLUSIONS

The results provide initial support for the clinical utility of machine learning approaches in the prediction of outcomes of patients with BDD.

TRIAL REGISTRATION

ClinicalTrials.gov ID: NCT02010619.

Network analyses reveal which symptoms improve (or not) following an Internet intervention (Deprexis) for depression

BACKGROUND

Depression is a heterogeneous collection of symptoms. Prior meta-analyses using symptom sum scores have shown the Internet intervention, Deprexis, to be an efficacious treatment for depression. However, no prior research has investigated how Deprexis (or any other Internet intervention for depression) impacts specific symptoms of depression. The current study utilizes symptom-level analyses to examine which symptoms are directly, indirectly, or minimally influenced by treatment.

METHODS

Network analysis and mean-level approaches examined which symptoms, assessed by the Quick Inventory of Depression Symptoms, were affected by an 8-week course of Deprexis compared with a waitlist in a nationally recruited sample from the United States (N = 295).

RESULTS

Deprexis directly improved the symptoms of sadness and indecision. Changes in these symptoms, in turn, was associated with a change in early insomnia, middle insomnia, self-dislike, fatigue, anhedonia, suicidality, slowness, and agitation. All of these symptoms (except for agitation and early insomnia) show decreases with Deprexis compared with a waitlist after correcting for multiple comparisons. Six additional symptoms, particularly the somatic symptoms, were not impacted by Deprexis compared with a waitlist.

CONCLUSIONS

In this sample, the efficacy of Deprexis was due to its direct impact on sadness and indecision. Examining the treatment-related change in specific symptoms may facilitate a more nuanced understanding of how a treatment works compared with examining symptom sum scores. Symptom-level approaches may also identify symptoms that do not improve and provide important direction for future treatment development.

Neo-adjuvant chemoradiotherapy response prediction using MRI based ensemble learning method in rectal cancer patients

OBJECTIVES

The aim of this study was to investigate and validate the performance of individual and ensemble machine learning models (EMLMs) based on magnetic resonance imaging (MRI) to predict neo-adjuvant chemoradiation therapy (nCRT) response in rectal cancer patients. We also aimed to study the effect of Laplacian of Gaussian (LOG) filter on EMLMs predictive performance.

METHODS

98 rectal cancer patients were divided into a training (n = 53) and a validation set (n = 45). All patients underwent MRI a week before nCRT. Several features from intensity, shape and texture feature sets were extracted from MR images. SVM, Bayesian network, neural network and KNN classifiers were used individually and together for response prediction. Predictive performance was evaluated using the area under the receiver operator characteristic (ROC) curve (AUC).

RESULTS

Patients' nCRT responses included 17 patients with Grade 0, 28 with Grade 1, 34 with Grade 2, and 19 with Grade 3 according to AJCC/CAP pathologic grading. In without preprocessing MR Image the best result was for Bayesian network classifier with AUC and accuracy of 75.2% and 80.9% respectively, which was confirmed in the validation set with an AUC and accuracy of 74% and 79% respectively. In EMLMs the best result was for 4 (SVM.NN.BN.KNN) classifier EMLM with AUC and accuracy of 97.8% and 92.8% in testing and 95% and 90% in validation set respectively.

CONCLUSIONS

In conclusion, we observed that machine learning methods can used to predict nCRT response in patients with rectal cancer. Preprocessing LOG filters and EL models can improve the prediction process.