Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
A collection of papers published in Nature Mental Health that highlight innovative means of incorporating digital advances in mental health research and psychiatry, such as machine learning, artificial intelligence and telemedicine.
Machine Learning Methods and Applications in Mental Health
Osimo et al. developed two models to predict the risk of treatment-resistant schizophrenia in patients with a first-episode psychosis using blood-based biomarkers and sociodemographic data routinely collected at psychosis onset in psychosis early intervention services in the United Kingdom. They used clozapine treatment as a proxy for treatment-resistant schizophrenia using data from 785 patients for model development and 1,110 patients for external validation.
Using machine learning, Zhang et al. identify EEG signature to predict psychotherapy outcomes in PTSD, paving the way towards the development of scalable biomarkers.
Will connectome-based predictive modeling change how we care for people at risk of late-life suicide? A novel two-step modeling approach used by Gao et al. in their study sheds light on the road ahead.
Using magnetic resonance imaging and connectome-based predictive modelling, Gao et al. find that brain connectivity data can predict suicide risk in patients with late-life depression.
Using data-driven disease-progression modelling, Jiang, Wang, Zhou et al. characterized and replicated two distinct ‘trajectories’ of brain atrophy in patients with schizophrenia.
In this Article, Singh and coauthors put forth a new machine-learning approach to evaluate inclusion and exclusion criteria from psychiatry abstracts to automate systematic reviews.
This Review discusses the adverse consequences of phenotypic imprecision for discovering reproducible biological correlates of psychopathology and provides recommendations for precision phenotyping that will help to overcome these challenges.
This Review summarizes the advances in personalized medicine and drug discovery in psychiatry and suggests a framework for the development of clinically relevant biological subtypes in the field.
Loechner et al. present an expert consensus on how to e-mental health, from the development to research and evaluation of e-mental health interventions, a useful guide with practical recommendations for researchers and practitioners wanting to explore the field of e-mental health.
Mental health is essential to a person’s wellbeing, and mental health is a crucial component of the positive functioning and flourishing of families, communities and societies. At CNS Summit 2022, held 17–20 November 2022, Murali Doraiswamy asked Joshua Gordon from the National Institute of Mental Health to explain current limitations in the field of psychiatry and future steps to overcome these impediments.
Panayiotou et al. performed a panel network analysis to investigate the relationship between time spent on social media and mental health in a large cohort of UK adolescents.
In this paper describing the impact of the coronavirus disease 2019 pandemic on a large public mental health system, the authors examined the availability and service uptake of telepsychiatry over 2 years in a regional network of community mental health centres covering approximately 10 million people in Italy.
Neurotechnologies that measure and modulate brain activity have not yet reached widespread clinical relevance. To accelerate translation into patient care, we propose three strategic adjustments in neurotechnology research — to consider the scope, scalability and stakeholders.