Internet-delivered Psychological Treatment Research Articles

Deep Hierarchical Attention Active Learning for Mental Disorder Unlabeled Data in AIoMT

In the Artificial Intelligence of Medical Things (AIoMT), Internet-Delivered Psychological Treatment (IDPT) effectively improves the quality of mental health treatments. With the advent of COVID-19, psychological tasks have become overloaded and complicated for medical professionals due to the overlap of sentimental values. The development of an AIoMT tool requires labeling of data to achieve clinical-level performance. Text data requires an appropriate set of linguistic features for vector latent representation and segmentation. Emotional biases could lead to incorrect segmentation of patient-authorized texts, and labeling emotional data is time-consuming. In this article, we propose an assistant tool for psychologists to assist them in mental health treatment and note-taking. We first extend the word and emotion lexicon and then apply a hierarchical attention method to support data labeling. The learned latent representation uses word position prediction and sentence-level attention to create a semantic framework. The augmented vector representation helps in highlighting words and classifying nine different symptoms from the text written by the patient. Our experimental results show that the emotion lexicon helps to increase the accuracy by 5% without affecting the overall results, and that the hierarchical attention method achieves an F1 score of 0.89.

Hyper-Graph Attention Based Federated Learning Method For Mental Health Detection.

Internet-Delivered Psychological Treatment (IDPT) has become necessary in the medical field. Deep neural networks (DNNs) require large, diverse patient populations to train models that achieve clinician-level performance. However, DNN models trained on limited datasets have poor clinical performance when used in a new location with different data. Thus, increasing the availability of diverse as well as distinct training data is vital. This study proposes a structural hypergraph as well as an emotional lexicon for word representation. An embedding model based on federated learning was developed for mental health symptom detection. The model treats text data as a collection of consecutive words. The model then learns a low-dimensional continuous vector while maintaining contextual linkage. The generated models with attention-based mechanisms as well as federated learning are then tested experimentally. Our strategy is suitable for vocabulary diversification, grammatical word representation, as well as dynamic lexicon analysis. The goal is to create semantic word representations using an attention network model. Later, clinical processes are used to mark the text by embedding it. Experimental results show the encoding of emotional words using the structural hypergraph. The 0.86 ROC was achieved using the bidirectional LSTM architecture with an attention mechanism.

Internet-delivered psychological treatment of functional gastrointestinal disorders in youth: Study protocol for exploration of cognitive biases

Introduction Functional gastrointestinal disorders (FGID) are common in children and adolescents (CA), cause functional disability and high health care use. Internet based cognitive behavioral therapy (i-CBT) have shown promising effect in Sweden. The treatment is exposure based and target avoidance behavior. Cognitive biases regarding bodily symptoms are suggested to be part of development and maintenance of functional disorders in adults, and could therefore be an important treatment target. Little is known about cognitive biases in CA with FGID, and hence the potential importance, it is crucial to explore more in depth. This study is embedded in The Danish FGID Treatment Study which aims to test Swedish i-CBT treatment in a Danish context.Objectives1) Examine cognitive biases in CA with FGID, compared with healthy controls. 2) Asses these biases before and after treatment for FGID to investigate changes and impact on treatment effect.MethodsWe developed a novel experimental approach to examine possible cognitive biases in CA. It consists of a picture task and a FGID-specific version of the Health Norms Sorting Task. We will examine 60 CA with FGID, who are included in The Danish FGID Treatment Study before and after treatment. In addition we will perform the test on 100 healthy controls.ResultsThe results are expected to improve our understanding of maintaining cognitive factors in CA with FGID, and how these may affect outcome and change during treatment, and how they differ from the general population.ConclusionsThis study can provide new targets for treatment approaches.DisclosureNo significant relationships.

Towards adaptive technology in routine mental health care.

This paper summarizes the information technology-related research findings after 5 years with the INTROducing Mental health through Adaptive Technology project. The aim was to improve mental healthcare by introducing new technologies for adaptive interventions in mental healthcare through interdisciplinary research and development. We focus on the challenges related to internet-delivered psychological treatments, emphasising artificial intelligence, human-computer interaction, and software engineering. We present the main research findings, the developed artefacts, and lessons learned from the project before outlining directions for future research. The main findings from this project are encapsulated in a reference architecture that is used for establishing an infrastructure for adaptive internet-delivered psychological treatment systems in clinical contexts. The infrastructure is developed by introducing an interdisciplinary design and development process inspired by domain-driven design, user-centred design, and the person based approach for intervention design. The process aligns the software development with the intervention design and illustrates their mutual dependencies. Finally, we present software artefacts produced within the project and discuss how they are related to the proposed reference architecture. Our results indicate that the proposed development process, the reference architecture and the produced software can be practical means of designing adaptive mental health care treatments in correspondence with the patients' needs and preferences. In summary, we have created the initial version of an information technology infrastructure to support the development and deployment of Internet-delivered mental health interventions with inherent support for data sharing, data analysis, reusability of treatment content, and adaptation of intervention based on user needs and preferences.

EANDC: An explainable attention network based deep adaptive clustering model for mental health treatment

Internet-delivered Psychological Treatment (IDPT) has been shown to be an effective method for improving psychological disorders. Natural language processing (NLP) requires an appropriate set of linguistic features for word representation and emotion segmentation. For psychological applications, models must be trained on extensive and diverse datasets to achieve expert-level performance. Labeling psychological texts authorized by patients is challenging because emotional biases can lead to incorrect segmentation of emotions and labeling emotional data is time consuming. In this paper, we propose an assistance tool for psychologists to explore the emotional aspects of mentally ill individuals. We first use an NLP-based method to create emotional lexicon embeddings, and then apply attention-based deep clustering. The learned representation is then used to visualize the emotional aspect of the text authorized by patients. We expand the patient authored text using synonymous semantic expansion. A latent semantic representation based on context is clustered using EANDC, which is a Explainable Attention Network-based Deep adaptive Clustering model. We use similarity metrics to select a subset of the text and then improve the explainability of learning using a curriculum-based optimization method. The experimental results show that synonym expansion based on the emotion lexicon increases accuracy without affecting the results. The attention method with bidirectional LSTM architecture achieved 0.81 ROC in a blind test. The self-learning based embedding then visualizes the weighted attention words and helps the psychiatrist to improve his explanatory power of the qualitative match for clinical notes and the remedy. The method helps in labeling text and improves the recognition rate of symptoms of mental disorders.

Attention-Based Deep Entropy Active Learning Using Lexical Algorithm for Mental Health Treatment.

With the increasing prevalence of Internet usage, Internet-Delivered Psychological Treatment (IDPT) has become a valuable tool to develop improved treatments of mental disorders. IDPT becomes complicated and labor intensive because of overlapping emotion in mental health. To create a usable learning application for IDPT requires diverse labeled datasets containing an adequate set of linguistic properties to extract word representations and segmentations of emotions. In medical applications, it is challenging to successfully refine such datasets since emotion-aware labeling is time consuming. Other known issues include vocabulary sizes per class, data source, method of creation, and baseline for the human performance level. This paper focuses on the application of personalized mental health interventions using Natural Language Processing (NLP) and attention-based in-depth entropy active learning. The objective of this research is to increase the trainable instances using a semantic clustering mechanism. For this purpose, we propose a method based on synonym expansion by semantic vectors. Semantic vectors based on semantic information derived from the context in which it appears are clustered. The resulting similarity metrics help to select the subset of unlabeled text by using semantic information. The proposed method separates unlabeled text and includes it in the next active learning mechanism cycle. Our method updates model training by using the new training points. The cycle continues until it reaches an optimal solution, and it converts all the unlabeled text into the training set. Our in-depth experimental results show that the synonym expansion semantic vectors help enhance training accuracy while not harming the results. The bidirectional Long Short-Term Memory (LSTM) architecture with an attention mechanism achieved 0.85 Receiver Operating Characteristic (ROC curve) on the blind test set. The learned embedding is then used to visualize the activated word's contribution to each symptom and find the psychiatrist's qualitative agreement. Our method improves the detection rate of depression symptoms from online forum text using the unlabeled forum texts.

Moderators of Internet-Delivered Cognitive-Behavioral Therapy for Adolescents With Chronic Pain: Who Benefits From Treatment at Long-Term Follow-Up?

Cognitive behavioral therapy (CBT) is effective for pediatric chronic pain, but little is understood about which youth are most likely to benefit. The current study aimed to identify individual characteristics for which CBT yielded the greatest (and least) clinical benefit among adolescents with chronic pain participating in a multicenter randomized controlled trial of Internet-delivered CBT (WebMAP2). A total of 273 adolescents ages 11 to 17 with chronic pain (M age = 14.7; 75.1% female) were randomly assigned to Internet-delivered CBT or Internet-delivered pain education and evaluated at pretreatment, post-treatment, and 2 longer term follow-up periods (6 and 12 months). Multilevel growth models tested several adolescent- and parent-level moderators of change in pain-related disability including 1) adolescent age, sex, pain characteristics, distress, and sleep quality and 2) parent education level, distress, and protective parenting behavior. Younger adolescents (ages 11–14; vs older adolescents ages 15–17) and those whose parents experienced lower levels (vs higher levels) of emotional distress responded better to Internet CBT treatment, showing greater improvements in disability up to 12 months post-treatment. This study expands knowledge on who benefits most from Internet-delivered psychological treatment for youth with chronic pain in the context of a large multicenter randomized controlled trial, suggesting several avenues for maximizing treatment efficacy and durability in this population. PerspectiveThis study identified adolescent- and parent-level predictors of treatment response to Internet-based CBT for pediatric chronic pain up to 12 months later. Younger adolescents and those whose parents had lower levels of distress may particularly benefit from this intervention. Older adolescents and those whose parents exhibit higher distress may require alternative treatment approaches.

Learning machines in Internet-delivered psychological treatment

A learning machine, in the form of a gating network that governs a finite number of different machine learning methods, is described at the conceptual level with examples of concrete prediction subtasks. A historical data set with data from over 5000 patients in Internet-based psychological treatment will be used to equip healthcare staff with decision support for questions pertaining to ongoing and future cases in clinical care for depression, social anxiety, and panic disorder. The organizational knowledge graph is used to inform the weight adjustment of the gating network and for routing subtasks to the different methods employed locally for prediction. The result is an operational model for assisting therapists in their clinical work, about to be subjected to validation in a clinical trial.

Internet-delivered psychological treatment for patients with health anxiety: Results from a randomized, controlled trial

Self-rated health (SRH) has been shown to be a stable predictor of illness and mortality. Improvement in SRH, even in the absence of change in objective health, predicts better health and reduced mortality. Severe health anxiety (SHA) is characterized by fear of illness and distorted health perception. The objective of the present study was to investigate if exposure-based cognitive behavior therapy (CBT) for SHA can lead to improvement in SRH and whether this improvement is mediated by reduced health anxiety.Data were collected from a randomized controlled trial where participants with SHA were allocated to 12 weeks of exposure-based CBT (n = 99) for SHA or to a no treatment control condition (n = 33). The mediation analysis was based on SRH- and health anxiety data collected weekly during the treatment phase.Linear mixed effects models analysis showed a significant interaction effect of group and time indicating superior improvements in SRH in exposure-based CBT compared to the control condition (Z = 2.69, p = 0.007). The controlled effect size was moderately large (d = 0.64) and improvements were stable at 1-year follow-up. Reduced health anxiety was a significant mediator of improvement in SRH.12 weeks of exposure-based CBT for SHA can lead to significant improvements in SRH. Considering the previously established importance of SRH as a predictor for disease and mortality, exposure-based CBT for severe health anxiety may lead to improvements on several important health parameters, possibly even increasing the likelihood of longevity.

Internet-Delivered Psychological Treatment Options for Panic Disorder: A Review on Their Efficacy and Acceptability.

Background Internet-delivered psychological treatments have been suggested as a chance to expand the access to professional help. However, little is known about the usefulness of different support formats and approaches of digital treatments for panic disorder among clinicians.Objective This narrative review aimed to explore the recent evidence base on the efficacy and acceptability of different internet-delivered treatments for adults with panic disorder.MethodsA systematic search in electronic databases (Pubmed/Medline, PSYNDEX) and a hand search were performed to identify articles on randomized controlled trials published within the past five years (2012/12/10–2017/12/12) in English peer-reviewed journals.ResultsEight studies (1,013 participants) involving 10 interventions met the inclusion criteria. Nine interventions were primarly based on Cognitive Behavioral Therapy principles. Most interventions were effective, when compared to a control condition (6 of 8 comparisons). Minimal guidance was associated with improved outcomes in one study and adherence in two studies (3 comparisons). Furthermore, no differences were found based on treatment approach (2 comparisons). Regarding acceptability, the attrition rates were moderate to high, ranging from 9.8% to 42.1% of randomized participants. Adherence rates also varied largely (7.8–75%), whereas participant satisfaction of program completers was assessed overall high (5 studies).ConclusionDiverse effective internet-delivered treatments are available for the self-management of panic symptoms. Especially selfguided and transdiagnostic Cognitive Behavioral Therapy approaches appear being efficient options for the dissemination in routine care. However, due to the limited evidence base, further efforts are required to improve the actual uptake of internet-delivered treatments and identify moderators of outcomes.

Does cognitive flexibility predict treatment gains in Internet-delivered psychological treatment of social anxiety disorder, depression, or tinnitus?

Little is known about the individual factors that predict outcomes in Internet-administered psychological treatments. We hypothesized that greater cognitive flexibility (i.e. the ability to simultaneously consider several concepts and tasks and switch effortlessly between them in response to changes in environmental contingencies) would provide a better foundation for learning and employing the cognitive restructuring techniques taught and exercised in therapy, leading to greater treatment gains. Participants in three trials featuring Internet-administered psychological treatments for depression (n = 36), social anxiety disorder (n = 115) and tinnitus (n = 53) completed the 64-card Wisconsin Card Sorting Test (WCST) prior to treatment. We found no significant associations between perseverative errors on the WCST and treatment gains in any group. We also found low accuracy in the classification of treatment responders. We conclude that lower cognitive flexibility, as captured by perseverative errors on the WCST, should not impede successful outcomes in Internet-delivered psychological treatments.

Trajectories of change during a randomized controlled trial of internet-delivered psychological treatment for adolescent chronic pain: how does change in pain and function relate?

Although pain and function improve at immediate posttreatment for youth receiving cognitive behavioral therapy for chronic pain, limited data are available to understand changes that youth make during psychological treatment. We sought to characterize distinct trajectory patterns of change in pain and function to understand the temporal association of these changes during internet-delivered cognitive behavioral therapy (CBT). Weekly repeated assessments of pain and function were conducted during 8 weeks of treatment among 135 adolescents, aged 11 to 17 years, with chronic pain who were randomized to the cognitive behavioral intervention arm of an ongoing trial of internet-delivered CBT (Web-based management of adolescent pain; Web-MAP2). Using random-effects growth mixture models, we characterized pain and functional disability trajectories finding distinct trajectory groups indicating patterns of both linear and quadratic effects. Trajectories of change showed that some patients' pain and functional disability were improving, others worsened or changed minimally. Paired t tests compared the within-subject relative change rate in pain and function demonstrating similar change range for pain and function during the treatment period. There was no support for improvements in either pain or function to precede changes in the other domain. Findings may be useful in informing future studies of psychosocial treatments for pediatric chronic pain to consider how to target treatment strategies to distinct patient response profiles. This may lead to the development of intervention strategies that can both more effectively target children's pain and function during treatment and lead to sustained changes after treatment.

Experiences of non-adherence to Internet-delivered cognitive behavior therapy: A qualitative study

Many trials on Internet-delivered psychological treatments have had problems with nonadherence, but not much is known about the subjective reasons for non-adhering. The aim of this study was to explore participants' experiences of non-adherence to Internet-delivered psychological treatment. Grounded theory was used to analyze data from seven in-depth interviews with persons who had non-adhered to a study on Internet-delivered cognitive behavioral therapy for generalized anxiety disorder. The process of non-adherence is described as an interaction between patient factors and treatment factors. A working model theory was generated to illustrate the experience of nonadherence. The model describes a process where treatment features such as workload, text-content complexity and treatment process don't match personal prerequisites regarding daily routines, perceived language skills and treatment expectations respectively, resulting in the decision to nonadhere. Negative effects were also stated as a reason for non-adherence. Several common strategies used for increasing adherence to Internet-delivered therapy in general are by these non-completers regarded as factors directly related to their reason for non-adherence.

Clinician-Supported Internet-Delivered Psychological Treatment of Tinnitus.

Internet-delivered psychological treatments for tinnitus distress have existed for more than 15 years, and there are a slowly growing number of studies. The aim of this brief report is to review the evidence and to comment on the future potentials of Internet treatments for tinnitus. Studies were retrieved, and in total 6 controlled studies were included in the review with 9 different comparisons (6 in which the treatment was compared against a control group and 3 in which Internet treatment was compared against group treatment). Moreover, 2 open studies based on clinical samples in regular care were also included in the review. The outcomes for the 2 controlled sets of studies were analyzed using meta-analytic methods. For the 6 studies comparing Internet treatment against a no-treatment control condition, a moderate effect size was found (Hedges's g = 0.58). The 3 studies comparing Internet treatment against face-to-face group treatments showed a small difference of Hedges's g =0.13. Internet-delivered psychological treatment holds promise as a treatment alternative to other standard forms of treatment delivery, including group treatment. Larger studies are needed as well as ways to blend information technology with regular services.

Internet-Delivered Acceptance and Values-Based Exposure Treatment for Fibromyalgia: A Pilot Study

Background: Acceptance and commitment therapy (ACT) is a promising treatment option for fibromyalgia (FM). Studies have shown that many cognitive behavioral protocols can be transferred to the Internet with sustained efficacy. However, no study has investigated the effect on an Internet-delivered ACT-based protocol for FM. This study evaluated the efficacy, acceptability, and the health economic effects of an Internet-delivered acceptance and values-based exposure treatment for FM. Methods: This open pilot trial included 41 self-referred women with a FM diagnosis. The 10-week Internet-delivered treatment included acceptance, mindfulness, work with life-values, and systematic exposure to FM symptoms and FM-related situations. Participants also had regular contact with an assigned online therapist. Assessments were made at pretreatment, post-treatment, and 6-month follow-up. Results: The treatment was completed by 70% of the participants. Attrition rates were low, with 98% completing the post-treatment assessment and 90% completing the 6-month follow-up assessment. Multiple imputations were used to replace missing values. Pre- to post-treatment within-group effect sizes were in the moderate to large range (Cohen's d = 0.62–1.56) on measures of FM symptoms and impact, disability, quality of life, depression, anxiety, fatigue, and psychological flexibility. All improvements were maintained at follow-up. Economical analyses revealed significant societal cost reductions that offset the treatment costs within 2 months of treatment completion. Conclusions: An Internet-delivered psychological treatment based on acceptance and exposure principles seems to be an efficacious, acceptable, and cost-effective treatment for FM. Randomized controlled trials are needed to confirm these results.

Internet administered guided self-help versus individualized e-mail therapy: A randomized trial of two versions of CBT for major depression

Internet-delivered psychological treatment of major depression has been investigated in several trials, but the role of personalized treatment is less investigated. Studies suggest that guidance is important and that automated computerized programmes without therapist support are less effective. Individualized e-mail therapy for depression has not been studied in a controlled trial. Eighty-eight individuals with major depression were randomized to two different forms of Internet-delivered cognitive behaviour therapy (CBT), or to a waiting-list control group. One form of Internet treatment consisted of guided self-help, with weekly modules and homework assignments. Standard CBT components were presented and brief support was provided during the treatment. The other group received e-mail therapy, which was tailored and did not use the self-help texts i.e., all e-mails were written for the unique patient. Both treatments lasted for 8 weeks. In the guided self-help 93% completed (27/29) and in the e-mail therapy 96% (29/30) completed the posttreatment assessment. Results showed significant symptom reductions in both treatment groups with moderate to large effect sizes. At posttreatment 34.5% of the guided self-help group and 30% of the e-mail therapy group reached the criteria of high-end-state functioning (Beck Depression Inventory score below 9). At six-month follow-up the corresponding figures were 47.4% and 43.3%. Overall, the difference between guided self-help and e-mail therapy was small, but in favour of the latter. These findings indicate that both guided self-help and individualized e-mail therapy can be effective.