Comparison of three machine learning models to predict suicidal ideation and depression among Chinese adolescents: A cross-sectional study

Huang, Yating; Zhu, Chunyan; Feng, Yunhe; Ji, Yifu; Song, Jingze; Wang, Kai; Yu, Fengqiong

doi:10.1016/j.jad.2022.08.123

Cited by 17 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the IMAGEN cohort, Toenders et al used LR (AUC = 0.70) and identified baseline depressive symptom severity, female sex, neuroticism, prior bullying, adverse life events, and surface area of the supramarginal gyrus as the most important predictors of later adolescent depression [ 14 ]. Finally, Huang et al found that RF outperformed CART and SVM algorithms (AUC = 0.90), with suicidality, anhedonia, lack of social support, emotional neglect in childhood, non-suicidal self-injury and poor maternal relationship being the most important final predictors of depression in youth [ 11 ].…”

Section: Discussionmentioning

confidence: 99%

Predicting individual cases of major adolescent psychiatric conditions with artificial intelligence

de Lacy,

Ramshaw,

McCauley

et al. 2023

Transl Psychiatry

View full text Add to dashboard Cite

Three-quarters of lifetime mental illness occurs by the age of 24, but relatively little is known about how to robustly identify youth at risk to target intervention efforts known to improve outcomes. Barriers to knowledge have included obtaining robust predictions while simultaneously analyzing large numbers of different types of candidate predictors. In a new, large, transdiagnostic youth sample and multidomain high-dimension data, we used 160 candidate predictors encompassing neural, prenatal, developmental, physiologic, sociocultural, environmental, emotional and cognitive features and leveraged three different machine learning algorithms optimized with a novel artificial intelligence meta-learning technique to predict individual cases of anxiety, depression, attention deficit, disruptive behaviors and post-traumatic stress. Our models tested well in unseen, held-out data (AUC ≥ 0.94). By utilizing a large-scale design and advanced computational approaches, we were able to compare the relative predictive ability of neural versus psychosocial features in a principled manner and found that psychosocial features consistently outperformed neural metrics in their relative ability to deliver robust predictions of individual cases. We found that deep learning with artificial neural networks and tree-based learning with XGBoost outperformed logistic regression with ElasticNet, supporting the conceptualization of mental illnesses as multifactorial disease processes with non-linear relationships among predictors that can be robustly modeled with computational psychiatry techniques. To our knowledge, this is the first study to test the relative predictive ability of these gold-standard algorithms from different classes across multiple mental health conditions in youth within the same study design in multidomain data utilizing >100 candidate predictors. Further research is suggested to explore these findings in longitudinal data and validate results in an external dataset.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting individual cases of major adolescent psychiatric conditions with artificial intelligence

de Lacy,

Ramshaw,

McCauley

et al. 2023

Transl Psychiatry

View full text Add to dashboard Cite

show abstract

“…Here, our results congruent with the small number of comparable ML studies that have included parental traits as candidate predictors, where parent total behavioral problems and poor maternal relationships were leading predictors of depression. (15,48) Sleep disturbances may affect up to ~40% of elementary school age children and youth with both internalizing and externalizing disorders are at elevated risk. (49,50) We found that sleep disturbances in the late elementary school age group (9-10yrs) predicted the later (11-12 yrs) onset (anxiety, SSD) and prevalence (depression) of internalizing disorders, congruent with recent research showing that disturbed or short duration sleep predicts later internalizing symptoms.…”

Section: Common and Specific Themes Across Internalizing Disordersmentioning

confidence: 99%

“…Outside the US, national registries or school system data have been available offering large sample sizes (n>10,000) but these typically lack physiologic information such as neuroimaging data. (14,15,16,17) An alternative strategy is to combine data from multiple studies offering neuroimaging or genomic data to boost sample size such as the datasets offered by IMAGEN or ENIGMA, though pooling across heterogenous studies may inherently limit features (variables) available for analysis to those that are shared across all studies. (18,19,20) Consequently, to promote comparative discovery at scale, federal and other organizations have recently sponsored the formation of large, longitudinal cohorts collecting a wide variety of multimodal data types with standardized protocols.…”

Section: Introductionmentioning

confidence: 99%

Predicting the onset of internalizing disorders in early adolescence using deep learning optimized with AI

de Lacy,

Ramshaw

2023

Preprint

View full text Add to dashboard Cite

Internalizing disorders (depression, anxiety, somatic symptom disorder) are among the most common mental health conditions that can substantially reduce daily life function. Early adolescence is an important developmental stage for the increase in prevalence of internalizing disorders and understanding specific factors that predict their onset may be germane to intervention and prevention strategies. We analyzed ∼6,000 candidate predictors from multiple knowledge domains (cognitive, psychosocial, neural, biological) contributed by children of late elementary school age (9-10 yrs) and their parents in the ABCD cohort to construct individual-level models predicting the later (11-12 yrs) onset of depression, anxiety and somatic symptom disorder using deep learning with artificial neural networks. Deep learning was guided by an evolutionary algorithm that jointly performed optimization across hyperparameters and automated feature selection, allowing more candidate predictors and a wider variety of predictor types to be analyzed than the largest previous comparable machine learning studies. We found that the future onset of internalizing disorders could be robustly predicted in early adolescence with AUROCs ≥∼0.90 and ≥∼80% accuracy. Each disorder had a specific set of predictors, though parent problem behavioral traits and sleep disturbances represented cross-cutting themes. Additional computational experiments revealed that psychosocial predictors were more important to predicting early adolescent internalizing disorders than cognitive, neural or biological factors and generated models with better performance. We also observed that the accuracy of individual-level models was highly correlated to the relative importance of their constituent predictors, suggesting that principled searches for predictors with higher importance or effect sizes could support the construction of more accurate individual-level models of internalizing disorders. Future work, including replication in additional datasets, will help test the generalizability of our findings and explore their application to other stages in human development and mental health conditions.

show abstract

“…This geographical divide intensifies the difficulties faced by individuals in remote locations, compounding the impact of mental health disorders on these communities. [3] Furthermore, economic inequalities contribute to disparities in mental health care access, Volume 47 Issue 3 (September 2023) https://powertechjournal.com limiting the ability of certain populations to afford or access necessary services. A shortage of mental health professionals represents another critical barrier.…”

Section: Prevalence and Barriers In Mental Health Carementioning

confidence: 99%

AI-Enabled Mental Health Assessment and Intervention: Bridging Gaps in Access and Quality of Care.

Sivasubramanian Balasubramanian

2023

pst

View full text Add to dashboard Cite

This paper explores the transformative potential of artificial intelligence (AI) in revolutionizing mental health assessment and intervention, aiming to address critical gaps in both accessibility and quality of care. Mental health disorders pose a global challenge, with a significant portion of the population facing barriers to timely and effective treatment. Traditional mental health care systems often struggle to meet the increasing demand for services, leading to prolonged waiting times and inadequate support for individuals in distress. Moreover, stigmas surrounding mental health further hinder many from seeking help. The advent of AI technologies offers a promising solution to these challenges. Through advanced machine learning algorithms, AI can enhance the accuracy and efficiency of mental health assessments, enabling early detection of disorders and personalized treatment plans. [1] AI-driven chatbots and virtual assistants can provide continuous, accessible support, offering a bridge to care for individuals who may be hesitant or unable to access traditional services. Furthermore, these technologies can help destigmatize mental health discussions by providing a private and non-judgmental space for users. The paper reviews the current landscape of AI applications in mental health, highlighting successful case studies and identifying key ethical considerations. It also delves into the potential pitfalls and challenges associated with AI implementation, emphasizing the importance of maintaining a balance between technological advancement and human-centric care. Ethical considerations, privacy concerns, and the need for ongoing human oversight are crucial elements in ensuring the responsible deployment of AI in mental health contexts. As AI continues to evolve, it is imperative to establish robust frameworks for regulation, accountability, and transparency to safeguard against unintended consequences. The paper concludes by proposing a holistic approach that integrates AI seamlessly into existing mental health care infrastructures, fostering collaboration between technology developers, mental health professionals, and policymakers to create a future where AI contributes significantly to bridging gaps in access and elevating the overall quality of mental health care.

show abstract

Comparison of three machine learning models to predict suicidal ideation and depression among Chinese adolescents: A cross-sectional study

Cited by 17 publications

References 45 publications

Predicting individual cases of major adolescent psychiatric conditions with artificial intelligence

Predicting individual cases of major adolescent psychiatric conditions with artificial intelligence

Predicting the onset of internalizing disorders in early adolescence using deep learning optimized with AI

AI-Enabled Mental Health Assessment and Intervention: Bridging Gaps in Access and Quality of Care.

Contact Info

Product

Resources

About