Vulnerability and Protective Factors for PTSD and Depression Symptoms Among Healthcare Workers During COVID-19: A Machine Learning Approach

Background: Healthcare workers are at high risk for developing mental health problems during the COVID-19 pandemic. There is an urgent need to identify vulnerability and protective factors related to the severity of psychiatric symptoms among healthcare workers to implement targeted prevention and intervention programs to reduce the mental health burden worldwide during COVID-19.Objective: The present study aimed to apply a machine learning approach to predict depression and PTSD symptoms based on psychometric questions that assessed: (1) the level of stress due to being isolated from one's family; (2) professional recognition before and during the pandemic; and (3) altruistic acceptance of risk during the COVID-19 pandemic among healthcare workers.Methods: A total of 437 healthcare workers who experienced some level of isolation at the time of the pandemic participated in the study. Data were collected using a web survey conducted between June 12, 2020, and September 19, 2020. We trained two regression models to predict PTSD and depression symptoms. Pattern regression analyses consisted of a linear epsilon-insensitive support vector machine (ε-SVM). Predicted and actual clinical scores were compared using Pearson's correlation coefficient (r), the coefficient of determination (r2), and the normalized mean squared error (NMSE) to evaluate the model performance. A permutation test was applied to estimate significance levels.Results: Results were significant using two different cross-validation strategies to significantly decode both PTSD and depression symptoms. For all of the models, the stress due to social isolation and professional recognition were the variables with the greatest contributions to the predictive function. Interestingly, professional recognition had a negative predictive value, indicating an inverse relationship with PTSD and depression symptoms.Conclusions: Our findings emphasize the protective role of professional recognition and the vulnerability role of the level of stress due to social isolation in the severity of posttraumatic stress and depression symptoms. The insights gleaned from the current study will advance efforts in terms of intervention programs and public health messaging.

Download Full-text

180 The role of social constructs in placental insufficiency outcomes: an unsupervised machine learning approach

American Journal of Obstetrics and Gynecology ◽

10.1016/j.ajog.2020.12.202 ◽

2021 ◽

Vol 224 (2) ◽

pp. S121-S122

Author(s):

Ramamurthy Siripuram ◽

Nathan R. Blue ◽

Robert M. Silver ◽

William A. Grobman ◽

Uma M. Reddy ◽

...

Keyword(s):

Machine Learning ◽

Placental Insufficiency ◽

Learning Approach ◽

Social Constructs ◽

Unsupervised Machine Learning ◽

Machine Learning Approach

Download Full-text

A Study on the Repercussions of the COVID-19 Pandemic in the Mental Health of the Common Public: Machine Learning Approach

Artificial Intelligence and Technologies - Lecture Notes in Electrical Engineering ◽

10.1007/978-981-16-6448-9_23 ◽

2021 ◽

pp. 215-231

Author(s):

Anusha Jayasimhan ◽

Preetiha Jayashanker ◽

S. K. Charanya ◽

K. Krithika

Keyword(s):

Mental Health ◽

Machine Learning ◽

Learning Approach ◽

Machine Learning Approach ◽

The Common

Download Full-text

Using Machine Learning To Understand Suicide: A New Approach To Classifying Australian Coroner’s Court Decisions

10.21203/rs.3.rs-640308/v1 ◽

2021 ◽

Author(s):

Ravi Iyer ◽

Elizabeth Seabrook ◽

Suku Sukunesan ◽

Maja Nedeljkovic ◽

Denny Meyer

Keyword(s):

Mental Health ◽

Machine Learning ◽

Mental Health Disorder ◽

Supervised Machine Learning ◽

Feature Engineering ◽

Court Case ◽

Learning Approach ◽

Machine Learning Classification ◽

Machine Learning Approach ◽

Case Files

Abstract We aimed to demonstrate how a large collection of publicly accessible Australian Coroner’s Court case files (n=4459) (2009-2019) can be automatically classified for determination of death by suicide, presence of mental health disorder and sex of deceased via Natural Language Processing (NLP) methods - supervised machine learning and unsupervised dictionary-based and string search based approaches. We achieved superior levels of accuracy in the machine learning classification (Gradient Boosting vs. Random Forest baseline) of deaths by suicide of 83.3% (sensitivity = 85.1%, Specificity = 79.1%) and an accuracy of 98.3% for the dictionary-based classification of mental health disorder, as defined by the OCD-10 (sensitivity = 99.0%, specificity = 97.9%). Our machine learning approach automatically classified 24.2% (1078/4459) of the case files as referring to deaths by suicide while 63.7% (2940/4459) where classified as exhibiting a mental health disorder1. We employed a two-stage machine learning approach involving feature engineering, followed by predictive modelling in the second. Feature engineering involved several steps including removal of low value text, parts of speech analysis, term document weighting and topic clustering. Predictive classification involved extensive hyperparameter tuning to yield the most accurate model. We validated our models against a manually pre-coded subsample of case files, and also via binary logistic regression to test the contribution of each classified mental health disorder against determinations of deaths by suicide according to extant literature. This validation step confirmed elevated odds of suicide attributed to diagnoses of Depression, Schizophrenia and Obsessive Compulsive Disorder. Finally, we offer a short case study to demonstrate the efficacy of our approach in investigating a subset of case findings referring to suicides resulting from family violence. We offer a proof of concept model that demonstrates an objective and scalable approach to the analysis of legal texts. The use of NLP methods in analysing Coroner's Court case findings has important implications for the ongoing development of a real-time surveillance of suicide system in Australia.

Download Full-text

Addressing Depression Symptoms among University Students under COVID-19 Restrictions—The Mediating Role of Stress and the Moderating Role of Resilience

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph182312752 ◽

2021 ◽

Vol 18 (23) ◽

pp. 12752

Author(s):

Chang Liu ◽

Melinda McCabe ◽

Sebastian Kellett-Renzella ◽

Shruthi Shankar ◽

Nardin Gerges ◽

...

Keyword(s):

Mental Health ◽

University Students ◽

Social Isolation ◽

Perceived Stress ◽

Structural Equation ◽

Depression Symptoms ◽

Targeted Interventions ◽

Mediating Role ◽

Moderating Role

Background: The COVID-19 pandemic has contributed to a decline in mental health globally. Compared to the general population, university students have been identified as a group vulnerable to developing depression symptoms during the pandemic. Social isolation, a signature mental health consequence under physical-distancing regulations, is a known predictor of depression symptoms during the pandemic. Yet, more research is required to understand the mechanism that underpins the isolation–depression association and identify psychological factors that may attenuate the association. The current study aimed to understand the role of stress and resilience in the isolation–depression association among university students. Methods: Data were collected from 1718 university students between 28 and 31 May 2020. Partial least squares structural equation modelling (PLS-SEM) was used to examine the mediating role of perceived stress and the moderating role of resilience in the isolation–depression association. Results: We found that perceived stress partially mediated the association between social isolation and depression symptoms. Both the direct and indirect effects were moderated by participants’ resilience levels. Conclusions: Social isolation during the pandemic may contribute to depression symptoms both directly and through elevated stress levels. As an internal strength, resilience may buffer the adverse effects of isolation and stress on depression symptoms. Targeted interventions including mindfulness and physical exercise training may provide promising results in reducing depression symptoms among university students and should be considered by university administrators particularly during times of imposed physical-distancing measures.

Download Full-text

Effects of the Disastrous Pandemic COVID 19 on Learning Styles, Activities and Mental Health of Young Indian Students - A Machine Learning Approach

2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS) ◽

10.1109/iciccs48265.2020.9120955 ◽

2020 ◽

Cited By ~ 4

Author(s):

Anuradha Khattar ◽

Priti Rai Jain ◽

S. M. K. Quadri

Keyword(s):

Mental Health ◽

Machine Learning ◽

Learning Styles ◽

Learning Approach ◽

Machine Learning Approach ◽

Indian Students

Download Full-text

Machine learning-based analyses support the existence of species complexes for Strongyloides fuelleborni and Strongyloides stercoralis

Parasitology ◽

10.1017/s0031182020000979 ◽

2020 ◽

Vol 147 (11) ◽

pp. 1184-1195 ◽

Cited By ~ 2

Author(s):

Joel L. N. Barratt ◽

Sarah G. H. Sapp

Keyword(s):

Machine Learning ◽

Population Level ◽

Strongyloides Stercoralis ◽

Human Infection ◽

Learning Approach ◽

Multiple Datasets ◽

Species Complexes ◽

Machine Learning Approach ◽

Serious Disease

AbstractHuman strongyloidiasis is a serious disease mostly attributable to Strongyloides stercoralis and to a lesser extent Strongyloides fuelleborni, a parasite mainly of non-human primates. The role of animals as reservoirs of human-infecting Strongyloides is ill-defined, and whether dogs are a source of human infection is debated. Published multi-locus sequence typing (MLST) studies attempt to elucidate relationships between Strongyloides genotypes, hosts, and distributions, but typically examine relatively few worms, making it difficult to identify population-level trends. Combining MLST data from multiple studies is often impractical because they examine different combinations of loci, eliminating phylogeny as a means of examining these data collectively unless hundreds of specimens are excluded. A recently-described machine learning approach that facilitates clustering of MLST data may offer a solution, even for datasets that include specimens sequenced at different combinations of loci. By clustering various MLST datasets as one using this procedure, we sought to uncover associations among genotype, geography, and hosts that remained elusive when examining datasets individually. Multiple datasets comprising hundreds of S. stercoralis and S. fuelleborni individuals were combined and clustered. Our results suggest that the commonly proposed ‘two lineage’ population structure of S. stercoralis (where lineage A infects humans and dogs, lineage B only dogs) is an over-simplification. Instead, S. stercoralis seemingly represents a species complex, including two distinct populations over-represented in dogs, and other populations vastly more common in humans. A distinction between African and Asian S. fuelleborni is also supported here, emphasizing the need for further resolving these taxonomic relationships through modern investigations.

Download Full-text