Search

The current study aims at assessing the performance of a multi-modal algorithm developed by the start-up Sentimentics (http://www.sentimentics.com/) in recognition of stress in an individual. This algorithm estimates stress according to information obtained from audio (both acoustic as well as verbal) and visual features, using machine learning methods, namely artificial deep neural networks.

Common mental health disorders, more specifically mood and anxiety disorders, are highly prevalent and lead to decrease in general health, quality of life as well as social functioning, contributing substantially to the global burden of disease. Objective, real-time assessment of their active and passive signs, presented as expressions of sentiment in text, specific vocal features or facial expressions, may contribute to earlier recognition, diagnosis of and treatment for these disorders. Techniques allowing for such assessment are in their infancy but appearing in fundamental as well as applied experimental research. The current scoping review aims to systematically synthesize the results of studies which focus on the use of text, audio and video recognition algorithms, their feasibility and effectiveness in the identification of symptoms of depression, anxiety and related mental outcomes, in technical or healthcare setting. To the best of our knowledge, this is the first systematic review of in this emerging field. Its results will inform researchers and health professionals about the current state of the art, the applicability and potential evidence base for the use of and research into such algorithms. In addition, their technological and clinical possibilities for interventions for mental disorders will be addressed. This protocol is structured according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses - Extension for Scoping Reviews (PRISMA-ScR). After the identification of the research question, searches will be conducted in 7 databases (PubMed, Embase, PsycINFO, Web of Science, Scopus, ACM digital library and dblp computer science bibliography). In order to be included, studies need to focus on the use of a recognition algorithm (text, audio or video) for symptoms related to depressive or anxiety states. Two researchers will independently screen and select references, first according to their titles/abstracts and subsequently based on full-texts, and extract data. The analysis will synthesize existing knowledge, provide overview of these methods and suggest what their implications could be for clinical purposes and research.

It is estimated that depression affects about 6% of the worldwide adult population every year (Bromet et al., 2011). In populations with somatic disorders the odds of suffering depression are significantly higher, as indicated by epidemiological and meta-analytical evidence (Moussavi et al., 2007; Chang-Quan et al., 2010; Read et al., 2017). Moreover, in individuals with somatic conditions, depression has been associated with poorer functional health outcomes and an increased risk of mortality (Baune, Adrian & Jacobi, 2007; Evans & Charney, 2003). Several meta-analyses have shown that psychological interventions are effective in reducing the severity of depression symptoms (Cuijpers et al., 2019). However, no recent meta-analysis has examined the effects of psychological interventions for depression in individuals with somatic conditions. We aim to conduct a meta-analysis to examine the effects of psychological interventions for depression in patients with various somatic diseases.

Background. The number of trials on psychotherapies for adult depression is very large and is quickly growing. Because of this large body of knowledge, it is important that the results of these studies are summarized and integrated in meta-analytic studies. More than a decade ago we developed a meta-analytic database of these trials, which was updated yearly through systematic literature searches. Recently, we developed a new version of this meta-analytic database, built on the systems and experience from our earlier database, but with completely new searches and improved methods. In this paper we will describe the methods and some first results of this database. Description. We conducted systematic literature searches in bibliographical databases (PubMed, Embase, PsycINFO, Cochrane Register of Controlled Trials to identify all trials on psychotherapy for adult depression (deadline January 1st, 2019). We excluded trials on maintenance and relapse prevention, dissertations, collaborative care, and studies not published in English, German, Spanish or Dutch. After reading 21,976 records (16,701 after exclusion of duplicates), we included 661 randomized trials. We distinguished the following categories of trials: Psychotherapy versus pharmacotherapy (65 studies), combined treatment versus pharmacotherapy alone (46), combined treatment versus psychotherapy alone (29), combined treatment versus psychotherapy plus placebo (18), psychotherapy versus control (335), psychotherapy versus another therapy (109), psychotherapy for inpatients (34), unguided self-help interventions (48), comparisons of different treatment formats (38), cognitive bias modification (14) and other comparisons (99). Over the years we have published several dozens of meta-analyses using this databases (including its previous versions). Conclusion. Psychotherapy for depression is definitely the best studied type of psychotherapy for any mental health problem. We hope that our database can be used as a resource for researchers who want to conduct systematic reviews and meta-analyses of subgroups of these studies.

Introduction: Common mental health disorders, more specifically mood and anxiety disorders, are highly prevalent and lead to decrease in general health, quality of life as well as social functioning, contributing substantially to the global burden of disease. Objective, real-time assessment of their active and passive signs, presented as expressions of sentiment in text, specific vocal features or facial expressions, may contribute to the timely recognition of and intervention for these disorders. Techniques allowing for such assessment are appearing progressively in primary, both experimental as well as observational, research. The current scoping review aims to systematically synthesize the results of studies which focus on the use of text, vocal and facial recognition algorithms, their feasibility and effectiveness in the identification of symptoms of depression, anxiety and related mental states. To the best of our knowledge, this is the first systematic review of in this emerging field. Its results will inform health professionals and researchers about the evidence base for the use of and research into such algorithms as well as their technological and clinical possibilities for interventions for mental disorders. Methods and analysis: This protocol is structured according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses - Extension for Scoping Reviews (PRISMA-ScR). After the identification of the research question, searches will be conducted in 7 databases (PubMed, Embase, PsycINFO, Web of Science, Scopus, ACM digital library and dblp computer science bibliography). In order to be included, studies need to focus on the use of a recognition algorithm (text, vocal or facial) for symptoms related to depressive or anxiety states. Two researchers will independently screen and select references, first according to their titles/abstracts and subsequently based on full-texts, and extract data. The analysis will synthesize existing evidence, provide overview of these methods and their implications for clinical purposes. Discussion: The results will lead future research endeavours as well as advance of current clinical practice with promising innovative tools and approaches which embark on the power of artificial intelligence.

Psychological interventions are effective in the treatment of adult depression (Cuijpers et al., 2021), but it remains to be confirmed what the optimal duration of a psychological intervention is and how often sessions should be delivered in order to achieve the best outcomes. Answering these questions is crucial, similarly to the search of the optimal dose of medication in pharmacological does-response studies. Missing this information may lead to the client receiving too little or too much of treatment, causing insufficient efficacy in the former and unnecessary costs for the society, who could invest into treatment for more patients instead, in the latter case. Moreover, understanding what the optimal amount and intensity is may shed light on the mechanisms of change inherent to psychotherapy. Previous research investigating the influence of duration of psychotherapy for depression brought mixed results (Barkham et al., 1996; Dekker et al., 2005; Molenaar et al., 2011; Shapiro et al., 1994). Similarly, contradicting conclusions have been reported in studies assessing the impact of intensity of psychological interventions for anxiety on this effectiveness (Abramowitz et al., 2003; Bohni et al., 2009; Herbert et al., 2004). Evidence for psychological interventions for depression suggests that delivery of twice weekly sessions of Cognitive Behavior Therapy and Interpersonal therapy improves treatment outcomes as opposed to weekly sessions (Bruijniks et al., 2020). A previous metaregression analysis investigating the association between treatment duration, contact, and intensity, and effect size of psychological interventions for depression (Cuijpers et al., 2013) found a strong association between number of sessions per week and the effect size. Moreover, a small association was found between the number of sessions of psychotherapy and the effect size, which was not significant after adjusting for other study characteristics. In addition, the effect size was not associated with the total contact time between the therapist and the client, and the duration of the therapy. However, since the aforementioned meta-analysis was published almost a decade ago and the current causal evidence comes from only one randomized controlled trial, these conclusions need to be updated by using the most recent evidence coming from multiple recent randomized controlled trials, pooled in a meta-analysis. Therefore, the aim of the current study is to assess whether an association exists between effectiveness of individual, face-to-face psychological interventions for adult depression and (1) the number of treatment sessions, (2) duration of psychotherapy, (3) total contact time, and (4) the number of sessions per week.